I I m -N Al B&B INFoRmA6iriom & IMASE MANASEMENT 300 @omcc Mmmuc"n R=uLxvmm Ulmosen P4^"Lnooto, MAL owl* ZD'772 11 Us^ * taol) 24@l 110 w I A STUDY OF THE REGIONAL MEDICAL PROGRAM VOLUME II HISTORICAL BACKGROUND _VGIONALIZATION FACILITATION EVALUATION RELATIONSHIPS Prepared for Regional Medical Programs Service Health Services and Mental Health Administration Department of Health, Education, and Welfare Contract No. PH-43-1014 By ARTHUR D. LITTLE, INCORPORATED and THE ORGANIZATION FOR SOCIAL AND TECHNICAL INNOVATION Cambridge, Massachusetts November, 1970 Arthur D Little, Inc A STUDY OF THE REGIONAL MEDICAL PROGRAM TABLE OF CONTENTS Volume II Introduction ............................. Chapter I Background ............................... Chapter II Regionalization .......................... Chapter III Facilitation ............................. Chapter IV Evaluation ............................... Chapter V Relationships ............................ Chapter VI Program Planning and Budgeting ........... Appendix A Other Regionalization .................... Appendix B Roles and Functions of Relationships ..... Appendix C Arthu r 1) l,i t t le, Inc. I. INTRODUCTION Arthur 1) l.ittle, Inc I. INTRODUCTION This report has been prepared jointly by Arthur D. Little, Inc. (ADL), and the Organization for Social and Technical Innovation (OSTI), both of Cambridge, Massachusetts, from work done under contract PH-43-1014 with the Division of Regional Medical Programs, DRMP, since replaced by the Regional Medical Programs Service (RMPS)@ The study which began in July of 1968, was to be a 2-year comprehensive analysis of three central aspects of the Regional Medical Program: (1) The concept of regionalization as it applies to the@Regional Medical Program and elsewhere in other Government and non- Government programs. (2) The evolving relationship between Regional Medical Programs Service in Washington and the individual Regional Medical Programs. (3) The need to develop a comprehensive framework for evaluating the Regional Medical Program at both the national and regional levels. The contract also,required us to consider the applicability of program planning and budgeting systems (PPBS) and other economic cost/benefits analyses to RIV. Early in the study, responsible officials of RMPS agreed with us that these deserved only secondary attention. The reasons are pointed up in Appendix A to Volume II, which deals with this subject. Finally, we were required to look rather broadly at other re- gionalizing experiences to see whether they might provide clues for RMP development. The limited findings of this investigation are given in Appendix B to Volume II. We have chosen to present our report in three separate volumes. Volume I -- Summary presents a concise overview of the Regional Medical Program. Volume II Background, Regionalization, Facilitation, Evalua- tion, and Relationships -- discusses regionalization processes, strategies of planning and action, a unique approach to evaluation in a scheme of systems transformation, and finally the basic relationships between RMPS and the regions themselves, and the outside community. The volume also contains three appendices. The first concerns program planning and bud- geting for RMP; the second is a brief paper on regionalization efforts outside the Regional Medical Programs; the third presents in table form roles and functions of various relationships among those in the RNP system.** * The change accompanied a major reorganization of HEW health agencies in June, 1968. ** To orient the readerwho may not read all of this report, the Introduction is reproduced in both Volume I and Volume II. Arthur D Little, Inc VolumeIII presents descriptions of five regions operating within the RMP structure; viz., North Carolina, New Jersey, Greater Delaware Valley, Northlands, and Memphis. A. CHANGES IN RNP AND ITS ENVIRONMENT DURING THE STUDY During the two years in which the study was in progress, many developments have occurred both in the Regional Medical Program itself and in the larger societal context. The main areas have concerned: k'l) Shifting societal values toward more concern with the pgor, with the environment in which we live, with the costs of health care, and with the need for local initiative as a way of obtaining genuine commitment and action. (2) There has been a growing sense in our country of the need to rationalize, supplement, or otherwise improve health care delivery in the face of indications (reflected by comparisons such as infant mortality and life expectancy figures) that the delivery of health care in the United States, in spite of renowned medical education and reseatch institutions, is not adequate, particularly for those who cannot afford medical care at prevailing rates. Many people believe that the solution lies in the evolution of a more effective pattern of health care delivery within the present system; others, seeing no Itope for the present system, are pressing hard for more radical solutions. Meanwhile the demand for health care grows at a fast pace. (3) There have been many changes in personnel in the Regional Medical Programs Service (RMPS) and its parent, the Health Services and Mental Health Administration. Dr, Marston, the first Director of the Division Regional Medical Programs, left to head up the Health Services and Mental Health Administration and then al- most immediately thereafter to head NIH. Other people such as Karl Yordy, Deputy Director, DRMP; Steven Ackerman, Chief,- Planning and Evaluatl.on Branch; Daniel Zwick, and Maurice Odoroff, Special Assistant for Data and Analysis, have left @ S. Also departed are Dr. William Mayer, Chief, Continuing Education Branch: his successor, Dr. Alexander Schmidt; Dr. Michael Manegold, Associate Director, Division of Professional and Technical Development; and Mrs. Martha Phillips. Recently, both Dr. Joseph T. English, the second administrator of the Health Services and Mental Health Administration, and Dr. Stanley W. Olson, the second Director of the Regional Medical Programs Service, have left as well. During 1968 and 1969, @S, along with other Federal agencies, experienced a severe personnel freeze, which left the Regional Medical Programs Service unable to add quali- fied staff during a period of rapid program expansion. In i-2 Arthur D Little, Inc part, to compensate for this development. the Chronic Disease Control Program was transferred to the Regional Medical Programs Service in the hope that some of the energies and talents of its people could augment the human resources available to the Regional Medical Programs. Un- fortunately, the process of acquiring and integrating the Chronic Disease Control Program consumed an unexpectedly large amount of the time and energy of top,RMPS people. Thus in terms of personnel, capacity, and program management, the shift was for many months a net drain. (4) Over the past two years, the individual regions have evolved and matured considerably. Many have taken on new forms of organization as the@imensions and needs of the program be- came clear. All but one of the fifty-five have now moved from the planning stage to full operational status. The concepts of the nature of a region, its function, and the functions of the regional core staffs have evolved considerably. Two developments are of special note: A marked shift in emphasis has occurred in some Regional Medical Programs from primary concern with the categorical diseases, continuing education, and technological transfer to the functions of a health system change agent-ultimately affecting (although not delivering) primary care. From this shifting view, projects can be both desirable activities in themselves and vehicles for collaborative efforts leading to desirable systems change. RMP has emerged as the only authentic organization on a national scale for "connecting up" the Federal government with the medical establishment and particularly the practicing physician. (5) Finally, the past two years have been marked by increasing fiscal constraint, manifested-in many ways including the per- sonnel freeze mentioned above. During the early stages of the program, more money was available than could be usefully spent considering the amount of time needed for the regions to get organized and plan before "going operational." But as more regions came on stream and built needs for more funding, the financial situation tightened to the point where there were, as of June 30, 1970, about $30 million in approved but unfunded projects. In other words, a reasonably clear balance between funds available and the need for funds has never really been achieved and maintained. The current deficiency of fundstd support even completely approved (and therefore presumptively worthwhile) projects has added a substantial element of uncertainty to the con- fusion of newness and its accompanying lack of positive program definition. i-3 Arthur 1) 1-ittle, Inc. In the Table on the following page there is a summary of the authorizations, appropriations, and amounts obligated from the beginning of the program through fiscal year 1970. Also shown is the rate at which regions entered the planning stage and became operational. It can be seen from this Table that all but one of the regions has now gone operational. What does not show in the Table is that the amount of funds approved by the National Advisory Committee exceeds the amount of grants because the amount available for obligation would not permit full funding. B. METHODS USED IN THE STI'DY To carry out this study, the ADL/OSTI team interviewed people both within RMP and in the medical field outside RMP. We interviewed staff at all levels of HSMHA and NIH, Congressmen and congressional'staffers, and experts on special aspects of health and health care delivery. Of central importance to the study were the investigations under- taken in the field to give us an understanding of processes and problems in the individual regions. ADL and OSTI staff visited 18 regions in all. Of these, four.were chosen for intensive study of 8 to 12 man-weeks each.- These regions were Greater Delaware Valley, New Jersey, North Carolina, and North- lands. In these regions we sought as much information and as many points of view about RNP as we could find, including the reactions of those engaged in the program, those who know little or nothing about it, and even those known to be outspokenly opposed to it. We engaged in frank and open discussions with: (1) Practicing physicians, nurses and other medical professionals, (2) Representatives of medical societies and nursing associations, (3) Deans, department heads and other professional staff of medical schools and schools of public health , (4) Hospital administrators and directors of medical education, (5) Administrators of extended care facilities, (6) Directors of hospital planning councils, (7) Representatives of voluntary health agencies, (8) Directors of state and areawide comprehensive health planning agencies, (9) Staff of OEO, Model Cities, Neighborhood Health Agencies, and the like. i-4 Arthur D Little, Inc Budget and Grant History (Dollars in thousands) FY 1966 FY 1967 FY 1968 p@' 1969 FY 1970 Authorization -------- $50,0(YO $90,000 $200,000 $65,000 $120,000 Appropriation: grants ----------- 24,000 43,000 53,900 56,200 73,500 Amount available for obligation 24,000 43,934 48,900 72,365 78,500 Amount obligated grants ------------ 2,066 27,052 43,635 72,365 78,500 ------------------------ ---------------------------------------------------- Regions in: Planning @ctatus New ------------------ 7 41 6 1 - Total ---------------- 7 48 54 55 55 Operational Status New ------------------ 0 4 18 19 13 Total ---------------- 0 4 22 41 54 Includes carryover amounts I-5 Arthur 1) Little- Inc Within the RMP offices,, we interviewed: (1) RMP coordinators and their staffs, (2) Members of Boards of Directors and Executive Committees, (3) Members of RAGs and sub-regional advisory groups, (4) Key participants in task forces engaged in solving a wide variety of problems, (5) Project leaders and participants. In connection with RMP-interviews, we reviewed operational plans, reports of activities (including projects), budgets of both core staffs and projects, minutes of policy-making boards, and internal staff memoranda. We also attended meetings of boards of directors and other executive boards, RAGs and sub-regional advisory committees, core staffs, and task forces. In addition, the ADL/OSTI team visited 14 other regions for short periods: Alabama, Arkansas, California, Connecticut, Georgia, Intermountain, Iowa, Maine, Memphis, New Mexico, Northeast Ohio, Northern New England, Tri- state, Western Pennsylvania. Volume III of this report describes five of the RMPs visited: the four selected for detailed study, plus Memphis, which proved to be of special interest. During the course of this study, we met at frequent intervals with the people in RMPS in Washington to appraise them of what we were doing and thinking, and during the latter months of our work we involved them in our field trips. We are grateful to them and to the individual regional co- ordinators, RAG members, core staffs and others for the support, cooperation, and the generous contributions of time they gave us. The membership of the ADL/OSTI team included: from ADL, Phi-lip Donham (project leader), Diana Beatty, John Bruckman, James J. Dunlop, Homer J. Hagedorn, Edward M. Kaitz, Moshe Katz, James Mitchell, Alexandra Walcott, and N. Conant Webb, M.D.; and from OSTI, Ralph Muller, Evelyn Murphy, Gerald Rosenthal, and Donald Schon. C. PERSPECTIVES ON RMP We attempt in this report to bring to the surface the realities that RNP people talk about when they are off the record and not preoccupied with procedures. When the National Advisory Council and the Review Committee put their papers aside, they are concerned with who has captured the program; what is the price of involving some particular community or institution; what are the health politics of an area -- in terms, for example, of such i-6 Arthur D Little, Inc issues as private and public medicine, academic and private practice -- and where the power is. The national staff and the National Advisory Council and Review Committee informally evaluate regional programs in terms of these issues. We have interpreted the work statement to include an invitation to say what we believe RNP should be in light of the activities actually going on in RNP, and in light of e'merging national health issues. For us, these realities organize themselves into a theory asserting RMP's role as an agency assisting "systems transformation" in the delivery of health care. While this assertion is found principally in the chapters on regionalization and facilitation, some perspective concerning it is neces- sary at the outset of this report. We studied RMP at a time of national transition. 1. Three Views of the Program We saw three principal positions taken: (1) With the history of NIH, it was easy and accurate for' a number of the national staff to regard RMP entirely as a grant program in the NIH mode. (2) Others sought "strong central leadership," a view that had consistency with the notion that the headquarters of a Federal program ought to administer the program (and eventually would because all Federally operated programs turn out that way.) (3) The third view, more amorphous, emphasized the notion that RNP was somehow "about change." Many saw RNP as a combination of (1) and (3) -- a program of local initiatives to bring about change, supported by a familiar grant mechanism. Everybody could agree that in some sense PM was a "change agent." Those who took the concept of the grant program or the concept of the administered program as their principal position could still see that RMP was affecting the relationships among components of the health care system. Few in DRNP, however, seemed to regard change-agentry as the essence of the program. 2. RMP as a Change Agent During the two years we have worked with the program, there has been considerable change in the viewpoints of people espousing all three positions. Though the grant program exponents continue to favor a hands- off view with respect to the regions, there is no question that they see many differences between a grant program under NIH and a grant program under HSMHA. Not only does project content have to change, but the criteria used in grant review must also change,- and they change in the direction of many of the criteria one might use if one were trying to shape RMP to be literally and exclusively a change agent in the health care delivery system. I-7 Arthur 1) Little, Inc Similarly, some of those in RMPS who favor central leadership and who want to respond positively to what they perceive as regional requests for direction are now more clearly aware that whatever happens will happen in the regions. They are coming to view central direction as guidance, enabling the regions to produce strategies, to think in programmatic terms rather than project terms, and to deal with the local issues of the health delivery system. These shifts in viewpoint seem to show a convergence toward the feeling that RMP is in fact a change agent, though one constrained by the historical process by which RNP was created, the terms of the legislation authorizing the program, and the beliefs, interests, needs, and capabilities of the constituency available to it. We see RNP as a program about change, whose essence lies in social and institutional change processes, and not one for which these processes are merelv incidental. The central aim of our report is to present this view of RMP, with its implications for the future shaping of the program. The administrative machinery available to the division, however, is that of a grants program or a centralized government program; as a result it is historically easier to view RMP as a program about change within one or both of these structures. We wish this report to suggest that RNP can be explicitly modeled on a third basis that in our view would be more completely consistent with the pattern imposed by its legal con- straints and the emerging health issues of the 1970's. This report is a still picture of what is essentially a moving target. It thereby suffers from at least two limitations. First, it cannot adequately convey the sense of motion and change which characterizes the Regional Medical Program. Second, it cannot really convey the diversity of viewpoints, the drama, and the differences in development among the many regions. The report does, however, detail our findings and conclusions in the three main areas of investigation -- regionalization, relationships, and evaluation -- and further tries to convey a sense of what the Regional Medical Program was, is, and can become. i-8 Arthur D 1-ittle, Inc. II. BACKGROUND TO PL 89-239 Arthur D Little, Inc II. BACKGROUND TO PL 89-239 A. FORCES SHAPING THE REGIONAL MEDICAL PROGRAM The enactment of PL 89-239 reflected some trends that had been developing for a long time. These trends had to do with the concept and practice of regionalization, the role of medical research and the research establishment in the United States, changing public attitudes and values with respect to health and health care, the general nature of the medical care system in the United States, and the national political situation in the United States in the 1960's. 1. Regionalization . For decades, a succession of American public health leaders has been urging regionalization of health services. These leaders have deplored weaknesses stemming from what they call the fragmented nature of our health care system, the lack of connection among community hospitals and between them and the major teaching centers, and the independent and entrepreneurial nature of practicing physicians. In most discussions and reports of commissions dealing with regionalization, the concept of regionalization under discussion has been a center-periphery system built around major medical school-teaching hospital complexes, with links between these and the community hospitals for teaching purposes and for referral to the teaching center of the more difficult cases that community hospitals could not handle. Over the years, a variety of reports have come out urging regionalization and continuing education for physicians built on this model. Among these reports, some are cited with particular frequency: the Lord Dawson report of 1920 in England; the findings of the Commission on the Costs of Medical Care of the early 19@O's; the 'Dryer Report of 1962, Lifetime Learning for Physicians'; and the Coggeshall Report, Planning for Medical Progress Through Education in 1965. There have been several reasonably successful and highly publicized examples of such regionalization, which have served to provide empirical support for the theoretical regionalization model. The examples used in the books about "regionalization" are repeated over and over: the Bingham Associates plan linking the Tufts New England Medical Center with community hospitals in Maine; the Albany Regional Hospital Program linking the Albany Medical College with five hospitals in New York State and Massachusetts for post-graduate medical education and consultation; the Rochester Regional Hospital Council linking the School of Medicine and Dentistry of the University of Rochester to 18 hospitals in the Rochester area for joint planning, joint operation, and teaching; and the Hunterton Medical Center in New Jersey, which joined with the NYU School of Medicine for teaching and referrals. Two major examples, often proposed as models for regionalization, come from the Armed Forces. These are the military hospital system and the Veterans Administration hospital system. The military rationale calls for ii-i Arthur 1) 1-ittle, Inc- battle casualties to be treated at forward stations, with the severe cases being sent to intermediary or base hospitals. The VA hospitals some years ago linked with medical schools in order to serve systematically as teaching hospitals for those institutions. Also, within the Veteran's Administration hospital system, there are selective referral patterns; for example, in Boston the Roxbury VA hospital takes care of all the spinal cord injuries in the New England VA region by referral. These examples are not intended to constitute a prehistory of regionalization in the medical care system. Once regionalization has happened it will be possible to identify the significant precedents and contributors. But these examples do serve to point up two widely held opinions with regard to the practice of medicine and delivery of health care: (1) that academic medicine (medical faculties and large teaching hospitals) must be directly involved in important changes in medical practice and organization, and (2) that the system should relate specialized resources at medical centers to less specialized peripheral institutions. The examples listed all involve the actual or proposed deliberate creation of institutions. Such institution-building attempts are by no means new; they take place all the time in every branch of American life -- religion, business, municipal organization, education, and family or communal life. However, the great majority of such experiments die with the initial enthusiasm of their advocates or are exhausted with the resources that first support them. These examples share still another characteristic. They are attempts to create a "public" medicine. They by no means all depend upon any sort of governmental control or sanction beyond those already imposed on private medicine. But they are all attempts to create connections between patients and physicians that take account of broader relationships between people and the institutions that care for them. Proponents of regionalization in the health care field have long been convinced that federal legislation is needed to bring it about. This conviction was'frustrated in the actual form taken by the original Hill- Burton Act, which provided matching funds for hospital construction, based, presumably, on a regionalization plan. The Hill-Burton program was to be the first in a series of legislative acts recommended initially by Surgeon- General Thomas Paran in 1944 and intended to rationalize the health care system along lines based on his understanding of the Bingham Associates program. However, in the view of regional health planning proponents, the Hill-Burton legislation (or its eventual administrative interpretation) accomplished little if anything in the way of true regionalization, succeeding only in dotting the landscape with small community hospitals. The more recent trial and failure of many of the voluntary local hospital planning councils (most of which have as their mission'to coordinate planning among hospitals in a particular area so as to avoid costly duplication of services) leaves the situation unchanged. Nevertheless, the idea of regionalization persisted. Coordination among hospitals, linkages with university medical centers, and graded levels II-2 Arthur1) l,ittle, Inc of care appeared to make sense in economic terms and in view of increasing specialization. To deliver relatively simple primary care in offices and clinics, to augment these services at small community hospitals, and to concentrate highly sophisticated care at the university medical centers seemed a credible way to organize for meeting needs in terms of their frequency and in terms of using scarce resources efficiently. Corresponding referral patterns would provide a way to get people to the care they needed. Interaction and communication between community-based and university-based physicians would be strengthened. The whole pic,ture was simple and rational in the terms stated, and easily visualized. And this is the model of regionalization which was incorporated into the report of the President's Commission on Heart Disease, Cancer, and Stroke and partl7 embodied in the first draft of the legislation which was to become the Regional Medical Program. In the thinking-that led to this legislation, the limited American precedents, the reorganization of the Puerto Rican health care system, the British National Health Service, and other European systems all were interpreted as center-periphery regionalization. This was what one meant if one talked about regionalization at all. Advocates and critics alike could agree that a system consistent with these examples would give more power', prestige, and eventually relatively more income to professors in medical schools, and that it would be regulated by the government. 2. Medical Research A second trend influencing the development of the RMP legislation was the phenomenal build--up in government-supported biomedical research. The genesis and swift growth of the National Institutes of Health represents the institutionalization par excellence of this trend. Of particular influence on the formation of RMP were the following developments which were evident by the early 1960's: o As a result of general public acclaim for research, the apparent success of medical research, and the natural concentration of research grants and contracts in medical schools and their teaching hospitals, most medical schools became substantially or partially dependent on research money from NIH to augment their programs, construct their buildings, and train additional researchers and potential teachers. Organized medicine tacitly or openly consented to this avenue for providing government funds to medical schools. o The sheer size of expenditures made the medical research budget a vulnerable political target, particularly once NIH reached and spectacularly surpassed annual expenditures of a billion dollars. Like the Department of Defense, whose research expenditures had come under criticism when they became large by established and popular standards, NIH encountered growing criticism of its own research budget. There was mounting and continuing pressure to translate the results of research into clinical practice, or, to put it another way, to demonstrate the applicability of the research and thus justify the billions of dollars spent on it. II-3 Arthur D Little, Inc oThe effects of the research boom on medical manpower reinforced other national trends that were stripping the countryside of its supply of physicians, and probably also intensifying the shortage of physicians in urban ghetto areas. The research boom tended to siphon off medical manpower by encouraging medical students to seek research careers, by encouraging further sub-specialization in clinical practice that could be carried out only in major medical centers, and by impressing young physicians with the idea that proper medicine could be delivered only in very highly developed hospital settings or in-conjunction with such hospitals. The combination of these tendencies made the proposal to regionalize resources for the treatment of heart, cancer, and stroke subject to some rather extreme interpretations. What was ultimately to become RMP looked to some like a defensive bid on.the part of the Lasker group to shore up the edges of the NIH effort by demonstrating that real efforts were being made to apply the results of clinical research. Another interpretation held that RMP must be a means of organizing academic medicine in order to sharply increase its power to encroach on direct patient care. At best, these interpretations were partial, and at worst they were exaggerated, but they were considerations that proposals to create a regional medical program had to deal with; they represented part of the emotional and political atmosphere into which RMP had to emerge. 3. Changing Values and Expectations A third trend has been increasing public awareness of the benefits and need for health care. This attitude has been stimulated by the medical profession, the voluntary agencies, the media, and the experiences of millions of Americans in the Armed Forces. Health care, as the saying goes, has become a right rather than a privilege reserved for those who can pay for the service. The implication that medical care could really accomplish almost any miracle was a part of this belief. The belief in miracles upheld the popular support over the past couple of decades for providing lithe best of health care to every American." The statistical formulation that heart disease, cancer, and stroke were responsible for 70% of the deaths in the United States, and the presumption that some of these deaths could be prevented by getting health care to people who were not receiving it, provided a graphic justification for RMP. But these attitudes also set RMP up for trying to meet some impossible expectations. 4. Politics A fourth trend, located in time more closely to 1965 when the RMP legislation was developed, had to do with the,politics of the Johnson administration. President Johnson hoped for a major legislative program connected with health care. In his message to Congress in January 1965, he presented a monumental legis'lative package dealing with health that included, in addition to what ultimately became RMP, Medicare, increasing appropriations for maternal and child health and crippled children's II-4 Arthur D Little, Inc. services, medical assistance to the poor, improved community mental health services, rehabilitation centers,- an extended program for the mentally retarded, increased Hill-Burton expenditures, support for group practice arrangements, increase in support for the health profession's Education Assistance Act, grants to medical schools, scholarships for medical and dental students, increased spending for health research and research facilities, and consumer protection in the field of health. Most of these bills were redrafts or resubmissions, or otherwise represented a long process of development and siow public education. Not so with RMP. It was a Johnson bill, and was the piece of major legislation in the 1965 legislative package on health which was developed and drafted entirely after the Johnson administration began. B. THE PRESIDENT"S COMMISSION ON HEART DISEASE, CANCER, AND STROKE: 1964 The President's Commission on Heart Disease, Cancer, and Stroke, appointed in 1964, took nine months to do its work. The Commission was headed by Dr. Michael DeBakey and included sub-committees dealing with heart disease, cancer, stroke, rehabilitation, manpower, communications, facilities, and research. In summary, the Commission recommended that the Federal Government give financial or administrative support, or both, to the following: Regional centers for heart disease, cancer, and stroke Diagnostic and treatment stations The development of medical complexes The development of additional centers of excellence A national stroke program unit Community health planning (grants) Community health research and demonstrations Community-based medical programs Statewide programs for heart disease control A national cervical cancer detection program Continuing education of the health profession Public information on heart disease, cancer, and stroke Establishment of biomedical research institutes Specialized research centers Research projects (grants) Contracting authority for research and development General (not earmarked) research funds A standard goverrment-wide policy for payment of full costs of research Expansion of resources for preparation of health manpower Increased recruitment for the health professions Undergraduate training in medical and dental schools (grants) Training for research Clinical training Stabilization of academic physician supply and support Training of health technicians Training of specialists in health communications II-5 Arthur D Little, Inc Continuous assessment of health manpower needs Expansion of patient care facilities Strengthening of the federal hospital program Medical libraries National medical audio-visual center Statistical programs Increased animal resources for biomedical research A clearing house for drug information International research and training programs. The first three of the Commission's recommendations formed the basis for the original bill, S-596, considered by the Sub-Committee on Health of the Senate Committee on Labor and Public Welfare. (In later hearings before the House Appropriations Committee, Dr. Shannon, Director of NIH, said that the other recommendations of the DeBakey Commission -- directed toward improving co=unity-based programs for the application of medical knowledge, the expansion of facilities and-support for development of new knowledge through research, the expansion of resources to train new manpower, and the enlarging of facilities and resources available for teaching, research, and community service -- could largely be accomplished through existing NIH programs in the national Institute of General Medical Sciences, the National Cancer Institute, the,National Heart Institute, and the National Institute of Neurological Disease and Blindness.) C. SENATE HEARINGS ON S-596 S-596 was drafted by Dr. Edward Dempsey, then Special Assistant to the Secretary of Health, Education, and Welfare (also a member of the President's Commission and Chairman of the Manpower Task Force) and his assistant, Dr. William Stewart, who was shortly to become the Surgeon-General. The bill as drafted had the intention of establishing (over a period of five years) about thirty regional complexes, each built around a university medical center (a medical school-teaching hospital-research institute combination) and serving a given geographical area having a radius of about 100 miles and encompassing on the average about two million people. It was planned that in these complexes there would be about 450 diagnostic treatment stations in total for heart disease, cancer, and stroke. The medical centers would assume the initiative for planning and developing each complex. On February 9 and 10, 1965, two months after the publication of the President's Commission Report, hearings on S-596 were held before the Senate Sub-Conmittee on Health of the Committee on Labor and Public Health, which was chaired by Senator Lister Hill. The bill had a fairly easy time in the Senate Sub-Committee hearings. (Congressman Fogarty, in the House Appropriations Sub-Committee held later, said that he was "told that the Senate hearings weren't the best ever held before a legislative committee.") Senator Hill called the proposed S-596 a logical outgrowth of the clinical research center program of NIH begun in 1959. He was supported in this statement by Dr. DeBakey, who stressed the research nature of the proposed centers. HEW Secretary II-6 Arthur D Little Inc Anthony J. Celebrezze stressed in the hearings that the complexes would pull together existing components as much as possible, thus reducing the need for new construction. Dr. Dempsey, his Special Assistant, suggested that perhaps $10-15 million might be enough to establish a fully developed complex. Support for the bill during the Senate hearings came from a variety of sources. Three important supporters, however, wanted some modifications. The Association of American Medical Colleges (AAMC) recommended that the National Advisory Council for regional medical complexes be given more power relative to the Surgeon-General. It also suggested locating the Regional Medical Complex Program (RMC) within NIH. The American Heart Association also came out for a stronger National Advisory Council for regional complexes and for administration by NIH. It urged that the government increase its support of medical schools in the production of more doctors. The Americiifi Cancer Society recommended that Regional Medical Complexes avoid dilution by concentrating only on the categorical diseases mentioned and that "other major diseases" be stricken from the bill. It also urged inclusion of major cancer research centers such as the Sloan- Kettering Institute and the M. D. Anderson Hospital as potential candidates for "centers." Other supporters, such as the APHA and the American Physical Therapy Association, confined their remarks primarily to statements of approval plus an urging of the inclusion of their interests on the National Advisory Council. About the only discordant notes were supplied by Mr. Marion Folsom, former Secretary of HEW, and by the -American Hospital Association. Mr. Folsom talked about the need for community planning of hospitals, the problem of rapidly rising hospital costs, the need for ambulatory services and organized home care; he proposed that local advisory councils supervise the Regional Medical Complexes, that state health departments participate in the program, that expenditures be coordinated with the Hill-Burton and other state or federal-state plans, and that the Regional Medical Complexes serve as demonstration projects and try to get more than the 10% participation from local sources on construction projects. His testimony was different from and d'issonant with that of other people; few of his remarks seemed to be picked up. The American Hospital Association urged that medical schools not be allowed to dominate the program and that small hospitals be used as diagnostic and treatment stations within the context of the proposed plan. The final version of S-596, as reported out by the Senate Sub-Committee, incorporated: (1) A more powerful National Advisory Council, with the Surgeon-General authorized to make grants only upon recommendation by the NAC. (2) More power at the local levels through a requirement for a local advisory group. (3) Most importantly, no funds for new construction, although alterations, remodeling, and renovation were allowed. II-7 Arthur D Little, Inc The version reported out by the Senate represented the first step back from what could be viewed as federally financed (and certainly medical school-controlled) centers. By giving more power to the National Advisory Council and the local advisory groups, this version weakened federal bureaucratic control. By excluding new construction funds, it reduced the possibility of setting up federal centers or stations. MODIFICATIONS IN THE HOUSE OF REPRESENTATIVES The House Committee on Interstate and Foreign Commerce, chaired by Representative Oren Harris, held extensive hearings on the proposed RMC legislation, HR-3140. Between the Senate hearings in February Iand the hearings in the House in July, there had been much speculation about the proposed program and quite a bit of activity by organized groups, particularly the American Medical Associatioh, which had then just lost the Medicare battle. The House hearings started with support for the bill coming from the President, Secretary Celebrezze, members of the President's Commission, and the AAMC. A lot of criticism was expected. Chairman Harris stated that great concern over the proposed legislation was being expressed by a number of people in the health professions. Secretary Celebrezze tried to provide reassurances that no one was trying to put the government into the medical business, that the complexes would operate under local control, that there would be local coordination of available manpower, and that the Regional Medical Complexes would not attempt to duplicate existing resources. Other proponents of the bill also stressed its provisions for local control. In support of the complexes they argued the need for closer communication between researchers and practitioners. Serious challenges were directed to the witnesses by some of the Committee members, notably Representatives Carter of Kentucky, Springer 'of Illinois, Nelson of Minnesota, and Rogers of Florida. The challenges were aimed at some of the premises on which the DeBakey Report and the RMC bill were based. These included challenges to the implication that technology was not available in the smaller hospitals and that there was need to overcome a technology gap or lag. Worries were also expressed about the effect that the proposed program would have on the already short supply of physician manpower. The Minnesota Heart Association recommended a delay of one year in passage to allow sufficient time for study. The University of Minnesota recommended a demonstration or pilot program rather than the elaborate program proposed in the bill. Spokesmen for the private practitioners questioned the degree to which MDs (general practitioners) had had a voice in the preparation of the legislation and the amount of protection offered to the general practitioner from the "monstercenters" which were proposed to be established. They were also worried about pulling manpower out of the rural areas. ii-S Arthur D Little, Inc County medical society representatives asserted that the bill, if enacted, would have an adverse effect on nonparticipating hospitals, would discourage physicians from locating in rural areas, would not effectively combat heart disease, cancer, and stroke, would not improve communications, would be detrimental to those medical schools which were not leaders in the complexes, and would heighten the physician shortage. And besides, they said, there is no serious lag in the dissemination of those new discoveries that are really valid. They quoted the report of the Sub-Committee on Research of the President's Commission on Heart Disease., Cancer, and Stroke, which stated that there wa s no major research breakthrough related to these diseases still awaiting clinical application -- and, in fact, that it knew of no significant body of fundamental medical information that was not being applied. The American Academy of General Practice came out against large specialized complex regio'nal centers. The AAGP objected to the stress on remedial as opposed to preventive care. In addition, it felt that the DeBakey Commission had been excessively dominated by academic medical men. But the big thrust for changing or amending the proposed bill came from the American Medical Association, which deferred testimony before the Sena'te but did testify before the House. The main thrust of its testimony was as follows: o The legislation as proposed was vague. o In fact, there was no serious time lag between discovery and application of research results. o There was ample continuing medical education being conducted by the profession, in particular by the AMA. o There was a well-operated referral system which included hospitals; therefore, the need for linkages in new referral patterns was nonexistent. o Coordinated arrangements among hospitals already existed. 0 The manpower was not available to meet the purposes of the Regional Medical Complex. 9 Regional Medical Complexes would discourage doctors from locating in rural areas. o The Regional Medical Complexes would not improve the communication of new ideas. o RMC would have an adverse effect on nonparticipating hospitals and medical schools. 9 RMC would overburden present facilities. o Patterns of care in the United States would be changed negatively. ii-9 Arthur D L-ittle, Inc o The AMA wanted to write a review and approval of all RMCS. Various other suggestions were made in modification of the bill. At times during the hearings, Chairman Harris injected the notion of collaborative arrangements among equals as opposed to coordinated arrangements with some organizing body in charge. The American Heart Association suggested the inclusion of training among the objectives of the program and suggested that a minimum of two years would be needed for planning the complexes. The Heart Association also suggested that local advisory groups should have broad representation and should be charged with local planning and operational responsibilities. The bill, then, was subjected to many powerful push6s from a variety of interests but few pulls from champions of its objectives. It did have the strong support of President Johnson, but only in general terms. Observers said he had no investment in the detailed contents of the bill, although he did want legislation passed on heart, cancer, and stroke. The enactment of this legislation in the 89th Congress was actually quite remarkable in view of the criticism directed toward it by various components of organized medicine. Also, and even more remarkable, it was the first'piece of major health legislation which had not languished through several congressional sessions before being passed. Medicare, for example, was subject to intense negotiation and major battles before it was passed. By comparison, the legislation for RMP sailed through Congress quite easily. The form in which the bill was finally enacted into law combined the results of pressures from organized medicine, academic medicine, the voluntary agencies, and the professional regionalizers all acting to modify the original concept. It came out looking somewhat the same as the original bill, but in reality it was substantially different. The report of the Committee on Interstate and Foreign Commerce on HR-3140 of September 8, 1965 (House Report No. 963) lays out the changes and the support for them: "Testimony favorable to the legislation was submitted on .behalf of the American Heart Association, the American Cancer Society, the American Hospital Association, the American Public Health Association, the Association of American Medical Colleges, several deans and officers of medical schools and others. "Testimony in opposition to the legislation was submitted by the American Academy of General Practice, the American Medical Association, several state medical societies, and others .... "The AMA President said he was gratified that as a result of these meetings, some 20 amendments to the bill recommended by the AMA Committee were accepted by the administration. filpresident Johnson told us [the AMA] he could not support deferment of the bill, that he favored it and wanted it passed in this session of Congress,' Dr. Appel said. 'President Johnson did, however, direct Secretary Gardner to work with the AMA committee to make the bill II-lo Arthur 1) Little, Inc less objectionable.,,, ... "Dr. Appel said he told administration officials that passage of the original bill would have been followed by a severe adverse reaction from the medical profession.... "The committee has therefore substituted for the phrase regional medical complexes" the phrase "regional medical programs," so as to emphasize the local nature of this.program, its limited scope, and the fact that the primary thrust of the program will be to facilitate arrangements among existing institutions .... the only construction which will be permitted under the reported bill will be alteration, major repair, remodeling, and renovation *of existing buildings, and replacement of obsolete built-in equipment of existing buildings. No new construction will be permitted under this definition.... "...The committee has deleted the phrase 'other major diseases' and substituted 'related diseases.' If at some time in the future it is in the public interest to establish a program for major diseases not related to heart disease, cancer, or stroke, the Congress will give consideration to the establishment of such a program at that time; however, @t present the committee feels that this program should be limited to the three named diseases and other diseases which are related to them. For example, it is known that there is an apparent relationship between diabetes and heart disease .... The committee feels that research should be conducted into diabetes under the program dealing with heart disease insofar as diabetes is related to heart disease. Similarly, certain kidney diseases are associated with high blood pressure which, in turn, is associated with stroke and heart disease. The committee feels that insofar as they relate to stroke or heart disease, these kidney diseases would be appropriate for coverage under the programs established under the bill. ..In several places, the introduced bill provided for coordination' of programs, arrangements, or activities. Fears were expressed to the committee that these words implied the possibility of Federal control of medical practice. The committee feels there is no basis for these fears; however, in those places where 'coordination' is referred to, the committee has substituted 'cooperation' instead .... "...The committee has adopted a further amendment ... which provides that no patient shall be furnished care incident to research, training, or demonstration at any facility unless he has been referred to that facility by a practicing physician.... ... The Committee has been very careful to establish machinery in the bill which will insure local control of the programs conducted under the bill .... Before an application m@y be received and acted on under the bill, the applicant must have designated an advisory group which will include practicing physicians, medical center officials, hospital administrators, representatives from appropriate medical societies, voluntary health agencies, representatives of other organizations Arthur 1) l,ittle, Inc concerned with the program, such as public health officials, and members of the public .... "...At least 2 of the members fof the National Advisory Council], in addition to the 3 previously referred to, shall be practicing physicians. In addition the Surgeon General may not make a grant for any program under the bill, except upon recommendation of this Council .... "...The introduced bill ... provided that one of the components of local programs was to be one or more "diagnostic and treatment stations," defined as a 'unit of a hospital or other health facility providing specialized, high-quality diagnostic and treatment services., The committee has deleted this concept from the bill and har, provided that as a substitute for the diagnostic and treatment station, the local program must include participation by hospitals .... "The Committee notes the agreement among all concerned that full participation of practicing physicians is required for the successful operation of this program.... "One of the objections to the legislation expressed to the committee was that it would have an adverse affect upon the supply of scarce medical manpower, and would discourage physicians from locating in suburban or rural areas. These objections appear to have been based in part upon the theory that the programs established by the bill would involve massive construction of new facilities which would be required to be staffed with doctors and other medical personnel admittedly in scarce supply. Since, as has been pointed out, the bill does not provide for such a program, it will not have the effect feared in this area.... "Fears were expressed during the hearings that the enactment of this legislation would adversely affect medical schools and hospitals which do not participate in the programs set forth in the legislation. ...The fact that one medical school may benefit from a program whereas another school which does not participate is not benefited is not, in the committee's opinion, a valid reason for saying that neither institution should be permitted to participate.... "It would be desirable as an ultimate goal for all medical schools to be involved in programs of the sort contemplated by the reported bill, but some may choose not to participate, and others may become involved in the program at a later stage. "With respect to the effect of the program on hospitals, the committee points out that the intent of this program is not to centralize medical capabilities in a single, or a few, medical centers within a region, but rather is to extend the capabilities now present in the medical centers more widely throughout the region .... The bill is not intended to support programs in competition with existing activities and one of the strengths of the bill is that it provides the flexibility necessary to accommodate the many different patterns of medical II-12 ArthLir 1) l.ittle, Inc institutions, population characteristics, and organizations of medical services found in this Nation." E. COMMENTARY What started out as a series of care-providing complexes mostly based in academic medical centers with a strong, continuing medical education thrust became a program emphasizing continuing medical education and relying on locally controlled regional cooperative arrangements. In order to get off, the ground, the program had to have the cooperation of the practicing physician, and to Congress and the President that meant that the objections of the American Medical Association had to be taken into account. The program probably also had to start categorically in the NIH tradition. Placement of RMP within NIH seemed a foregone conclusion (most witnesses testified in support oj' :Ci). The apparent alternative was the Bureau of State Services, although that organization was having difficulty because its traditional approach was not well accepted or supported. The whole development of the Regional Medical Program legislation and its subsequent history as an operational program can be viewed as a series of steps back from the original concept of "categorical" regionalization built around the center-periphery model. These steps gradually pushed the task of regionalization to lower and lower levels as a price to pay for getting anything done. Dr. Marston, a former medical school dean, was named the first head of the RMP. Under his initial leadership, a philosophy was established in the RMP, which permitted the regions to develop pretty much on their own their regional boundaries and their regional organization. Much of the experience of RMP to date probably results from the fact that the idea was still new when it was enacted into legislation. Since the bill was passed the first time it was submitted, RMP became a reality before many people had a chance to think about it. Small wonder that it was subject to wide variation in interpretation. RMP could be viewed as a kind of political accident, in that no very permanent coalition had been formed to lobby for it. Who would stand up to support it in the long run? To whom would it really belong? Not the Public Health Service, presumably, out of which emerged PL 89-749 (Comprehensive Health Planning) only after RMP passage became a certainty. Apparently not organized medicine; the AMA never endorsed it in the course of its passage. Not the President's Commission; this law was not what they had asked for. Possibly the medical schools, though not quite all rushed to join. Certainly not the Hospital Association. The RMP, in its formative stages, thus became in a sense a projective vehicle for what people wished to see in it. The program was never sharply defined, and therefore people who were interested in research could project research into it. Those interested in continuing medical education could view it as a vehicle to that end; people interested in regionalization could view it potentially as a regionalizing vehicle; those II-13 Arthur D Little, Inc interested in supporting medical schools could potentially view RMP as a source of some support for that effort; people who were interested in not changing the health care system could view RMP as a vehicle for no change (because of local control). Those who were for change in the health care system could look on the RMP as a program to facilitate system transformation. RMP, then, was something to everyone, but not the same thing. In each of its guises it had a few strong supporters, but it lacked unified backing. It had no sanctions nor coercive power to enforce its will. However, it had been passed; it would be funded beyond its early power to spend the money. RMP from its birth was authorized to work with all the major forces in health and medical care service delivery -- but it was also constrained by all the realities, both political and economic. II-14 Arthur D Little, Inc t III. REGIONALIZATION Arthur D Littk Inc III. REGIONALIZATION The purposes of the Regional Medical Program-(RMP) as stated in Public Law 89-239* are: "(a) Through grants, to encourage and assist in the establishment of regional cooperative arrangements among medical schools, research institutions,, and hospitals for research and training (including continuing education), and for related demonstration of patient care in the fields of heart disease, cancer, stroke, and related diseases; "(b) To afford to the medical profession and the medical insti- tutions of the Nation, through such cooperative arrangements, the opportunity of making available to their patients the latest advances in the diagnosis and treatment of these diseases; and "(c) By these means, to improve generally the health manpower and facilities available to the Nation, and to accomplish these ends without interfering with the patterns, or the methods of financing, of patient care or professional practice, or with the administration of hospitals, and in cooperation with practicing physicians, medical center officials, hospital administrators, and representatives from appropriate voluntary health agencies." By its proscription against interfering with the patterns or methods of financing of patient care or professional practice, or with the administration of hospitals, as well as by stipulating a process for creating regional cooperative agreements, the Law has located effective power with- in the regions. The May 1968 Guidelines** underscored the reliance on regional autonomy; the formulation continues to be central to RMP practice. Regionalization was the only thread running through all three purposes as stated in the Law, which called repeatedly for regional co- operative arrangements. But cooperation cannot be viewed as an end in itself, so regionalization from the national point of view must neces- sarily be viewed as a strategy leading to something else. That "something else" is now understood rather broadly as the use of cooperative arrangements to bring about improvement in health care, with emphasis on the categorical diseases. This formulation represented a compromise during the legislative process that fell short of a sharply defined system of regional centers and affiliated stations for the dis- covery and treatment of heart disease, cancer, and stroke. To show how the actual experience of the 55 regions has further defined the meaning of "regional cooperative arrangements" is the objective Title IX, Section 900 Guidelines, p.2 last para. iii-i Arthur D Little, Inc of this chapter. Given the constraints of law, conflicting interests, human capability, time, differing interpretations and emphasis, andfinally the money allotted to RMP, the program could evolve only within the limits these constraints permitted. The result has been something approaching a reversal of ends and means. Where earliest proponents wanted to use re- gionalization to fight heart disease, cancer, and stroke, RNP has come closer to using the categorical disease focus as a vehicle for "re- gionalization," meaning "regional cooperative arrangements." Once this perception is accepted -- or even tolerated -- in a region, the way is open to encompassing the original, prelegislative pur- pose of regionalization (war on heart, cancer, and stroke) with a,new one. Regionalization is what the RMP does in a specific region to help effect systems transformation: to create linkages and patterns that deal with undesirable conditions resulting from the fragmentation of the health care system. (This is set forth in more detail later in this chapter and in Chapter V.) This chapter categorizes the styles of regionalization we have observed, and then presents more fully the concept of regionalization as a voluntary (and, therefore, legal) systems transformation in RMP. RMP regionalization should be looked at on two levels -- the national scene and the individual regions. Examination of the regionali- zation processes occurring at the regional level reveals currently ob- servable alternative forms, strategies, and processes, and offers some options for proceeding in the task of building a region. But to provide context, let us first consider how regionalization has developed on a national scale, and some possible alternative ways it could have developed. A. THE DEVELOPMENT OF REGIONALIZATION FROM A NATIONAL PERSPECTIVE AND POSSIBLE ALTERNATIVES In recommending the establishment of regional centers of excel- lence in heart disease, cancer, and stroke, the Presidentis Commission apparently intended these centers to be located in the major teaching and research complexes typically associated with our medical schools. The DeBakey concept, as it has become known, envisioned a kind of "solar system" approach with medical schools at the center, auxiliary treatment centers in major community hospitals, and less sophisticated diagnostic and treatment stations at the periphery. As such, it represented an ap- plication of the basic "center-periphery" model of regionalization for health resources developed earlier by various health planners, as noted in the preceding chapter. Public Law 89-239, which authorized RMP, did not, in fact, legis- late a center-periphery system, partly because the American Medical Asso- ciation and others resisted the anticipated effects of such a system in 'increasing the power of both academic medical centers and the Federal Government over the patterns of medical practice and patient referral. III-2 Arthur D Little, Inc Also, the concept of upgrading the skills of the private practitioners by exposing them to the techniques employed in the academic medical centers was perceived as imputing lower-quality medical skills to them. This insinuation aroused serious resentment among practitioners everywhere, many of whom felt that the kind of medicine practiced in the academic medical centers does not recognize some of the realities they encounter in private practice and so falls short of true excellence. A system that looked down on the private practitioner was unacceptable. Instead of a national system of centers of excellence surrounded by a diagnostic and treatment station, the Law set up a flexible regional program with considerable local autonomy. The possibilities for diversity were myriad. The diversity began with the definition of regional boundaries. As it turned out, regional populations vary from 20 million to 500,000, and in area from Washington-Alaska to the Metropolitan Washington, D.C. region. Some regions overlap others, such as New Jersey and the Greater Delaware Valley RMPS, which share South Jersey; the Bi-State and Illinois, which share Southern Illinois; and the Tri-State and Albany RMPS, which share Western Massachusetts. Thirty-one regions confirm to pre-existing state boundaries, (somewhat to the surprise of RNPS), 24 encompass parts of states or are multi-state regions. Some regions have one medical school, others a number of medical schools, and a few others (such as Maine) no medical school at ali. Some contain large cities; others do not. The result is a pattern of RMP regions which does not consistently conform to any other existing regionalization pattern. This has produced some problems. For example, where RMP regions have cut across political lines, there have been questions of how to relate to state-based Public Health departments and CHP agencies. On the other hand,, the new pattern does seem to have taken advantage of, or created, entities potentially capable of dealing with emergent health care issues in ways significantly different from pre- existing state-based agencies. In many regions, the RNP has been organized in a way that builds new links among the health care professions. The RNP has secured at least nominal commitment and involvement of thousands of physicians, members of voluntary associations, nurses, allied health personnel, hospital admini- strators, government health officials, and lay people; and it has created regional structures relatively accessible to the influence of, and com- munications among, all these professions and some of the institutions in which they work. In some regions, RNP has permitted (and enabled) a strong alliance to emerge among different categories of health care providers, leavened by the presence of lay people. (In Northlands, for example, doctors and nurses are building closer working relationships than ever be- fore, using coronary care units as a means to that end. Maine, New Jersey, and North Carolina have also moved to bring the several health professions into closer relationships.) In other regions, a pre-existing providers' alliance has been broadened and its social utility potentially increased. (The Mid-South Medical Center Council of Memphis expanded its activities significantly as the Memphis RNP came into being.) Pressure to change III-3 Arthur D Little, Inc or adjust the regional boundaries has for the most part been met con- structively. In those regions where territory overlaps, competition appears to be manageable, and the local pattern of playing off one re- gion against another is not viewed as a significant problem. RNP's regionalization pattern has recognized several medical catchment areas that do not conform to political boundaries. For example, South Jersey relates for medical purposes strongly to Philadelphia. Yet politically and in some ways socially it is tied to New Jersey. Both re- lationsl,.ips were recognized inthe regionalization pattern, which includes South Jersey in two regions. Similarly, Southern Illinois relates strongly to St. Louis (Bi-State RMP), Northern Mississippi relates to Memphis (Memphis RNP), and the Pittsfield area in Massachusetts relates to Albany, New York. In the instances of markedly slow regional development of which we have been appraised, several factors are alleged to have been critical: 9 Program perceived to be "dominated" or "captured" by one of the parties of interest (medical school(s), medical society, core staff); o Program in the hands of an inappropriate program coordinator (unenergetic, unable to cope with a social process in a highly political mileu, unable to communicate with a broad enough spectrum of people, passive); o Program unable to deal quickly with the range of complexities facing it (big city with several medical schools, region with a raging, locked-in conclift that is built into basic RNP structure). Since we have visited no regions that could be judged total failures, we can only acknowledge that these factors would be strong nega- tive forces wherever they are found. They are conditions that are risked in any situation where regionalization is allowed to develop on a self- selecting basis. Other possible forms for regionalization can be envisioned,, but each carries with it certain inherent disadvantages that might have made it less viable than the voluntary self-selection process that actually took place. 1. Regionalization along State Lines - Regionalization mandated .in accordance with the political subdivisions of our country would perhaps be the most logical alternative to the present system and would have had the potential for securing more support from the political establishment in the governors' offices and in the states' departments of health. However, there are obvious drawbacks. Some of the large states cover areas with vastly different medical, economic, geographic, and demographic characteristics. New York City, for ex- ample, is far different from upstate New York; to have included the III-4 Arthur D Little, Inc entire state under the aegis of one program would have been immensely difficult and would have done justice to neither area.* In addition, a gulf often separates state health departments and pri- vate medicine. If a regional medical program were to come too heavily under the influence of the states' departments of health and "state politice," its potential for attracting the interest and constructive .attention of organized medicine would have been drastically reduced. 2. Reaionalization on a Medical School-by-Medical School Basis Regionalization centered around medical schools would have come closer to the original model suggested by the President's Commission on Heart Disease, Cancer, and Stroke, but, if attempted, would also have faced many basic problems. Many local physicians would have reacted in terms of the town-gown syndrome that exists pretty generally through- out the country. Furthermore, regionalization around medical schools and their teaching hospitals would probably have required substantial restructuring and rearranging of the relationships between medical schools and community hospitals, a prospect not necessarily welcomed by the latter. In addition, in most large urban areas with several medical schools, the "turf" overlaps, which could exacerbate competitive problems. We should note, however, that in some areas of the country such as California and upstate New York, the geographical distribution of medical schools does form a reasonable basis for regionalization. In other areas, relationships between medical schools and other health care institutions are tbinly developed. The problems of sorting out connections between peripheral hospitals and medical schools, severing some and establishing others, might well have posed an impossibly long and frustrating task. In summary, while there are ways in which regionalization could have developed on a consistent nationwide basis, each appears to have carried with it major difficulties in implementation, a critical risk of alienating practicing physicians, or both. In retrospect, the way that regional boundaries grew up under the RMP was functionally effective, though it may have looked chaotic at the time and certainly has resulted in "regions" put together for diverse reasons. Our general conclusion is that regionalization could have hap- pened in none of the other ways outlined because, in each case, the new program would have been viewed as beholden to a pre-existing activity, already well understood, and then would have been dealt with accordingly. The legislative process through which PL-89-239 emerged helped to keep FM free from such entanglements. The subsequent administrative history also helped: the switch of RNP from NI)i tothe new HSMHA structure made more valid the concept of RNP as a relatively free-floating entity that could be trusted not to reflect any of the familiar federal or private health or medical interests too narrowly. Of course, the change also left RMP without a strongly entrenched, well recognized champion. It is possible that some day the division of upstate New York into four regions will be viewed'as overdoing the recognition of differences. III-5 Arthur D Little, Inc B. REGIONALIZATION EFFORTS VIEWED FROM WITHIN THE REGIONS We see three archetypal patterns of regionalization being de- veloped, or at least employed, to varying degrees in the regions: (1) The center-periphery model, (2) The nucleation or subregionaliz-tion model, and (3) The centerless network model. None of these models exists in a pure state in any region; they are not necessarily mutually exclusive ways of carrying on the 'rocess of p regionalization. One (the center-periphery model) has not even been at- tempted on a region-wide basis in some of the regions, though we repeat that it was the original model f6commended by the President's'Commission on Heart Disease,, Cancer, and Stroke. These models should be discussed in some detail since they representthe patterns we have observed in RMP practice. The importance of these three regionalization models lies in their very different political implications. Specifically, the models differ significantly in the degree to which they force acceptance of power concentrated in one place as a precondition for anything else to happen: o Center-periphery regionalization defines the "peripheral" elements as subordinate in some respects to a more powerful center; a Nucleation or subregio-.ialization is ambuguous as to the con- centration of power; 9 The centerless network is a guarantee that power will be con- centrated only by consent of the governed, consent being granted under circumstances in which the governed have a reasonably good idea of what they are consenting to. Other differences exist and these will emerge in the descrip- tions and illustrations that follow. But one factor characterizes most of the regionalizing experience of RMP: at present, nobody is in a posi- tion to enforce center-periphery regionalization, and almost nobody wants it to happen except on the assumption that he will be identified with the "center." The other two approaches to regionalization, as will be shown, represent attempts to make feasible an otherwise unworkable model. 1. The Center-Periphery Model of Regionalization a. Structure and Operation The center-periphery model, based on a center of excellence (gener- ally or in terms of certain specialized resources) and related peri- pheral institutions, was developed and ramified in the Report of the President's Comission and is the basic conceptual model which many III-6 Arthur D 1-ittle, Inc health planners adopt when they think about "rationalizing" the health care system of the United States. The model is easy to visualize and is grounded in the logic of equating the level of care needed and the capability of the resources to give it. It is designed to develop graded health care delivery, education, and research. Small hospi- tals on the very edge of the periphery* typically provide routine primary care as well as certain kinds of specialized care which have to be located close to the population being served; for example, in- tensive coronary care or emergency and obstetrical services. The cen- ter of excellence, on the other hand,, is devoted to high-technology medicine and clinical research, and is familiar with, and qualified in, difficult, expensive, complex, and highly specialized procedures. Intermediate facilities for commonly experienced problems requiring equipment too costly for the periphery, but using procedures so well established that they do'not have to be confined to the research cen- ter may also exist. The center, in this model, is a teaching insti- tution where doctors are exposed to difficult and rare medical cases. The model is built on the principle of hospital-based, acute-care medicine as viewed from the perspective of the academic medical center of the early 1960's. The fLow of patients in the completely developed center-periphery model is inward and upward as the severity or complexity of ailments increases. For example, in surgical terms, given the conditions of the 1960's, appendixes are removed at the peripheral institutions (community hospitals) and hearts are repaired at the medical center. Routine X-rays are taken at the periphery, and neuroradiology and angiography are performed at the medical center. Intermediate pro- cedures, like hemodialysis, may be carried out at larger community hospitals. The flow of information and expertise in this model is in the op- posite direction, i.e., outward and downward. Techniques which are developed or refined in the center are disseminated to hospitals and practitioners at the periphery, usually through a program of continuing medical education or communications media such as newsletters, tele- phone tapes, closed circuit television presentations, and the like. Planners collect information about regional resources, the skills available at each level, and other kinds of data needed to ensure a rational, orderly, sensible flow of patients and techniques, and this information is shared with physicians and administrators. In this model, continuing education programs bring doctors from the periphery to the center for refresher training. Perhaps missionaries or licircuit riders" are used to participate in rounds and perform other kinds of teaching activities in community and local hospitals. Distance between center and periphery here refers to size and sophisti- cation of the hospital: geographical distance from the center of excel- lence may play a part, but is not the governing factor. III-7 Arthur D Little, Inc b. Purposes Served by Center-Periphery Regionalization The linking of peripheral institutions to the great teaching centers, which the center-periphery model encourages, can increase the attractive- ness of internships and residencies in the outlyiag community insti- tutions because of the academic affiliations. House officers, typically in short supply in community hospitals, can assist in the work of the hospital at relatively low cost and provide the medical staff of hospi- tals with a climate of intellectual challenge that is not present with- out them. In return, the medical center can insist on exercising some degree of control over the clinical training and operation of the peri- pheral hospital. v Another intended purpose of the center-periphery model is to bring about a more rational allocation of resources, meaning the avoidance of unnecessary duplication. Tlle example of radiation therapy facili- ties is frequently cited in this connection. Similarly, physicians can usually agree (and laymen can easily understand) that open-heart surgery, organ transplants, and other complicated procedures should be carried out only in those institutions where the volume of work will be sufficient to keep the surgical teams "tuned up". The center- periphery.model allows people to address this issue directly by deter- riining where in a region particular kinds of work will be done and providing a kind of template for the construction of new facilities. The rational allocation of resources postulated in this model is highly compatible with the interests of the medical centers and associ- ated physicians who need to have access to cases for teaching and re- search purposes. It also matches the interest of the public in mini- mizing the cost of facilities while ensuring access to highly trained people when highly complex procedures are needed. However, the center- periphery model can work to the disadvantage of doctors who are not affiliated with the medical center. They may feel excluded, unable to sharpen their skills, restricted in their referrals, and in some cases denied continuous access to their patients. The model may also conflict with the aspirations of certain community hospitals which are trying to become broadly capable medical centers. The patient who must travel and stay away from home when sick can also be considered at a disadvantage. In this model, institutions on the periphery do not have to feel iso- lated or constrained to work toward the costly objective of being com- pletely self-sufficient. When they are part of a center-Deriphery system, the community hospitals have access to the resources and talents of larger, more complex institutions that contain high competence in certain specialties or subspecialties and that have expensive faci- lities and equipment the peripheral hospital cannot afford. By the same token, the medical center, being assured that the routine needs of the community will be well served by the peripheral insti- tutions, can devote the bulk of its energies, talents, and resources iii-8 Arthur 1) Little, Inc. to working on solutions to challenging medical problems -- I)articulariy those judged to have teaching and research merit -- assured that their relationships with doctors and hospitals on the periphery will provide sufficient patients to meet this teaching requirement. The more tightly organized the system, the surer the referrals. In terms of power and influence, the center-periphery model has the effect of reinforcing power in the center, placing the academic medical center in a stronger position to influence reftrral patterns and to control the operation of individual hospitals to influence the alloca- tion of construction dollars, to control training and research, and to increase their staffing in the subspecialties. The shift does not neces- sarily imply an equivalent decline in the power of the individual hospi- tal or doctor. Through councils, boards, and affiliation bodies, indi- vidual hospitals can potentially exert more influence on the center than if the regional pattern did not exist, 1)ut the degree of influence and its overallsignificance depend on how the center-periphery system is organized and how the organization works in practice. Influence of the periphery over the center is not inherent in the model; some degree of centralized rationalization and control is. Peripheral institutions have to relinquish some of their independence. The center-I)eriphery system tends to promote stability, at least on the more obvious levels. It specifies -- or at least clarifies -- re- lationships, coC[ifies agreements, and prescribes and circumscribes be- havior. If hospital B ties to medical center A for teaching and patient care, the conditions of relationship are usually spelled out in some detail. It can improve the quality of care by concentrating specialized resources and talent. Its proponents view all these factors as being a more or less desir- able way of ultimately offering health services to a population in the most efficient, least expensive, and expeditious way possible. c. Experience with Center-Periphery Regionalization in the RNP We found a number of instances where relationships between teaching medical centers and outlying hospitals were encouraged through the ef- forts of the RMP, but in no place did we see anything approaching a fully developed center-periphery system on a region-wide basis. The most nearly complete examples were found in the Memphis and Intermountain regions, the latter centered in Salt Lake City. These two regions each have a single, large population center, and a single, dominant medical school closely interacting with a strong group of private physicians, many of them specialists having real interest in reaching a large popu- lation. The medical schools in both regions had traditional ties with hospitals and doctors in the surrounding areas that enabled a fluctua- ting but perceptible degree of practicing physician influence to permeate the medical schools. The RMP came to a situation, in both cases, where a fairly well-developed center-periphery system already existed in transportation, commerce, and finance, and to a significant degree, in medicine itself. iii-9 Arthur D Little, Inc In other regions where center-periphery regionalization was at- tempted -- or appeared to be attempted -- it usually met with con- siderable resistance or was converted into some other form of region- alization, such as subregional formation, as a reaction against the perceived imposition ofthe center-periphery model. Some examples: (1) In Connecticut, the State Medical Society, despite its involvement in the formation of the RW and thp development of its program, posed serious objections to the attempts of Dr. Henry Clark, the Coordinator, to develop a center-of- excellence model of regionalization and a "Third Faculty" based on community hospitals. In part, the Medical Society was opposed to the "planners -- who favor a system of cen- tralized, academic, and theoretical management of medical affairs, and, further, who evidently contemplate using non- voluntary leverage to impose that system on the Connecticut professions. . ." * Over the succeeding two years, the State Medical Society and the Connecticut RMP have become closer, but the originally proposed "grand design," incorporating affiliations between the Yale-New Haven Medical Center and the.35 community hospitals in Connecticut is still a long way from materializing. However, some ot@er kinds of re- gionalization have begun to appear. In several instances, for example, community hospitals have initiated joint planning efforts with neighboring hospitals to provide community servi- ces. Moreover, there seems to be a reasonably broad ac- ceptance of the subregional division of the state into 10 health service areas, though no subregional RNP organization has yet been formed. (2) In the Greater Delaware Valley (GDV-RNP), Philadelphia, with its six medical schools is explicitly referred to as "the Center." Everything else is in "the periphery." But there has been relatively little success so far in the attempt to build a center-periphery system between the academic centers and community hospitals outside Philadelphia. In fact, even regionalization planning has not been completed and accepted in any depth. People outside Philadelphia tend to resist domination by the center city in health care as in other sectors of activity. In part as a reaction tothe perceived power and dominance of the medical schools, and in part as a planned strategy, area-wide planning groups are emerging and are being developed by the GDV-RNP. While it is too early to say whether the "areas" will develop to the point of representing a substantial force to interact with the medical schools in Philadelphia, they have gained positions on the governing board of the GDV-RMP. Explicit center-periphery Correspondence from the Connecticut State Medical Society to the Division of Regional Medical Programs. III-lo Arthur D Little, Inc regionalization seems still possible in the Greater Delaware Valley, but less likely than regionalization in other forms. (3) In the Northlands Region, despite the existence of two geo- graphically separate centers of obvious excellence -- the Uni- versity of Minnesota Medical School and the Mayo Clinic -- no significant region-wide attempt has been made to implement a center-periphery model of regionalization. Historically, re- lationships between the University Medical Center andthe great majority @f community physicians and hospitals have been weak. The University was looked upon as a place which would never let you know what happened to patients you had referred there. Until recently, there was little noticeable outreach from the University Medical Center as viewed by physicians in the countryside. With Mayo, the situation is somewhat different. For years Mayo had a policy of cultivating relationships with community physicians in Minnesota and nearby states, and it has built its referral network carefully. But still, in nearby communities the local physicians fear being overshadowed by Mayo. In either case, had there been an assertion of "centrality" through the RMP, the community hospitals and physicians would al- most certainly have been alienated from the program. As it hap- pened, both Mayo and the University agreed that they should not control the MT, though neither took a totally passive role. Action an their part was imperative if hospitals, physicians, and allied health personnel were to be expected to assume active roles in the process of regionalization. Recently, the University Medical Center has undertaken some activities which, as they succeed, could lead to closer rela- tionships with community physicians. These include a family practice curriculum at the Medical School and the active seeking of referrals throughout the state. (4) In North Carolina, the Charlotte Memorial Hospital resisted the idea that it might be a "peripheral hospital" with respect to any or all of the three university medical centers in W71nston- Salem, Durham, and Chapel Hill. As a reaction it attempted to pull together the medical resources in Mecklenburg County, thus precipitating another form of regionalization -- a reinforce- ment of outreach and center-periphery development with Charlotte Memorial as the hub. While comprehensive center-periphery regionalization has not been a widely successful RMP strategy, nevertheless there have been a number of instances in which RMP has facilitated 'university medical center out- reach. None of these instances blankets a region. Those outreach efforts that have the broadest and most consistent coverage tend to represent special purposes. By no means do they intend to effect com- plete center-periphery regionalization$ but they do reflect the theory that there is a "center"'and that it can relate to entities outside Arthur D Little, Inc itself in an outreach mode and in such a way that information flows from the center to the periphery and referrals flow from the periphery to the center. The following examples describe projects that began in a form capable of becoming part of a center-periphery system, though some soon took courses that precluded that possibility and none depends on the full articulation of center-periphery regionalization for its viability. While awareness of what a center-periphery region could be is one of the factors that shapes these efforts, it is not the only factor and is usually not the dominant one. o In North Carolina, RNP supported the "Berryhill Project," which (as one of its several activities) linked the University of,North Carolina Medical Center to the large community hospital in Wilming- ton, North Carolina. Through exchanges of physicians, ties between the University and the locaL doctors in Wilmington were developed. The project enabled the local Wilmington physicians to visit and profit from the technology and expertise available at Chapel Hill. More surprisingly, in view of the original continuing education objectives stated, it enabled faculty members from the North Carolina Medical School to learn something of the very real excel- lence of health care in Wilmington and of the practical realities of first-line care. o In the Northlands Region, RMP is supporting Mayo, the University, and the American Rehabilitation Foundation (ARF)* in developing relationships with three distant parts of the region to introduce and develop stroke rehabilitation. Mayo is taking responsibility, roughly for the southern third of the state, ARF for a broad band in the middle, and the University of Minnesota for the northern part of the state. o In the southern Minnesota communities of Austin and Albert Lea,, RNP is helping to support merger discussions between the community hospitals. o Physicians in Austin, some 40 miles from Rochester and the Mayo Clinic, are actively considering ways of using Mayo as a diagnostic resource (perhaps through closed-circuit TV), whereas formerly Mayo had been viewed largely as a competitor for patients residing in the northeastern part of Mower County. In most regions we have visited, the center-periphery model was never considered, or, if considered, was immediately rejected as an RN]? strategy for the region as a whole. Maine, for example, stimulated by the Bingham Associates Fund,- had had a long potentially "regionali- zing" relationship with the Tufts-New England Medical Center in Boston. But the Maine RMP never seriously considered,developing its own center- periphery system around the Maine Medical Center in Portland because experience with the Bingham Associates Fund, while generally positive, had not convinced local doctors of the advantages of close (subordinate) connection with a Medical Center, or even of its feasibility. As the Formerly The Sister Kenney Foundation III-12 Arthur D Little, Inc hub of a regionalization scheme, the Maine Medical Center, in 1967 the most obvious candidate in the state for the "center of excellence" title by virtue of its size, staff capabilities, and teaching pro- gram, was no more acceptable than Boston in the eyes of local doc- tors and other leading community hospitals in @Aaine. For people in Bangor and Augusta, going to Portland for medical care was seen as undesirable -- and not justified merely to satisfy the theoretical advantages of "regionalization." We have chosen to describe Maine because of its earlier, somewhat related experience with the Bingham Associates Fund, and because an attempt to regionalize in some sense had been made. But the experience in Maine was no different from that in other places where there had been no prior experience with center-periphery regionalization: it simply did not match the perceived needs of the medical profession or their patients. In California, to select another quite different example, it was abundantly clear that no one medical school could be the model center, so the region could not have a single center-periphery system. It was equally clear that two center-periphery systems, one for the North and one for the South, would probably exacerbate the political and economic divisions between the two areas. The California RMP therefore settled on a division into eight subregions (nine, if Watts-Willowbrook is included as a subregion), each with its own medical school. Whether the activities of these subregions were themselves to develop into "center-periphery" regions (in any way except that each subregion is equipped with a medical school) was left for the subregions to decide. d. Some Conclusions about Center-Periphery Regionalization Among the things we can learn from RW experience with this form of regionalization are the following: (1) While center-periphery regionalization may not become the strategy of choice in a region -- not even attainable if chosen -- the suggestion of this model as a possibility, or steps taken in that direction, can precipitate other forms of regionalization. For instance, the most common response is a defensive reaction. Health providers goaded by the threat of center-periphery regionalization decide to band together -- at least among themselves -- in some other posi- tive cooperative arrangement. Thus, if an RMP coordinator can use center-periphery regionalization as a concrete starting point, he may well precipitate real movement, though not in the direction first indicated. (2) When attempted, center-periphery regionalization almost always remains limited in terms of its content and of realized relationships between the center and the periphery, even when it is pushed hard. What makes sense in terms of center-peri- III-13 Arthur D Little, Inc pheral regionalization in a single category (open heart surgery, medical information retrieval, radio networks) may not fit very many other categories. "Islands of excellence" can and do exist almost everywhere, and "centers of excellence" can and do contain extensive "islands of mediocrity." Accordingly, it is often unreasonable to extend a perfectly plausible center- periphery regionalization scheme based on one service, tech- nique, or type of facility to others. What we see in practice is thus usually limited to one purpose. It us;ually consists of a program in continuing medical education, with perhaps some coordination around a tumor registry, a DIAL-access system for information on the categorical diseases, or perhaps"the recognized leadership of the center in some particular aspect of heart disease, cancer, or stroke. This may be all to the good, but it is hardly what "regional planners" have in mind as a goal when they think about comprehensive regionalization. (3) Physicians on the periphery, by and large, tend to resist domination by a university medical center, particularly after they have been in independent practice for a while. To this group, it is their work that constitutes the center, and the university with its principal hospital facilities is merely a handy place to which they can refer patients from time to time for really specialized attention. The failure of a university medical center generally to recognize, acknowledge, and under- stand the centrality of the private physician as he perceives it, and its frequently observed arrogance in asserting its own centrality combine to inhibit the possibility of deep, rich, mutually beneficial relationships between the center and the local physicians. A monolithic view does not correspond to the realities ofthe medical system in this country. At the very least', this suggests the need and possibility of using a dif- ferent description of the model, in which centrality is ac- corded to people, professions, and institutions where they are central. In one view, the academic medical complex is a backup facility in terms of patient care, witl special capabil- ities to deal with exceptional problems. It is central (though not quite exclusive) for devising new approaches and techniques, and a very important resource for providing training and faci- lities to medical and health care students and trainees (in- cluding house officers, etc.) (4) The hospitals' need for and tradition of independence, particu- larly from the university-based medical centers, is important. Trustees, doctors, and administrators of community hospitals can view their institutions as delivering a different kind of care from that offered in the academic medical centers. However, the difference is not only a matter of size or degree of technological sophistication; in fact, many so-called com- munity hospitals that are not affiliated with university medical centers are quite large, employ the latest in sophisti- III-14 Arthur D Little, Inc cated technology, and are staffed by highly trained specialists and subspecialists. The major difference -- as perceived by some community hospital people -- seems to be in the approach to health care. They view the university medical center as being too preoccupied with teaching and research at the expense of the patient. "The medical school treats diseases; we treat patients." This is admittedly harsh, overgeneralized, and in its own way a stereo- type, but it does point out one of the major reasons why some community hospitals tend to be reluctant to affiliate too closely with university medical centers. t Many community hospitals fear that medical center control in an early phase of centralization might alienate general practi- tioners and other referring physicians who are depended upon to fill community hospital beds. They fear that medical center control of bed utilization would create unsolvable medical and financial problems for the community hospital. Finally, there is a sense that their own independence provides control over the very life of the hospital when it comes to fund- .raising and building programs. This independence will be sur- rendered only reluctantly, unless offsetting payoffs are clearly in evidence. RNP has little to offer to reassure the community hospital that questions center-periphery regionalization (except in the special- ized sense noted under (2) above). It would, in general, re- quire a good deal of,energy, money, and power to convert existing community hospitals to the new role that such a scheme calls for. In fact, if this were the only path to regionaliza- tion resistance to RMP by the community hospitals would be understandably high. (5) 'Until recently, there has been no central coercive authority to compel hospitals to relate to each other or to a university medical center. Even now, such an authority exists only in a few places. For example, in New York State, a hospital planning council has statutory power to veto hospital construction plans, and something similar has been created in California. In most other states, the Ilill-Burton agency can encourage, but is reluctant to coerce, hospital collaboration. Most comprehen- sive health planning organizations also lack authority, statutory or other, to compel patterns of regionalization. Given the conditions just mentioned, true center-periphery regionalization strongly involving or principally based on community hospitals is presently a practical impossibility. There is no reason for hospitals to accept it voluntarily, and as yet no sanctioned authority to force them to do so. III-15 Arthur D Little- Inc (6) The center-periphery model tends to be built around in-patient care and facilities -- hospital facilities -- because centrality is defined in part in terms of technological sophistication, and this means facilities. Not obviously included in the center- periphery model of hospital facilities are the voluntary agencies, the state health departments, physicians having primarily office practices, visiting nurses agencies and associations, dentists, and nursing homes. Thus, the model tends to enlist only partial involvement of the health caie delivery system; it is not all- inclusive. This is ironic in view of the patent conflicts between those elements of the system that it does involve. (7) A shift to the center-periphery mode of regionalization would require that at least some physicians in a region alter their referral patterns, a change that is very difficult to enforce Referral patterns are,--after all'based on a complex of factors,- and what will actually ensue-when one sets about to change physician relationships is hard to predict. The results could easily be the opposite of what was intended. (8) It is significant that the primary functional relationships between the institutions involved in the center-periphery .regionalization model are generally perceived to be one-to- one. Aside from local coalitions around such aspects of medi- cine as radiation therapy, the impetus in the model for collabora- tion among institutions away from the center seems minimal. The attempts that have been made to implement the model accordingly have depended on most of the energy coming from the centers, something that has been politically impossible and financially unsupportable. (9) While center-periphery regionalization can attempt to upgrade the quality of care, it can do little in any material way to directly affect the availability of and accessibility to care. There has been a distinct shift away from the diagnosis that led to RMP in the first place: the judgment that more highly scientific medicine was the primary requirement for solving the health care needs of the nation. There has been a growing con- cern about the national medical system -- concern for rising medical costs, the effective exclusion from the health care system of large numbers of disadvantaged people, shortages of medical manpower, and the difficulties of getting care when it is needed, even by ordinary middle-class people. (10) Finally, the poor dividends in terms of availability and accessibility of care from the huge investment in Medicare and Medicaid have begun to convince observers that no amount of investment in payment for care will suffice by itself to introduce necessary changes on the provider side. Clearly, people are saying that direct intervention of some kind on the provider side is needed as well. III-16 Arthur D Little, Inc Very powerful objections either to nationalized systems of care or to enforced decentralized solutions, such as wholesale reliance on commtmity-based group practices, continue to appear. Yet shortages of medical manpower suggest that changes in the system will have to work with largely existing personnel, and the immediacy of the need dictates heavy dependence on existing institutions. This means, to a great extent, attempting to facilitate (voluntary) rearrangements of existing providers and institutions. We have already reviewed the forces that make center-periphery regionalization difficult in these terms. 2. The Nucleation Model of Regionalization: The Characteristic RNP Regionalization Process a. Description Regionalization through nucleation refers to a pattern of linkages among providers, such as community hospitals, in which one of them typically provides most of the initial unifying energy, but in which the agreement to unify is derived from mutual self-interest rather than a dominant coercive force. While the initial source of enthusiasm, initiative, and ideas may act as a "center," the real key to the process of nucleation is response and reciprocation from the "nuclei." The signals sent out by the "center" at first are attempts to get inter- action by identifying interested listeners, determining the listeners' interests, and creating an arrangement for doing something of mutual benefit. This implies that when action begins it may look very dif- ferent from what the listeners originally had in mind. As contrasted with the center-periphery model, the nucleation model can be incomplete; that is, unlike the center-periphery system, it need not aim ultimately to include every significant health care provider in a particular class in a particular region. Thus, several nuclear patterns can coexist in one region without necessarily being related in a self-consistent whole. Nor does nucleation have to in- volve the entire region. In a nucleation model, as in the center-periphery model, the center can be a primary locus of power, but the difference in power between center and its related nuclei is not so large as in the center-peri- phery system. In the nucleation system, the basis of power is the ability of one element (perhaps a hospital, perhaps RNP itself) to convince other elements to come together for a common purpose. The process of union can be accomplished through negotiation, mutual collaboration, or defensive alliance -- and the only imposed power may be that of calling meetings and stimulating committee work. A subregional structure (organizationally and/or geographically) may be the outcome of a process of nucleation. Alternatively, the establish- ment of subregions or an attempt to effect center-periphery regionaliza- ti.on by the RNP or other agency may precipitate nucleation. But geo- graphical saturation --- or division of the turf --is not essential. III-17 Arthur D Little, Inc For example, a continuing education program linking a medical center and a distant hospital by mutual agreement can be a form of nucleation, even if both institutions are also forming separate linkages with still other agencies. The nuclear pattern of regionalization can be illustrated thus: F OB E D where element "A" is the nucleators i.e., the one originally causing the linkage to develop, and the one which tends to retain the initiative with respect to determining the nature and purpose of the linkages, especially in the early stages of their development. in the case of community hospitals, "A!' P)ay develop into a "regional" hospital. In the case of referral patterns, "A" may be a multi-specialty, hospital-affiliated group practice, with B, C,D,E, and F as referring physicians or groups. But "Alt may also be an RMP core staff member, with little or no license or competence to intervene in the substance of work done once the link- ages are perfected. His competence and role may instead be directed toward making the links among others work effectively and acceptably. The process of nucleation within regions is closely analogous to the process.by which the RMP regions themselves were established-. "Central" sometimes identified and sometimes encouraged the local creation of alliances, joint purposes, momentum, boundaries, and budgets, but the energy was for the most part locally generated. Furthermore, on both levels, "regionalization" is seen as in process as soon as a nucleus exists somewhere in what was otherwise an empty (i.e., unknown or hostile) "cell." Reliance on nucleation seemed to be the most practical way of getting the program started across the country; reliance on nucleation turned out to be the most acceptable way to launch the program regionally. Some of the circumstances favoring it were: The lack of coercive power in the hands of the regionalizers; o The availabil-ability of modest monetary incentives through the regionalizers; III-18 Arthur D Littk Inc Voluntary cooperation as the watchword; Existing isolation among persons, professions, institutions, and communities; Much disagreement on the proper shape of the future; Expectations for "planning" on the part of both friends and enemies; 0 Some clarity about ce-nter-periphery regionalization, at least on the conceptual level, and no alternative commonly acceptable view of regionalization. The result on all levels has been a "downward push," i.e., the delega- tion from each level to.the next more localized level of the privilege of discovering whether the people there could create genuine center- periphery regionalization, or get it to take hold. By the time this kind of subregionalization is experienced, it becomes apparent that a broader, more inclusive, or region-wide center-periphery system can-. not be made to work -- at least not voluntarily or immediately -- with the limited energy that is available to stimulate its development. Nucleation, then, is not only a political fallback position from center-periphery regionalization; it is also a method for exploring the extent to which center-periphery regionalization can be pushed locally through subregionalization. Even when fairly active nucleation is going on, it is likely to be criticized as "opportunism" or regionalization without planning. Ad- mittedly nucleation, when encountered, may evidence a "cop-out" rather than a virtuous choice. The issue is whether nucleation is the strategy of choice, a substitute for strategy, or something that happened in the course of spending money. This becomes an important question for the evaluation of RNP regions, and our views on how to deal with it are laid out in Chapter V. Of the 18 regions we have visited, not one seemed to us to_be using nucleation blindly or randomly; almost all used it because nothing else that was believed legal worked as well. Some had begun to treat it as a primary process toward regionalization, and to think of it in quite conscious and even positive terms. As our examples suggest, it was a powerful part of the North Carolina strategy during Dr. Musser's tenure two years ago (when we were there). It is a highly conscious aspect of the Northlands approach, and it is what Greater Delaware Valley is in fact doing under the label of "center-periphery regionali- zation" as it subregionalizes. In New Jersey, the fourth of the regions we studied intensively, nucleation has been attempted with community hospitals, has been made to work with the Model Cities Program in a number of urban areas. The RNP-Model Cities health planning coordinator is at first himself a "nucleus," works to identify and develop additional "nuclei," and aims to strengthen what he builds around him eventually to the point that it can do III-19 Arthur D Little, Inc without him. We know of no region in which nucleation is not the major process going on, whether consciously viewed as instrumental to a regionalizing strategy or otherwise. Onthe level of entire regions as established in RMP, in our judgment nucleation is the most widespread and characteristic regionalization process tried to date. Accordingly, it is very difficult to understand stand how the process is accomplished, a subject touched on repeatedly in this report and dealt with more systematically in Chapter V. It is also important to understand where it leads, which is the subject of Section C of this chapter, "Systems Transformation and Regionaliza- tion." b. Purposes Served by Nucleation The nucleation model of regi@iialization can have many reasons for coming into being and, when formed, can have a variety of effects. Below are described some of the most common purposes and outcomes of nucleation with the awareness that others can be derived from the specific experiences in the individual regions: (1) Nucleation can be a reaction against a real or perceived threat of dominance by a powerful medical center. The example given earlier of the Charlotte Memorial Hospital taking steps to form a more tightly knit region around it- self is one form of regionalization by nucleation, begun as a reaction to what was feared to be an assumption of more power by the three medical school teaching centers in the state. (2) Nucleation can be a way of sharing the elements of a job that none of the participating institutions could easily do alone. Radiation therapy linkages can form the basis for nucleation, in which the expensive facilities are located in one hospital with the consent of the others. Joint labora- tory facilities are another form. (3) Nucleation can be an attempt to rationalize the referral patterns in a particular area so that a minimum of unnecessary referrals get made outside the subregional system. The re- lations between the State of Franklin and Asheville, North Carolina, are a case in point (see Volume III.) Historically, referrals were made from the State of Franklin to Charlotte, 150 miles away, or to the teaching hospitals of the medical schools further to the East, rather than to Asheville, which is much closer. People in the area talked about a conflict of interests between the physicians of the westernmost counties and Asheville. Recent subregional activities, based on the discovery of common interests,, are beginning to link the western counties to Asheville for planning purposes and have caused some shift in referral patterns to that city. (In this instance, III-20 Arthur D Little Inc the change in referral patterns will be accompanied sooner or later by changes in training and recruiting pattemsfdr para- medical personnel and new developments in hospital administra- tion that will make this example ultimately look more like our third type of regionalization process: "the centerless net- work.") (4) Nucleation can be a way of getting a needed job done at the local level. The discovery by RMP of a person or group in the region who is willing to take the lead in joining with others to do some work can be the start of a nucleation process that first appears as a project request. Such nuclei have sprung up, for example, in the Memphis region, notably at luka, Mississippi, and Greenville, Tennessee. (5) Nuclear regional-ization can result in the creation of new loci of power in the region as a whole, which can cause shifts and changes in the health care system of the region. In the GDV-RNP, as mentioned previously, the formation of subregional areas created, potentially, new sources of power which may counterbalance some of the power of the medical schools in that region. (6) Perhaps the most obvious outcome of nucleation can be a reduction in the degree of open competitiveness among the providers, the coalescing of parts of a region, and the formation of new allian- ces where none existed before. c. Subregionalization and Nucleation Subregionalization is perhaps the most common way of expressing either the fact of nucleation or a plan for achieving it. In numerous regions North Carolina, Connecticut, Northlands, Greater Delaware Valley, and Georgia, to name a few -- subregionalization has been used as a way of working toward greater inclusiveness and of tying together parts of a region. In North Carolina, the subregionalization scheme, which was developed as a result of an exhaustive demographic survey and proposed as the basis for planning, caught on in two out of the six named sub- regions. The other four were not immediately ready to declare them- selves subregions. Charlotte, for reasons mentioned above, was in- volved in one of the subregions which declared its intention to proceed as a unit.' The State of Franklin, together with Asheville, was another. The subregional scheme was not the sole basis for the coalescing which took place, but it did provide an additional context and rationale. A subregional pattern for Connecticut has been accepted by nearlv all of the major health planning bodies in the state including CHP, RNP, Hill-Burton, and the state's Public Health Department. This has provided a context for joint planning initiated by some of the community hospi- tals, such as in the Stamford area. III-21 Arthur D Little, Inc d. Some Conclusions about Re ionalization by Nucleation Regionalization by nucleation represents a genuine alternative to the center-periphery system of regionalization and, in fact, may emerge as a reaction to the perceived imposition of that system by RMP, or a medical school, or a medical school consortium. The encouragement of nucleation by RNP can lead to the discovery of the "movers" in the region and the potential for developing new foci of power. These, in turn, can lead to opportunities for c@an e in the system if shifts in 9 the power balance are a prerequisite to change. The change can take any form, including having the medical centers pay more attention to the community physicians and hospitals. I If nucleation is a strategy of choice for the coordinator and his Regional Advisory Group, there are some specific tactical moves he can make to promote that strategy, (1) He can actively seek out the "movers" in a region and en- courage them to take the lead in forming new alliances with others; (2) He can encourage the development of subregions (particularly if they emerge out of local activity) and provide for represen- tation of those subregions on RMP boards, committees, and ad- visory groups; (3) He can provide staff support to a nucleus which is trying to form links and can support negotiation processes which occur in the nucleation process; (4) He can push for project approval guidelines that call for linkages between two.or more elements and require that one of the elements be the "leader" or grantee in the application; (5) He can ask for sanction for emerging regional hospitals from his Regional Advisory Group or other power groups in RNP, al- though he must be careful that an attempt to invoke sanctions does not irrevocably damage the ability of the aspiring re- gional hospital to form links with other hospitals; and finally, (6) He can encourage various health provider organizations to take the lead in building coalitions for planning and care delivery. The discerning reader will have noted that: 9 Some examples of center-periphery regionalization slide into nucleation (reactive nucleation, as in Charlotte Memorial Hospital; the attempt of Greater Delaware Valley to create subregions that are peripheral to the Philadelphia center.) In fact, nucleation is the most likely result when center-periphery regionalization is undertaken. III-22 Arthur D Little, Inc. 3. The Centerless Network Model of Regionalization a. Description As in the nucleation model of regionalization described above, the centerless network model refers to a pattern of linkages based on mutunlly perceived needs or goals among elements of the health care system -- doctors, hospitals, voluntary societies, medical schools, nurses associations, and the like. The difference arises in the lack of any recognized center or nucleus of power. A true "collaborative arrangement" between any two or more provider elements could be called a centerless network. No special power is attributed to a "center" in the network; in fact, coercive power may be explicitly built out of the network. Like nucleation, the centerless network does not aim to be all-inclusive of the providers in the region. The primary quality which differentiates it from the nucleation model is that power is more completely shared, and in all probability the pattern of cross linkages among the elements of the networks is richer, i.e., more of the possible two-way links actually exist. While the nucleation model may look like this: OB E c D the centerlesss network model may look like this: III-23 Arthur D Little, Inc It may not be clear at first glance whether an existing set of linkages is a nucleation model or a network model; and indeed it may be a matter of perspective. What looks like nucleation to a com- munity hospital of 300 beds may look like a centerless network to the 50-bed hospitals around it, which do not acknowledge -- and were not asked to acknowledge -- the hegemony of tne larger institution. But removal of the nucleus from the nucleation model may de-energize or dissolve all the linkages, whereas in the centerless network model there is no single focal point and the network may be better able to sur- vive the withdrawal of one element. b. Purposes Served by the Centerless Network The purposes and outcomes of a centerless network are essentially similar to those described for_ the nucleation model. It can serve as: o A means of reducing competitiveness, coalescing specific rela- tionships within part of a region, and establishing new alliances-, o A reaction against medical center dominance or other assertions of impending center-periphery regionalization; o A way of jointly getting a job done; o A rationalizing of referral patterns; 9 A means of generating RNP projects; o A means of building new loci of power (in this case the network itself); and o A means of offering maximum opportunities for identifying and developing leadership on a minimum risk basis, by allowing leadership to shift quickly from point to point in the network. c. Network Development in the RNP The coalition of community hospitals in the State of Franklin in western North Carolina represents one of the best examples of center- less networks we have encountered. (See Addendum 1 to Chapter IV). These hospitals, all small and increasingly competitive as their isolation was gradually reduced, banded together to seek a group accreditation from the JCHA. No one of the hospitals has either as- sumed or been given permanent leadership of the consortium. It took the threat of disaccreditation, the facilitating presence of an RMP core staff member, and the vision and skills of a number of local leaders to provide the impetus for building this network against the background theme of the "State of Franklin" viewed as a cohesive region. III-24 Arthur 1) I itt le, Itic. A second example was found in another rural area, as described in Addendum 2 of Chapter IV. A group of institutional provider represen- tatives banded together to plan action programs in health care delivery for a poor territory, somewhat skimpily blessed with health care re- sources. The group itself was bent on program development but, at least in 1969, was not anxious to undertake program management. Its intention was to develop a communications network and channels through which choices could be made and local efforts mobilized to: o Create means for visiting nurses and private physicians to work together and much more extensively; 9 Tie the community hospitals together for sharing an"increasing range of services; and o Solve emergency-servzices and 24-hour coverage problems. The Regional Advisory Group is the level within RMP from which the idea of the centerless network most often evolves, and from which a region-wide network building process can develop. Each of the committees and task forces of a typical RAG can itself act as a centerless network if the chairman chooses to serve as a moderator-stimulator or if an RMP core staff member provides facilitative staff support. In turn, the committees and task forces can collectively become elements of a greater centerless network -- a network among RAG members and others outside the membership, such as the Board if there is one, and local action groups where they exist. Leadership is temporarily accorded the committee currently in the state of greatest activity; for example, the New Jersey Task Force on Urban Health Problems in 1969, the Northlands Committee on Continuing Education in 1970 as it created goals for itself that both reflected and clarified the goals for the region as a whole, and the subregional groups in the Greater Delaware Valley that pressed for revision of the Program Committee (Board) to include more local representation in 1969-1970. d. Some Conclusions about the Centerless Network While the center-periphery regionalization model is the most easily understood concept for visualizing a regionalizing process, and the nucleation model is the appropriate process for developing a feasible strategy for a region ag a whole, the centerless network is the most effective process for building local support for new ideas. It turns out in practice, that centerless networks help people to reach mutually acceptable decisions on a course of action, and to build the consensual commitment which is the only legitimate power RNP has available to it. In the process of building centerless networks, RMP core staff members usually play active roles, serving importantly as catalysts and faci- litators in the development of these networks. Typically, in the absence of a nuclear leader or institution acting to regionalize for its own purposes, some outside'presence is needed to call together the parties to potential agreement and articulate at least the beginnings of a network. III-25 Arthur D Little, Inc When RMP can be seen as neutral (in terms of the ideological posi- tions on how health care should be delivered) in can serve in this facilitating role or as a kind of broker among the various parties. This role is described in more detail in Chapter V. But why consider centerlets networks at all? They offer an alterna- tive regionalization strategy, even more removed than is the nucleation model from the center-periphery system. On the level of projects, feasibility studies and core staff activities intended to lead fairlv directly to changes in the quality of care available to people, the centerless network model offers the RMP staff member or project appli- cant a non-threatening way of organizing and presenting his proposals for improvement in the system, and also offers him a way of suggesting regionalization" that is beyond the immediate level of the project and just as real as nucleation or center-periphery regionalization. It keeps open what the network wif! decide to make of itself as it evolves, thus quite genuinely preserving the voluntary nature of its members. Additionally, by bringing people together in ways they never experienced before, it offers an opportunity for setting up uniformly high standards in the process of which RMP can usually exert a positive influence. But, we repeat, it is at the RAG level that the management of the kind of political processes which are appropriate to RNP usually takes place, and that the centerless network has its greatest power and its most widespread application. C. SYSTEMS TRANSFORMATION AND REGIONALIZATION A major message of the first part of this chapter is that center- periphery regionalization (even to achieve categorical disease centers and the diffusion of information), cannot be achieved over very strong opposi- tion based on the. actual distribution of excellence or power. That is behind the form the legislation took when passed, and it explains why the experience of 1966-1970 has been what it has been. With the original con- cept in doubt, the question becomes: Can regionalization work on any basis? Voluntary cooperative arrangements are proving to be a viable -- if limited -- answer, since they are vehicles for: 0 Overcoming fragmentation and isolation where the effects of divisiveness are judged to be harmful; o Permitting physicians to work with other providers of health care to solve "delivery problems" on a natural organizational level -- "natural" meaning whatever level emerges from the process of seeking cooperative agreements; * Providing local forums of people who reflect all aspects of professional knowledge and practice, to work on health care issues in a context broader than any particular profession or institution or community; III-26 Arthur D Little, Inc. 0 Creating an additional channel between public and private sectors of health care for improved communications of facts, opinions, and judgments. In this interpretation, the emphasis shifts from a goal of dealing with categorical diseases to one of fostering voluntary agree- ments. The categorical disease provisions of the law are taken as con- straints guaranteeing that: * RMP will recognize the realities that physicians and hospitals confront dread diseases and that their primary function is the care of individual patients; I a RMP will approach planning and operations in terms of specifics; its plans will be bui.It up from an understanding of such specifics rather than being simply based on a political or social theory (diagnosis of a cervical cancer, treatment of a stroke patient: how, where, by whom, at what cost?); o RMP will not prescribe changes in patterns of care, or changes in the professional practice of medicine, and therefore any changes it generates will have to be undertaken voluntarily by the providers themselves. It will, within this constraint, press for the most effective standards of practice. All this means that the primary role of RMP is to facilitate voluntary rearrangements of existing institutions and creative new rela- tionships among providers, using whatever incentives (for example, project grants) it can find but having no power to coerce. RMP is en- gaged in "voluntary" systems transformation. Initially RMP activity was directed in most regions toward the center-periphery technical diffusion model. Since then, systems transforma- tion has become a primary goal in some regions and a secondary or informal goal in others; it is gaining increasing acceptance as a dominant rationale for the programs as a whole. This acceptance is, of course, far from universal. Some RMPs see themselves essentially as solicitors and screeners of proposals and have difficulty conceiving of themselves as "programs" in any broader sense. Moreover, in nearly all regions RNP is regarded to some extent as a collection of individual projects centering around con- tinuing education, training, coronary care units, and the like. In those RMPs which have consciously adopted the goal of systems transformation, coordinators face the problem of how projects initiated under earlier views of RNP should best be pursued. In order to understand RMP's real role, a more detailed descrip- tion of systems transformation is in order. 1. The Elements of Systems Transformation There is, to begin with, a set of starting conditions. These may be reckoned from the beginning of RMP, or more appropriately, from III-27 Arthur D Little, Inc the time the coordinator and those with whom he works* begin to work deliberately and systematically at the task of transforming the regional system of medical care. Starting conditions include both the status of the medical care system (the quantity, quality, and deployment of medical resources, and the state of access to them), and in a broad sense, the politics of the system (the nature of the key actors and agencies, their interactions with one another, power relationships, and the like.) The coordinator's "diagnosis" of the starting conditions is to systems transformation what "baseline data" are to efforts to affect people's health. The process in which the coordinator engages is represented by the set of activities he undertakes and the chains of consequences which radiate from them as he seeks to unite the influence elements of the medical care system. While the nature of this process may vary from region to region and from time to time, it tends to vary around certain central themes -- for example, the manner in which regional identity is built or the attitudes toward centralization of health services. These provide part of the basis for comparing efforts at systems transformation from region to region. Further, the process of systems transformation is not an end in itself, but must always be understood to be moving toward some goals and, therefore, to be subject to certain process criteria based on those ends which are equally applicable to all regions. As the RMP coordinator and his staff work at systems transforma- tion, they develop and work toward ends-in-view -- these are the specific rearrangements the coordinator seeks to achieve (fcr example, the establish- ment of a clinical outreach center connecting a center-city neighborhood to a major hospital), and they derive more or less explicit models of medical service delivery. Ends-in-view also tend to reflect the hcalth care issues that have currency at a given time in a region; for instance, the cost of medical care or the problems of nutrition in rural areas. To the extent that the coordinator's efforts at systems transforma- tion are deliberate, they imply strategies of systems transformation -- general approaches to achieving ends-in-view. Strategies draw on specific techniques(for example, the use of critical data as a means of drawing - institutions together, or the use of advisory committees to encourage joint confrontation of medical issues.) This way of looking at regional programs and the process of regionalization permits assessment of progress at several levels of change. * The actual make-up of the group of those involved in deliberate efforts to effect a systems change in a region may vary widely from region to region and from time to time. For the sake of simplicity, we will refer to this group simply as "the coordinator, 11 even though the composition of the group of those involved in deliberate efforts at change is usually complex. III-28 Arthur D Littlo Inc In specific situations it permits, over time, assessment of substantive effects on health services (whether at the level of quality and distri- bution of resources, people's access to them, or actual change in health). But where it is not reasonable to expect changecf this sort within a given period of time or throughout a region, this frame of reference still per- mits assessment of the program as a whole and of specific projects within it. At the very-heart of RMP is the goal ot assisting in bringing about improvement in the health of people. Thus any systems transforma- tion that is sought has this goal ultimately in mind. Yet the connection between cause and effect through systems transformation it usually un- clear f or three reasons: (1) There are many forces working on the system at any one time, and tracing effects back to a single cause is a practical impossibility; (2) The time-lag between the introduction of systems change and any discernible improvement in national health can be very great, and (3) There is little agreement on which measures are really representative of the health ofthe nation. But if cause and effect are matched in smaller steps, it begins to be possible to deal with the problem of understanding what is happening. To accomplish this we propose that systems transformation be viewed in terms of several levels of change. Before we examine how this may be done, however, we should look at each of the elements of systems transforma- tion more closely. 2. Levels of Change Our classification is similar to several others developed else- where, and. include: (1) Change in people's health; (2) Change in the character of delivered care, including change in people's access to care; (3) Change in the quality and configuration of care-providing resources described by Bodenheim* in part as the "anatomy and physiology" of the system, and Bodenheim, Thomas J., Regional Medical Programs: Road to Regionaliza- tion, Medical Care Review, 26 (11), December 1969. III-29 Arthur D Little, Inc (4) Change in the process of planning and interaction within the system.* Each level suggests familiar issues or statements of national (or regional) health goals, as outlined below: Level 1: Change in PeoT)le's Health Increase in life expectancy; Increased freedom from disease; and In disease-specific terms, reduction in the number of persons threatened by, or suffering from, heart disease, cancer, stroke, and kidney disease, particularly those in the most productive years of their lives. Level 2: Change in the Character of Delivered Care, Including Change in People's Access to Care More nearly equitable access to care on the part of groups of people taken to be most in need or at least connected to the system: Expectant mothers and young children; The poor, including those eligible for public assistance, unable to pay basic costs, unable to afford the costs of extended care. Level 3: Changes in Configuration of Care-Providing Resources Placing facilities where the patients are; and Changingthe services available to reflect emergent needs. Level 4: Change in Planning Processes and Interaction within the System Developing processes to minimize duplication of facilities in neighboring hospitals; and Developing joint manpower planning and training programs. Similar versions of these levels of change have been described in other writings about RMP -- for example, in "Regional Medical Programs and Health Care" by Robert K. Ausman, M.D. (published by Florida aT). We are aware of similar thought in the California RMP and in the Tri-State RW . III-30 Arthur 1) i-ittle, Inc The problems with ultimate goals expressed at these levels is that by themselves they provide little or no guidance for the establishment or assessment of efforts at systems transformation. There are too many inter- mediate steps and too many poorly understood transitions between, on the one hand, efforts to change the relationships and interactions of key actors in the health care system and, on the other hand, changes in quality and distribution of care, access to care, or health. This is particularly true in large areas; for example, on the level of a region as a whole. To relate systems transformation to ultimate goals some connecting assumptions are needed.@ One such set of assumptions, for example, might be first, that the route to substantive health goals such asithose outlined above, lies in systems transformation through reduced fragmentation and, second, that the key to reducing fragmentation is regionalization in ac- cordance with the center-periphery model. "The DeBakey Commission recommendations envisioned (1) a network of regional centers with highly trained specialists and the most advanced diagnosis and therapeutic facilities for heart disease, cancer and stroke, (2) less elaborate diagnostic and treatment stations which would be in close contact with the regional centers and would refer patients requiring more specialized care to a regional center, (3) the affiliation of community hospitals with existing medical centers to form re- gional complexes, and (4) planning to coordinate services within communities ... health institutions within each region would be integrated around a university medical center."* This concept, if it could work, would have the advantage of making possible uniform criteria for measuring the achievement of systems transformation: the propess would consist of the steps that would lead to regionalization as described above. But, as we have already pointed out, there are inherent ob- stacles to the realization of such a model -- obstacles deriving broadly from resistance to the acceptance of centralization, structure, and coercion in American society, and deriving more specifically from re- sistance among key actors in the health system (hospitals, medical cen- ters, private practitioners, voluntary health associations). Further, the RMP legislation finally adopted in Public Law 89-239 explicitly rejects the nationwide imposition of specific models of regional medical complexes and shifts the language to that of the establishment of "re- gional cooperative arrangements." There are strong arguments to be made for regional cooperative arrangements and against the imposition of a particular regional medical complex. These derive from both the diversity of starting conditions in the regions and from the practical difficulties of effecting systems transformation in actual health care systems (characterized as they are by fragmentation and autonomy) on any other basis than by exploiting, in Bodenheim, op. cit. III-31 Arthur D Little, Inc ad hoc fashion, the particular issues and opportunities that present them- Serves. However that may be, the fact is that RMP's legislation requires just such a diverse, cooperative, open ended approach. The appropriate- ness of such an approach to substantive health goals becomes, then, the fundamental assumption underlying the program. This concept, too, presents a problem. What guidance does it provide? What criteria does it establish for systems transformation? Where it is not possible to establish a single, uniform model of institutional arrangements to be imposed on all regions, it is still possible to identify certain fundamental themes which need to be addressed in one way or another by all regions. Such theme must be dealt @ith, over time, through the development of broad strategies of change in insti- tutional arrangements, specific ends-in-view for delivery systems, ways of confronting particular probl@s that emerge, and techniques of faci- litation. These provide a basis for assessing the relevance and effective- ness of various processes of systems transformation, even though they leave open-ended the specific models of institutional arrangements that might best be adopted in a region. 3. Themes in Systems Transformation Among the fundamental themes of systems transformation are the following: a. Centralization/Decentralization Every region confronts issues over the level and kinds of central- ization to be effected In the medical care systems. To begin with, there is the distinction between administrative and structural de- centralization, the one referring to centralization of authority; the other to centralization of resources. Further,, each region faces the issue of whether to effect regional- ization through regional centralization (center-periphery regionaliza- tion),,subregionalization (nucleation), or the centerless network. Depending on the strategy taken, different questions become relevant from the point of view of assessment of progress toward systems transformation: t For the "centralizing" strategy: o What are the starting points for introduction of the centrali- zing pattern? o How are the key elements of the system to be engaged in imple- menting the centralizing plan? How are their interests to be served? What leverage can be exercised over them? III-32 Arthur D Little, Inc o What are conceived as the steps moving toward the centralizing plan? For the decentralizing strategy: o What are the ends-in-view for the delivery systems to emerge from application of the strategy? o In what ways do they meet the regionalization goals (rati<)nali- zation of resources, linkages)? a Where are the regional "sza-Ds" and how are they to be dealt with? The first starts with a plan, incorporating efforts to make optimal use of central resources through rationalization of planning and thrcx.,gh establishment of linkages; it raises ques- tions about the connection of the plan to real sources of activity and power. The second starts with the actual sources of activity and power, seeking to bind them together in a variety of ways; it raises questions about the adequacy of the delivery system toward which those efforts are directed. b. Regional Identitv The way in which the region defines itself as a regioninfluences both its strategies of systems transformation and the question relevant to those strategies. The region may define itself around geographical political boundaries, as in the case of Iowa, Georgia, Minnesota, Maine, and-others, or around boundaries related to the "catchment areas" of medical institutions, as in the case of Northeast Ohio, Western Pen- nsylvania, Intermountain, Memphis, and the District of Columbia. Identities within state boundaries may take very different forms: New Jersey has for two centuries struggled with the implications of having New York City at one end and Philadelphia at the other. California tries to deal with diversities and distances on a very large scale. In both cases, there is obvious impact of these conditions on the medical scene and on the meaning of regionalization, as already pointed out. The region may define itself partly around what is "left over" after other regional boundaries have been fixed. ("What is left over" is never the whole explanation of a region's identity, but the formation of Tri-State, Nebraska-South Dakota, and some other regions was the results of processes that included at least some of this element.) Where the boundaries are primarily geographical/political, the most urgent questions are these: o How are the considerations determining the boundaries of the region likely to influence its strategies and priorities? III-33 Arthur D Little, Inc o 14hat can the region's stance be toward relationships to insti- tutions beyond its boundaries which nevertheless serve or influence parts of the region (as in the New Jersey-New York, New Jersey- Philadelphia relationships)? o How can theregion respond to resource gaps (or institutional gaps) within the region (no medical school in Maine, few specialists in Arkansas, few or decreasing numbers of physicians in rural Iowa or coastal North Carolina, urban ghetto gaps in most big cities)? 9 Where there are divisions of space and distance between major health care institutions (as in California and North Carolina), what rationalizing strategies are likely to be viable? Where the boundaries are formed around institutions, e What stance would best be taken toward sections of the region which lack major institutions (southern rural New Jersey; the Imperial Valley region in Southern California)? * How does conflict or compartmentalization of major institutions af- fect the regionalizing strategy? Where the region is built ar.otmd "what is left over," the central questions concern w ys in which regional coherence, on any basis, comes to be established. It is, for example, sometimes pointed out that Tri-State RNP represents about the first time that Rhode Island, Massa- chusetts, and New Hampshire have ever tried to do anything together. The problem of regional identity need not present itself only at the beginning of the regionalizing process. It tends to present it- self in new forms as the region develops, leading to changes in available strategies. It is, accordingly, a continuing aspect of systems transformation. c. Inclusiveness Every region confronts, at several levels, the problem of including key elements of the medical care system in its regionalizing strategy. Issues arise around the extent to which the region seeks to be com- prehensive in its inclusion of key actors and agencies, the rate at which they are included over time, and the strategy of inclusion. The price paid for comprehensiveness may be unmanageability; the price paid for manageability may be an ex parte quality that destroyes credibility. The question of "what to include" presents itself minimally at three levels: o What key actors and institutions? Medical centers, community hospitals, private practitioners, public health officials, voluntary health associations, repre- sentative business and labor groups? III-34 Arthur D Little, Inc o What geographic areas? The institution- or resource-poor subregions (the "back county" section of North Carolina as well as the eastern segment of the state)? * What user groups? Representatives of groups :currently cut off from the system, as well as those closely related to it? The activities or "TAoves" in relation to which the issue of inclusion arises are the formation of advisory committees (the RAGs and the committee structures, primarily), the distribution of project funds as ways of drawing in resources, the deployment of programs (as ways of connecting to particular user groups), and the extent and types of contacts made by an RNP core staff in its "facilitative" activity. (See Chapter IV, Facili:@ation.) The choice may be guided by a consideration of which elements in the region are most necessary to systems transformation: 9 Must all parts of the region be taken into account? 0 Must the issues of quality of care and access to care be confronted for all user groups? 9 Must key actors and institutions from all parts of the medical system be taken into account? Given the limited resources of regional medical programs, the questions of inclusiveness quickly lead to questions of priority among elements for inclusion (emphasis on ghetto medicine, for example, as against em- phasis on knitting together elements of the medical care system that .is primarily directed to middle income persons), and to the question of sequence of efforts to include key elements in the light of an emerging strategy of regionalization. RMP's'broad substantive health goals (as outlined above) make no ex- clusions of subregions or of user groups, although proposed HGMHA priorities focus on those now dispossessed or cut off from medical care, and those (such as expectant mothers, very young children) for whom improved access to higher-quality care is judged to be of criti- cal importance. The question of inclusiveness then tends to resolve itself into an evaluation ofthe sequential strategy of inclusion -- a strategy to be assessed in terms of the direction in which it is tending. (Although the decision has been made in the GDV-RMP to begin with em- phasis on Philadelphia, Wilmington, Allentown and Scranton/Wilkes- Barre, how does the coordinator propose to get from there to inclusion of the other major towns, suburbs, and open country in between?) Can the coordinator most effectively work "downhill?" Should he begin with those actors and agencies he thinks he can knot together, in order to position himself to draw in others more resistant to inclusion? Such III-35 Arthur D Little, Inc questions touch closely on the related questions of the "linkages" to be effected and the "conflicts" to be confronted. d. Linkages The notion of linkage is built into the goals of regionalization, but there are a variety of open questions about the elements to be linked, the nature of the linkages to be established, and the strategies of establishing them. Regions share the need to confront these questions,, and differentiate themselves by the kinds of answers they give. 'r The kinds of linkages that are important to center-periphery region- alization are familiar. RMPS statements refer to: o Linking teaching hospitals and medical schools with unaffiliated hospitals, neighborhood health centers, and other community health facilities for the training of young physicians and improving the continuing education of physicians and allied health personnel; o Developing effective relationships between primary care units and specialized backup facilities and services, which include aspects of major community hospitals, teaching hospitals, and the like. In addition to these center-periphery linkages, the following direct linkages are also important to systems transformation at the level of the regions: o @unity hospitals with one another -- for sharing of scarce resources, rationalizing planning, mutual referral; o Private'practitioners and medical administrators with one another for learning about innovations; o Physicians to one another and to paramedical personnel in diagnostic and therapeutic teams outside the hospital framework; o Persons and agencies oriented to disease/therapy with those oriented to preventive medicine and to environmental control; o Health establishment and community organizations; and o Health establishment and those from business, labor, and political groups with power of financial resources. As in the case of inclusiveness, the issues of what kinds of linkages and what particular linkages ought to be sought, in what sequence the effort to establish linkages should be undertaken, and what tactics should be employed cannot be resolved a priori for all regions. Much depends on starting conditions in the region in question, and on the overall strategy of the coordinator as it emerges. Depending on starting conditions and §trategy, however, it is possible to monitor III-36 Arthur D Little, Inc the direction of systems transformation as a basis for raising con- tinuous questions as to whether the rate of movement is satisfactory and the tactics effective. For example, do attempts at sharing services spread either to ad- ditional services or additional institutions? The Upper Kennebec Valley Health Agency, started as an interhospital blood bank, has come to include shared visiting nurse programs and other services, and has been given a chance to operate a nursing program in additional com- munities. Thus, it appears to be moving in the direction of systems integration. The fact that the leadership has also chosen to convert the operation into a "(b)agency" under CHP is also indicative of the direction taken. They are working explicitly on the level of changing institutional planning processes; heretofore, they worked on the level of changing the type and quantity of delivered services. e. Conflicts The question of what elements are to be included and what linkages are to be established leads directly to consideration of the nature of conflicts existing among established institutions and actors, and to the questions of "what," "when, it and "how" concerning the confronting of conflicts. The nature of the conflicts among key elements of the medical care system -- and between elements of that system and of its environment -- influences the strategies of inclusion and linkage. Similarly, these strategies set the bounds and rate within which particular sets of conflicts are to be confronted. Among the major types of conflict are these: Between the "haves" (the large medical centers) and the "have nots." o Among medical schools -- for "turf," or even over the appropriate- ness of dividing turf; for resources, centrality, and prestige (in some regions, competition has been suppressed or dealt with piecemeal; in California, the strategy of subregionalization has overcome, or sublimated, a fair share of medical school competi- tion); o Between "town" and "gown" -- over the relevance of academic centers to community health problems; o Within the medical centers.over the relative importance of teaching,, research, and service, and over service "to whom"; o Among professionals: among private practitioners, between private practitioners and medical centers, and between medical practitioners and paramedical personnel (over role definition and prerogatives); III-37 Arthur D Little, Inc o Between community hospitals and medical centers -- over issues of autonomy, being "gobbled up," dominance; o Among community hospitals -- over command of resources, struggle for emerging pre-eminence, territory; * Between professionals and "outsiders" -- lay people being organized in relation to health -- around "poverty" and social welfare pro- grams, over relevance of the demands of outsiders for access to- service, and over priorities for the use of resources on the part of professionals (as in New Jersey and Denver); o Tension between entities fighting to be appointed as planners or "knitters" (as in the case of RMP and CIIP in some regions, RMP coordinators and state health departments in others, and key actors within RMP in still others.) The analysis of conflicts such as these becomes an important part of the "diagnosis" of regional starting conditions. The nature of the conflicts discovered suggests both where needs for efforts at linkage are critical, where difficulties of linkage are to be expected, what prices are likely to be paid for various strategies of inclusion, what appropriate starting points may be, and what strategies of regionaliza- tion make the most sense from a political point of view. The pattern of pre-existing conflicts among institutions sets the stage for a strategy of subregionalization: Will the pattern of sub- regionalization minimize destructive conflicts and channel energies into regionalization, or will it exacerbate conflict? Subregionalization in California, for example, was intended to minimize destructive conflict both between North and South and among individual medical centers. The device'seems to have worked rather well in that setting. From the point of view of systems transformation, the coordinator must ask himself the following kinds of questions (which are developed .more fully and from a different point of view in Chapter V): o Has he correctly diagnosed the patterns of pre-existing conflidt- in his region? 9 Does his strategy of regionalizing and subregionalizing take these into account? o Does his strategy of systems transformation -- especially as it relates to inclusion of important elements of the medical care system -- take into account the pattern of conflicts and build on it by emphasizing linkages among compatible institutions, or by resolving conflicts through negotiation, or by encouraging joint efforts among contenders on projects in which their interests are not in conflict? III-38 Arthur D Little, Inc. 9 Does he possess, or must he hire or develop, skills in facilitating these linkages and alliances as the need arises? (This will be discussed further in Chapter IV.) 4. Stages in the Process of Systems Transformation The process of systems transformation is one in which some actor or initiator seeks to alter the institutional-organizational-political system of medical care in a region or subregion, or among a particular group of agencies or individuals. The "initiator" may be the RMP coordinator, or he may be the coordinator supplemented by members of hts core staff, or by members of his RAG or review committees, or others influential in the medical care system of the region, with whom he has made common cause. Or there may be, within a single region, several initiators, each of whom is engaged in pursuing his own version of systems transformation collabora- tively with, competitively with, or independently of the RNP coordinator. Within a given RNP region, there will generally be more than one effort going on at one time, some may be attempts at systems transformation and, over considerable periods of time, these may affect one another little, if at-all. (California, with its carefully established subregions, is in many respects an obvious, highly formalized case in point.) Thus, in addition to the RNP region, we have what we might call a "systems transformation region," i.e., the set of elements of the medical care system and the geographic area encompassing them which figure in a particular systems transformation strategy at a particular time. One pattern of regionalization, then, is the process of extension through which several previously independent "systems transformation regions" begintD interact. While there is no single pattern of systems transformation, it is nevertheless useful to think in terms of very general stages of systems transformation common to all @ regions engaged in this processi The stages as described do not correspond to the concrete historical develop- ment of any one region, because they are an abstraction from manv regions; and they do not correspond to the reality encountered at any one time, because a region will generally have several activities reflecting dif- ferent stages going on at a given time. The usefulness of a concept of stages lies in the fact that it allows the same activities to be seen and used in different ways, depending on where they fit in the systems transformation process. Almost any RNP activity (project, feasibility study, planning process, etc.) can contribute to any stage of the process, but the priorities, strategies, and basic objectives appropriate to that activity are different, depending on how it is being used: Primarily to help REP "case" the region and obtain involvement; Primarily as a planning stratagem to help clarify ends-in-view and arrive at more concrete formulations of them, or to check out or discover the feasibility and appropriateness of chosen change processes; III-39 Arthur D Little, Inc. o Primarily to implement a coherent program, with sharply focused objectives (ends-in-view) already established; o Primarily (or at least importantly) to assist in revising the ends-in-view or clarifying new ones. The initiator begins by trying to understand the starting con- ditions, the baseline situation, in which he must function. He seeks to create the conditions for the operation of some strategy of change. In general, this includes the effort to involve in RMP activities those agencies, individuals, and institutions which are regarded as central to the medical care system and to the operation of any strategy of §ystems transformation. Depending on his sense of strategy, at the outset the initiator may seek to get every such element active and connected, or he may work initially only with a subset of the whole, leaving the in- volvement of others for later colmuitment. Then comes the stage of preliminary planning and interaction. Depending on the initiator's strategy and the political realities of the region, this interaction will consist to varying extents of processes of planning, bargaining, negotiation, and more or less open conflict and conflict resolution. The climate of this stage will be governed by the interaction of the interests of the people actually engaged in medical care in the region and by the ends-in-view of the coordinator and other key actors for changes in the system for delivery of health services. In this stage, themes of RMP activity begin to emerge and issues of priority begin to be confronted. Planning activities lead into a stage of implementation, in which projects or activities emerging from the planning stage begin to come to reality and changes in the health care system begin to be affected. The entire process must be regarded as cyclical and interactive. The initiator, once started, will not move sequentially through the three stages, resulting in accomplishment of his systems transformation. Rather he will be continually assessing his starting conditions, seeking to bring new actors and agencies into the fold, trying to guide or facilitate their interaction, observing or influencing implementation of the activities emerging from that interaction, and reassessing the starting conditions under the constraints within which he must now function. 5. Ends-in-View- We have talked about the process of systems transformation, but more needs to be said about its content. The levels of change described under c-2 above, suggest the broad directions in which systems transforma- tion is to move; there are to be changes in institutional and political relationships, changes in the configuration of health care resources, changes in accessibility, and so on. However, these are not explicit enough to suffice as guides for systems transformation in a particular setting. III-40 Arthu r D Li t t le, Inc. goals, e-%pressed at a level broad enough to encompass the national program, or even the program of a region, also tend to be too vague and general to provide much guidance or direction before-the-fact. It does not help to be told that RNP should "plan, develop, experiment with, and demonstrate new or improved systems for organizing and delivering health services that will improve the quality and efficiency of those services" (level 2), or to "generate a wider option among health ser- vice facilities, with heavy emphasis on ambulatory care facilities and services" (level 3), or to improve the health of "expectant-mothers and young children, the poor, residents of core city slums" (level 1), or even to "improve the linkage of -eaching hospitals and medical schools with unaffiliated hospitals, neighborhood health centers,,and other community health facilities" (level 4). It is not that these expressions of goals and priorities are meaningless, but only that they are not specific enough to guide regional operators in the development of a strategy of systems transformation.* Often there tends to be a gap between RNP goals at this level of generality and the particular objectives of clusters of projects which make up the "bread and butter" of so many RMPs (for example, "establishment of 13 coronary care centers in community hospitals,throughout the state"). We have applied the term, ends-in-view, to goals or objectives which are stated in a way that specifies (1) the intention of effecting systems transformation and, (2) the specific nature of the change. At the regional or subregional level,, examples might be as follows: o Bring 16 isolated community hospitals in eastern Maine into a joint planning process, so that they begin to share and exchange scarce resources, such as specialist personnel and equipment-, 0 Ihcrease the number of medical and paramedical personnel providing service in Imperial Valley, California, and establish a referral net which connects Imperial Valley to specialized medical re- sources in San Diego County; o Establish in Newark, New Jersey, an outreach center which con- nects a major teaching hospital to its adjacent ghetto community. Note, however, that they do provide a basis for the critique of specific ends-in@-view once these have been formulated at a regional level. It is possible to say of a region, for example, that it has concentrated on linkage of teaching hospitals and community hospitals to the exclusion of movement toward increased access of "target groups" to the medical care system. In this sense, national RNP goals and priori- ties do provide guidance for the formation of ends-in-view and for their critique. The point here is only that they do not replace formulation of the ends-in-view themselves, but merely offer additional criteria against which to make judgments about the appropriateness of what a specific region has chosen to do. III-41 Arthur D Little, Inc Ends-in-view may take the form of "models" of the delivery of medical care. In particular circumstances any one of several models can be seen as appropriate for accomplishing desired systems changes. There are nearly as many health service delivery models as there are providers capable of seeing themselves as the centers of health service delivery systems. Many of these are familiar through the literature and through demonstration units spread throughout the country. They include: o The Internist's Model: Hospital-based, it sees the sophisticated community hospital or teaching hospital as the center of care, treating all other forms of care as versions of outreach from or referral to the medical center. T 9 The Pediatrician's Model: The pediatrician, through his contact with children and mothers,,..serves as a provider of primary care, an initiator of preventive action, a responder to emergencies, and a screener of service systems. He serves, in turn, as a switching point to all other health facilities and resources which are conceived as specialized backup. o The Neighborhood Health Center Model: The Health Center, linked to it@ surrounding neighborhood by its physical presence, its family health workers, and its paraprofessionals drawn from the community, provides the bulk of primary care, directly serving as a quasi-independent unit for this purpose and drawing on backup re- sources provided by hospitals, specialists, and medical schools. o The Psychiatristfs Model: Often based on the concept of the commun- ity mental health center, the psychiatrist's model envisages the psychiatrist as a practitioner broadly concerned with individual and social well-being for his catchment area and, therefore, as the principal source of referral to outside resources and of liaison with institutions critical to health within the community. Obviously, these are highly schematic and oversimplified and do not begin to suggest the variations within each category. (Psychiatrists, for example, are by no means unified in their views on this subject.)_ Neither do these models exhaust the possibilities. Nurses, social workers, hospi- tal administrators, independent physicians, medical corpsmen, midwives -- all may be seen as appropriate "centers"of health service delivery systems under particular circumstances. Further, there are disease-specific models of health service delivery, such as the various models of preventive, screening, diagnostic, and treatment processes associated with stroke$ and there are more broad- ranging models which attempt to encompass all of the health care functions, and the correlative resources required to carry them out, in a region as broad as an @ . Versions of regionalization models associated with the DeBakey model (see Bodenheim*) are cases in point.' Bodenheim, op. cit. III-42 Arthur [) 1-ittle, Inc. Finally, there are the very specific ends-in-view associated with particular RMP projects, such as: o The establishment of transportation linkages connecting rural populations with medical centers; o The establishment of telephone links connecting medical specialists with independent community physicians; o The development of a cadre of specially trained nurses and their deployment at coronary care units throughout the region. Further comments are in order about ends-in-view as they function within a process of systems transformation: o They are the normative connecting links between broad RMP goals and priorities andthe specific conditions and project clusters of a region; o For a given stage of systems transformation they may be the products of that stage. Early activities , for example, may be judged suc- cessful to the extent that they result in broadly shared ends-in- view for the region or for subregions. o Because of the relative youth of RMP, the process through which ends-in-view came to be generated, the support they have received, the energy and effectiveness with which they begin to be implemented, and their relation to overall RNP goals, become the principal bases for assessment of the progress of systems transformation. 9 Ends-in-view are specific to the particular starting conditions of a region and to the regional strategies of the initiators. While each end-in-view may be criticized in terms of its desirability and feasibility for a particular region at a particular time, there is no reason to expect uniformity of ends-in-view across regional boundaries. o Ends-in-view may be expected to change, and indeed should change, over time. The nature of the changes will be influenced by the increasing scope or leverage of RNP in the region, the desirability of shifting attention from one part of the region to another, changing institutional and political conditions, or new conditions created by the results of earlier efforts to achieve ends-in-view. -6. Health Issues Ends-in-view tend to be formulated with reference to health care issues that are in good currency in the nation or in the particular region or subregion. Ends-in-view incorporate reference to particular target groups, institutions, health problems, aspects of health care, or regional districts. Each ofthese is at some time the subject of special attention, III-43 Arthur D Little, Inc around which controversy swirls and to which energy gravitates, and it is at such times that they become important from the viewpoint of systems transformation. The issues may be expressed in various terms: o They may concern the problems of access of health care of, the rural poor, core city residents, expectant mothers, the aged, or pre-school children. o They may be disease-specific, dealing with tuberculosis, the cate- gorical diseases of RMP, hunger and malnutrition, or venereal disease. a They may be related to broad aspects of medical care, such as the shortage of physicians, the status of paraprofessionals, the rising costs of hospital care, convalescent care, or medical costs generally; or the proliferation of care-providing facilities, and thethreats posed by new profit-making facilities. 9 They may be tied to specific regional entitites, such as the medical isolation of counties removed from centers of population. It is clear even from this brief list that the health care issues of a region may focus on any level of change -- they may deal with speci- fic aspects of peoplets health, with the quality or distribution of care or resources for providing care, with problems of access and availability, or with institutional roles and interactions. While certain of these issues (medical costs, for example) have national status and will encounter only small variations regionally, others are region-specific. They may reflect the changing awareness of needs'or problems within the medical care community, or broader aware- ness of problems within the community at large. They may be stimulated by initiatives already taken to alter health care delivery systems -- for example, the issue of relations between neighborhood health centers and major medical institutions, or the issue of the relation between new classes of paraprofessionals and the traditional health professions. At a given time, each region or subregion will present a profile of health-related issues in good currency -- that is, issues which have become powerful for controversy, action, and the commitment of resources. This profile will be part of the "starting conditions" confronting the initiator of a process of systems transformation in the region. His understanding of them and his ability to find ways of engaging them in his strategy will be signs of his effectiveness. Moreover, the profile of health-related issues changes over time. The resolution of an issue does not necessarily solve everything; it may simply draw attention to other issues, Bringing Wyoming County into the fold may cause its special health problems to come to broader attention and so spread recognition of similar problems across the entire region. III-44 Arthur f) l@ittle, Iiic Creation of a new cadre of health workers designed to respond to physician shortage may give rise to new issues of status and relative reward among health professionals. Successful response to one health hazard (polio being a classic case) may cause attention to shift to other health hazards. These new issues do not by any means engender less energy and concern than the issues they replace. It is, rather, as though there were a fixed quantity of energy and attention for deployment within a regional community, and it seems to distribute itself over the changing set of issues at hand. The pattern of change of health-related issues within a region or subregion is an indication of the direction of systems transformation. 7. Strategies of S@tems Transformation A variety of systems transformation strategies has been sug- gested in the preceding pages (for example, strategies of centralization). At this point, however, we wish to call attention to strategies at the broadest possible level -- that is, to the overall strategies by which the coordinator plans through his program to bring about change in the system of medical care in his region, using the range of techniques available to him. These techniques include, for example: The collection and analysis of data, The development of plans for changes in medical care, o The allocation of RNP project funds, and o The brokerage and entrepreneurial activities of core staff, all of which may serve a variety of general strategies. A coordinator may or may not have explicitly formulated a general strategy, and even if he has, his stated strategy may be more or less in tune with his actions. Moreover, there is no law which limits him to only one strategy. Five of these generalized strategies may be expressed as follows: a. Negotiation This strategy involves bringing about change in the system through negotiation among power centers, which the coordinator must help to generate and mediate. Among the assumptions underlying this strategy are these: o Systems change is a political process, dependent upon power relationships among key actors in and related to the medical care system. III-45 Arthur D Little, Inc 9 It is not possible (or desirable) to impose patterns of change on key actors from some central position. * Change will come about only as key actors find it in their interest (given incentives, constraints, and pressures on them) to change. 9 Such change implies bargaining as the main form of activity (Northlands and GDV). Given this way of looking at things, the coordinator's role includes efforts to identify key actors, to prepare them for negotiation with one another, perhaps even to strengthen weak actors judged to be important, to facilitate the negotiation process, and to support the development of increasing scope for negotiation. b. Unlocking the System Change can be brought about through.unlocking the system. The assumptions here are that: o Key actors and agencies are more or less frozen in relation to one another and to outside groups; that is, they are unable or unwilling to change relationships to one another in signifi- cant ways. 9 The coordinator's role permits him to intervene in such a way as to unlock these relationships, while not obligating him to guide the restructuring of the relationships. The unlocking of the relationships may be achieved by bringing into contact actors previously isolated from one another, by providing incentives or goals to interaction, by brokerage or mediation of interaction of actors and agencies, by drawing attention to new .possibilities for action, or by introducing new actors or strengthening existing ones in such a way as to upset the existing equilibrium, (Maine, North Carolina, Northlands, Tri-State). c. Master Plan Generation Change can be brought about by generating a Master Plan for modi- fication of the medical care system, and engineering conformity to it. Assumptions include these: o It is possible to draw up a Master Plan adequate to the re- quirements of the region for change in quality of,and access to, care. o It is possible to induce key actors in the medical care system to work towardor conform to, such a plan, through rational persuasion, III-46 Arthur 1) Little, Inc. incentives, or compulsion exercised from some central source. The coordinator's role shifts radically over time. He is first engaged in the development of the grand design and in involvement of that design of those he regards as crucial to it. He then becomes a salesman or manipulator on behalf of the plan and finally a guide or director for its implementation (Connecticut to a degree, and Intermountain and Missouri to a lesser degree). d. Imposition of Sanctions Change can be brought about through the imposition of sanctions designed to produce conformity to goals for regionalization -- that is, for rationalization of the allocation and use of resources and for establishment of linkages among agencies and actors. Assumptions include these: 9 Sanctions are required for systems transformation at least as much as incentives to voluntary action. o The imposition of sanctions can be undertaken centrally and in an ad hoc fashion -- that is, on an issue-by-issue or case- by-case basis rather than on the basis of a centrally conceived plan. The coordinator becomes an imposer of sanctions on the behavior of key elements of the medical care system, or an engineer of sanctions which can be so imposed. He concerns himself, for example, with seeing to it that the deployment of new facilities is monitored, reviewed, and (hopefully) controlled; similarly with respect to the development and deployment of medical manpower. (The Memphis Council appears perceptibly to invoke this model, though with aspects of others -- notably (e). But note that the Memphis Council acts for both RNP and CHP. e. Innovation/Sanction Combination Change can be brought about through a combination of (volunta@) innovations in systems for delivering care, coupled with the imposition of sanctions and incentives to enforce rationalization of,planning for and use of medical resources. The assumption here is that both facilitation and "teeth" are required -- the former for the develop- ment of new arrangements of medical resources, and the latter for the allocation and use of the resources. The coordinator plays both the role of facilitator or broker of innovations, and enforcer or arranger of enforcement for rationalized use of resources. Whether any coordinator can be both overt facilitator and overt "enforcer" remains to be seen. While the combination is possible in hierarchically structured organizations behaving permissively, the health care system is not one hierarchy, but many, and the RNP coordinator does not have organizational authority-except with respect to his own staff. We have seen no RNP examples of this strategy being openly used. III-47 Arthur D Little, Inc f. Incentives Changes will come about through the judicious use of positive incen- tives. RMP projects have had their origins, very often, in enabling someone to do what that someone wanted to do. But to the extent that he modifies his original aims, he may be responding to the incentive power of money. This strategy is implicit in all grants programs. The swift growth of RMP and the declining availability of money, of course, have combined to make this strategy one that cannot be used bv itself. This list of strategies is partial, each strategy being only sketchily formulated. Moreover, each raises questions of its own and has implications for judgments about starting conditions and for the conduct of regional programs. The list@carries no requirement that coordinators limit themselves to one strategy at this level of generality, or that strategies may not change over time -- except, of course, insofar as adoption of one may prove incompatible in practice with others. From the point of view of criteria for systems transformation, the following seems to be implied: o The coordinator should be held accountable for making and espousing explicit general strategies, and under controlled circumstances, for acknowledging his implicit strategies for the processes in which he is engaging; o He should be accountable for the ways in which his behavior confirms or denies his explicit strategies; o He should be accountable for answering the questions stemming from the' strategies he employs in his particular region. Chapter IV, Facilitation, is devoted to analyzing and illustra- @ing methods by which these strategies (with the partial exception of lisanctioning") may actually be implemented in the context of RMP. III-48 Arthur D 1-ittle, Inc. IV. FACILITATION Arthur D Little, Inc IV. FACILITATION If regionalization is to be brought about through alP, it must be done to a large extent voluntarily. This implies a slow, often undramatic change process. Some people are more successful than others in bringing about these voluntary changes. The skills necessary to the art of accel- erating voluntary change we call "facilitation." This chapter tries to convey the major elements of this art. It should help a number of core staff members and program coordinators to extend their perspective on what they do, or make better sense out of the activities they engage in. It should help others to improve their skills. Boards, @s, and coordin- ators should be able to spot facilitation when they see it, to identify the need for it when they do not see it, and to evaluate its quality. Those who already can claim-to be skilled "change agents-" need not read the chapter. We expect this chapter to be of little direct help to those regions in which the most effective facilitative work has been done, since they have already experienced much of what we here can only describe and illustrate. We would hope to provide them, however, with some reassurance that t@ere is recognition and appreciation for these facilitative activi- ties, by whatever name they may be called, and that facilitation plays an essential role in RMP viewed from almost any perspective. Of course, if RMP is regarded as itself a broker, a convenor, and a change agent, "facilitation" is central and primary, as well as essential. Since we view RMP as a change agent, we have to discuss faciliation in order to explain the extent to which we believe RMP to be workable. Facilitation can be accomplished in a number of ways. What characterizes all of these ways is the fact that the facilitator has little power beyond that of his own personality. The most obvious form of facilitation is verbal persuasion. Virtually everybody in RMP obviously tries to make others sympathetic to a concept at some time or another, or at least tries to gain consent to a specific project, or to engage in meetings leading toward "regional cooperative agreement." Facilitation on this level is so common and so familiar that concrete examples are not necessary. However, other modes of facilitation, also raised by now to the level of skilled practice, though familiar enough, are not always examined in terms of their facilitative content and impact. Three quite different approaches are: 1. Bargaining or negotiation, 2. Formal planning, and 3. ConfrontatiorL-resolution. Each can be used independently or together with others; each is more appropriate under particular circumstances; each suits the styles of some practitioners better than others; none is sure-fire. Arthur D Little, Inc In this chapter we do not seek to justify facilitation, but to illustrate how changes that cause or enable work to be done across insti- tutional lines are carried out among some of the regional programs which we have studied. In our viewpoint, it is significant that coordinators and other RMP people tend quickly to grasp and readily express their own roles as enablers of change, rather than as "mere administrators." Indeed, coordinators have been among the stronger proponents for recognizing that the regional core staff has a primary "catalyzing" job to perform, in addition to providing such secondary or supporting services as project- liaison, grants management, and stimulating the generation of new grant applications. To be sure, some coordinators have felt uncomfortable at-out the fraction of their total regional budget devoted to core staff activi- ties, but most recognize that a task-oriented regional cooperative agree- ment can neither be born nor indeed mature without sustained work on the part of a considerable number of people. Some of the people who carry out these mediating and facilitating tasks are almost inevitably members of the regional core staff, because they can be justifiably paid for this service, and for them alone the RMP program (and not just individual RMP projects) takes major concern. The work of these people is not all "facilitation," but much of their activity is intended to bring about changes consistent with the objectives of existing or anticipated regional cooperative agreements; i.e., to make these agreements real. In this chapter we describe some of the things we have seen in terms of a theory of facilitation which we believe is consistent with the purposes and practices of the Regional Medical Program. Our introductory remarks may be obvious to some readers; however, within the varied skills, experiences, and intentions included within RMP there are those who have perceived RMP as a facilitating pro- gram from the outset. A. PRINCIPAL CHARACTERISTICS OF FACILITATTON While the processes of facilitation can be described as though it.had a life of its own, in fact, we hold that "facilitation" is just a word, and we use it merely to convey an attitude toward one's job and the ability to apply some specific skills. The keys to facilitation are: 9 Process involvement (a sense that social, economic, political, and psychological processes are at work, and an ability to discern, describe, and relate to them); o A real desire to influence these processes (which implies having made reasonable, accurate, and necessary judgments about possible and desired outcomes of these processes). The facilitator's relation- ship to what is going on is active rather than passive; o Skill in conducting human relationships in ticklish situations not often susceptible to the application of raw power (usually because power is shared among a number of institutions and professions, any of which has the privilege of opting out, or opposing, and none of which has enough power to overcome the others). The skill can be native or acquired; IV-2 Arthur D Little, Inc A willingness at once to declare one's own position and intentions and yet to do so in a way that also positions the facilitator to be open to a variety of outcomes. This concept of facilitation implies a genuine willingness to work with what is available and to invent or call upon techniques appro- priate to the situation in which the facilitator finds himself, in a way that suits his own personal style. It also implies a willingness to be specific, and to be content with'taking small steps. Facilitation is the most likely process by which to succeed when power is fragmented or divided against itself. When organizations are so independent that the decision as to who should call a meeting is considered to be important, facilitative behavior can take care of initial suspicions so that work can begin. Thus facilitation is a way to begin work, to develop a common view of a problem and to arrive at a viable solution, when the real problem is the discovery'and development of a group designed to include the necessary persons appropriately -- a "community of solution" which is a community capable of reaching a solution. If this descriptive theory causes the reader to conclude that "facilitation" means about the same thing as active, constructive partici- pation in a political process, we would feel that our message is getting across reasonably well. Politicians, however, tend to be public figures, identified broadly with particular projected methods and objectives, and they often become known as politicians by allying with some and opposing others. Facilitators are generally not public figures; they are more or less anonymous "go-betweens" -- brokers who (in their chosen milieu) cannot be said to have taken up any specific ideological position, but can enable more genuine communication among all factions, levels, and locations. Of course, the contrast between politician and facilitator should not be overdrawn; in fact, the differences that do exist may be erased completely in the statesman, or the political patriarch. B. MODES OF FACILITATION Verbal persuasion can be facilitative as we admitted ' but we will not discuss it at length. Used alone, it tends to become barren. In most cases of honest difference of magnitude great enough to require facilitation, considerable differences in perspective exist. Mere words do not bridge these @aps,because the speakers send a message so different from what their hearers receive. Something additional is necessary to identify and clarify the messages sent -- and received -- before verbal persuasion can serve facilitatively. We believe we can extend the definition of facilitation, however, by this brief statement of the limitations on verbal persuasion before going into the major modes of facilitative behavior we have observed in RMP. IV-3 Arthur D Little, Inc 1. Bargaining or Negotiation There are several styles of facilitation that are different enough to be easily distinguished: Bargaining 'is one obvious mode, and one which almost every RMP facilitator (committee member, program coordin- ator, or core staff member) uses from time to time. "If hospitals X, Y, and Z can be committed to this project, their involvement will make your grant application a lot more attractive, at least to me" is a typical statement. Of course, there are obvious limitations to facilitation conceived exclusively as bargaining. In reality, RMP does not have much with which to bargain. No core staff member or regional coordinator ever has much free cash or anything else, except sporadic current information for bargaining purposes, and generally he has too small a voice in the grants review process to be able to deliver on any dollar promises he might make if he ventured presumptuously to indulge in dollar-promi-sing. A bargain does not inevitably bring about change; it is not necessarily facilitative at all. If a community hospital agrees to participate in a regional radiation dosimetry-calculating program, the local radiologist thinks he is buying a quicker and more accurate compu- tation of something he would have done in any case. He does not believe he is going to change his method of practice or his relationships with other practitioners or institutions. If he is asked to keep records on the effects of radiation therapy in a form useful to a researcher, he is still not explicitly agreeing to any real modification in his own behavior. He is not even promising to accept the results of research based on his own data. The "price" is likely to be set in terms of the services actually rendered. If changes in referral patterns, consultation, or methods of practice do eventuate, there was nothing in the origi,-,E-.l bargain that made this kind of change a necessary condition. The very fact that it takes two to make a bargain on terms comprehensible and acceptable each to himself does, however, make bar- gaining a very useful facilitative technique. Wherever people feel that changes are not inevitably improvements-- "what's in it for me" (or my institution) -- makes all the difference, and sometimes represents the only reality worth discussion. When these conditions prevail, a bargain is the strongest kind of agreement appropriate or attainable. In these cases, bargaining can become "facilitation of change" if it leads to opening up broader lines of communication or to perceiving new goals important enough to change priorities and relationships. From the view- point of someone trying to facilitate change processes, the worst risk in bargaining is that it will lead only to more bargaining and thus reinforce the very habits of institutional isolation which it is the facilitator's primary intention to open up for scrutiny and possible modification. Another risk has to do with what is communicated between the bargaining parties. Facilitators sometimes find that the party.with whom they have bargained believes that "change" was tacitly excluded from the bargain. IV-4 Arthur D Little, Inc But bargaining is not the only process available to the facilitator. As we discuss other styles of facilitation, the foundations for this judgement should become more obvious for those who need some additional demonstration. We are aware that people who have experienced no facilitative relationship other than bargaining are inclined to believe that nothing could be more effective; in their view, whatever can be accomplished can be accomplished by bargaining. But there are other possibilities. 2. Formal (Conceptual@ Planning Formal planning is frequently intended as facilitation and some- times works that way. The rationale behind using a plan to effect change is familiar to most people: create a blueprint, get people to understand it, mobilize power to implement it, and then do the job. Indeed, Americans assume that anybody embarkdd on a planning effort must want to change something, unless the proposal to plan is a reaction to some other proposal to deal with the problem. Reactive "planning" is perceived to be -- and sometimes is -- just a delaying tactic. Practical people more or less automatically assume that planners are self-interested too. Practical people assume that those who want to plan will invoke change processes which will accrue positively to their own benefit or at least not threaten their own interests. Just now, for example, highway planners are assumed to be in business by a lot of "practical" people because of the support of the highway construction contractors. The analogous example in RMP is the still rather widespread assumption that an RMP plan (and the planners) simply reflect medical school interests, and that the personal careers of the Program Coordinator and the core staff are bound up in the future of the medical school. This assumption quite naturally creates resistance, especially if projected results do lead to cost or-other disadvantage for someone else. In this context, planning tends to be perceived as a process advantageous to people who have (or can hire) lots of technical competence, but who do not have power enough to bring about their desired objectives by a direct onslaught. While formal planning can be facilitative in a climate of trust, it can take on qualities that make it obstructive rather than facilitative in a climate of mistrust. The proponents of formal planning processes see them as a way of building commitment and mobilizing support around progressively more specific approaches to progressively more specific problems. Opponents tend to per- ceive planning primarily as a way of modifying the power structure to the disadvantage of some of its occupants. When this view prevails, the formal planning process can lead directly to choosing up sides. Once the resulting battle has been fought, or even in the course of the battle, change may actually begin to occur. The process can then be perceived as inherently facilitative of change. However, it is likely also to be seen as a victory for some and a loss for others (lizero-sum gamelf) and, accordingly, without net overall gain. The typical FM example is a battle over the planning implica- tions of demographic and health data. The fight results in establishing IV-5 Arthur D Little, Inc a series of sub-regions that are different from those the planners proposed in the first place, the planners feel defeated -- but do have the satisfaction that something happened, a change occurred that would not have come except through some change-facilitating process. The planning process that starts with an ambitious, high- priority attempt to develop a conceptual plan runs the highest risk of suffering from the adverse effects of the process just laid out. In RMP, for example, none of the health planners who has suggested "regiona- lization" on any of the possible center-periphery models has been able to disassociate himself from the suspicion (or hope) that he was trying to accomplish a subordination of those on the periphery of thetscheme to those of the academic medical specialists and central teaching hos- pitals. Achiev-i-ng such a pattern is seen as regionalizing against" 9 or 90 communities to benefit one entity. When the need for such a pattern is asserted (unless developed in great detail and with circum- spection), it says to the local physicians and community hospitals that they are incompetent. It is no wonder that center-periphery plans generate much opposition. This kind of difficulty probably hindered the Connecticut RNP plan, which in September 1968 met with strong medical society resistance even after the Connecticut RMP staff had gone to some lengths to try to inform and involve interested Connecticut health professionals who could be interested. In this case, the center-periphery plan was also vulnerable for not being developed locally. It could be attacked (or dismissed) as a preconceived scheme, because its primary architect, Dr. Henry Clark, had been identified with attempts to develop somewhat similar plans elsewhere. Much of the battle around the plan and what it means has been fought in Connecticut. Passing judgement on it at this distance is dangerous, because the judgement would be backed by too little specific knowledge. Whether the battle was necessary, accordingly, is not the issue under discussion. The point has to do with conceptual planning, a mode of behavior that invites opposition. and creates suspicions on the part of outsiders particularly when used early in a planning process before interests and individuals previously dissociated have come to enough of an accommodation to permit a commonly acceptable conceptual plan to emerge. Conceptual planning serves a real purpose facilitatively; it rarely works well, however, as an exclusive device for bringing people together around change-related issues. Concurrently, fortunately, meetings are held in which many processes formally connected to the conceptual plan but actually separable from it take place. Groups are identified, or identify them- selves, and begin to create their own capability to initiate and agree to (some) changes. Data are generated, shared, and incorporated into increasingly credible diagnoses. Opportunities arise in which to test the self-serving qualities that may be imputed to some -- or all -- of the participants by one another. Another illustration is to be found in North Carolina where Prof. Harvey Smith, early in the development of RMP, initiated health resource data collection and analysis. In organizing the data he posited a series of six sub-regions,each to be developed around one major community IV-6 Arthur D Little, Inc hospital. Dr. Smith maintained that he saw nothing sacred about these sub-regions;facilities, services, and professionals merely represented important portions of the available health resources to be considered in any regionalization process. But many in his audiences were highly critical of his method, his findings, and the adequacy of these findings to justify the "division of turf" he advocated. It was easy for those so minded to band together to unite against his conceptual plan, in the development of which nobody, except Dr. Smith and his staff, had much at stake. Conflict is not necessarily bad and may, instead, be necessary. The trouble is that once a conceptual plan has been proposqd, it is very hard to make people believe it was intended merely as a starting point; and once the battle lines are drawn, it becomes increasingly diffi- cult for either its friends-or its enemies to believe that any outcome will be truly constructive. When a major battle shapes up over a conceptual plan, the plan is no longer viewed merely as a starting point, or a springboard for deeper penetration of the problem. It becomes more like a battle flag around which to rally or against which safely to levy an attack, because the plan itself is there to attack. The attacker can in his own mind avoid questioning the motives of the plan's sponsors, even though.suspicions about these motives are often more important than his objections to the plan itself. Yet no peaceable solution can be found until doubts about those motivations are somehow quieted. The wise facilitator will try to enable the participants to test specific motivations involved in the plan in direct and concrete ways. But unless the planners are skilled facilita- tors in their own right (as well as talented conceptualizers), they may neither deal with, nor even be aware of, what it is that the listeners impute to them, and how excluded and exploited the plan makes these listeners feel. A plan may be broader than the sympathies and active awarenesses of those who formulate it, but it will not often be perceived initially as more than a camouflage for self-interest. A plan always seems adequate (at least as a starting point for discussion) to those who have formulated the concept. They would not otherwise have taken the risk of talking about it nor gone to the trouble of developing it. But it is the limitations, ambiguities, and irritations in applying the con"cept that are likely to be most obvious to everybody else with a stake in the out- come of the plan. Any proposed conceptual plan prepared without prior broad support in the planning process and with shared belief in the utility of its out- comes tends to look like a solution imposed before adequate agreement has been achieved on the nature, seriousness, and tractability of the problems it purports to address. Nevertheless, conceptual planning can be a directly change- oriented and decision-forcing mechanism. In this respect it is different from negotiating processes. Bargainers, as was earlier suggested, need be neither explicit nor conscious about the need for change. Conceptual planners usually want (and are frequently perceived as insisting on) quite specific, very real changes. IV-7 Arthur D Little, Inc There are special perils connected with conceptual planning for anyone tarred with the RNP brush. The report of the President's Commission on Heart, Cancer and Stroke plainly outlined a conceptual plan that related peripheral institutions to "centers." Although PL 89- 239 did not make that model of regionalization mandatory -- indeed in a sense denied it -- the natural association that continues to exist .between the report of the President's Commission and the program that grew up under PL 89-239 has kept many physicians, hospital administrators, and others mindful of the possibility that RMP planners might be trying to institute center-periphery relationships based on the academic medical centers. Every regionalizing concept adds to the resistance that any conceptual plan would normally face, even in regions in which the RAG, Board*, Coordinator, and core staff have consistently regarded center- periphery regionalization as either wrong or unattainable. 3. Collaborative Confrontation-Resolution Critics of health services delivery in this country include a number who are very critical of RNP because it belongs to the medical establishment. Some critics also feel that RNP is inherently unsuitable as an agent-of health delivery system change because it has no clout. Because RMP is a grants program, its main mode of action is presumed by many to be that of inherent power to implement any of the plans it may produce; even its planning projects are seen as weak or meaningless threats. While these attitudes reflect partial truths, they overlook the basic strength of RNP as facilitator: its ability to create and carry out regional cooperative agreements devoted to shared aims that depend on energy drawn from a mixture of contributing institutions, professions, and communities. The basic, direct approach to facilitation is that of explicitly trying to draw people together long enough for them to discover how they can collaborate, and in enough of an atmosphere of mutual support to enable them to dissolve some of their differences and use their remaining differences as productively as possible. Getting these kinds of results is what is wanted and why RMP'anecdotes about bringing people together who "have never sat down in the same room at the same time before" can represent real progress -- even in 1970. This approach is built on the experience that people can (and may choose to) communicate with one another very much more effectively in the presence of a skilled facilitator, whose role it is to press for facing issues, and to insist on adequate recognition of the positions of the people involved. It helps if people are brought together by a neutral party. It minimizes the question, "Why does he want to meet with me?" "Board" here and hereafter is to be taken in the broad sense to include not only the separate boards of directors that are characteristic of the new corporations, but also to include executive or steering committees of the RAG, and similar groupings. The term "board" is being used to distinguish that smaller group that among other responsibilities hires and fires coordinators, provides continuity in decision and policy-making, and along with the coordinator, is the effective decision-making group in an FM . IV-8 Arthur D Little, Inc lo this mode of facilitation -- a process of building relation- ships and confronting issues jointly -- we will devote most of our attention in the remainder of this chapter. Like RMP itself, it is based on the premise that all interested parties have to be kept in genuine communica- tion with one another. It assumes also that the facilitator's commitment is to effect the most significant changes possible rather than any speci- fic prescribed change. (In this mode of facilitation, what "we" come to want tends to become more important than what either "you" or "I wanted initially.) Other functions are important, too, and have facilitative aspects. But the powers implicit in setting agendas and convening meetings (for example) is much better recognized as an instrument of leadership than the items we are choosing to emphasize. Furthermore,,they are ambiguous techniques in the sense that they can be used as readily to close down on a social or political process, as to open it up. The style of leadership and the time of action we are characterizing here is one alwayE concerned with keeping social change processes and political negotiations as open, and as open-ended as is consistent with a focus on a series of defined or definable problems. Let us illustrate. o Georgia's Shift in Objectives: In Georgia, for example, in its initial stages RMP sought to create regionalization around specific, hospital-based projects through which knowledge and techniques requiring a high degree of medical sophis- tication would be made more readily available to practitioners and patients in localities separated from the more academically oriented teaching hospitals. But, as Dr. Gordon Barrow tells the story, in the Judgment of the people living there and involved with the community hospitals, the problems of highest priority in rural Georgia turned out to be simpler questions: how do those in need find medical help? This was not a problem that RMP could solve by itself, but it was an issue RMP could not deny without discrediting itself. The Georgia RMP could not fund projects in the direct delivery of care nor sponsor projects without a categorical focus. RMP could use some core staff time to help develop jointly funded training- recruiting projects. In response, Georgia RMP flexibly shifted some of its earlier objectives -- but in the direction of dealing with what the communities perceived as central issues, rather than evading them. In doing so it became usefully credible to local people whose participation and support was a vital condition of any regionalization process, other than one based on brute force centrally imposed. C. FACILITATIVE POSSIBILITIES INHERENT IN DYNAMIC SHIFTS IN OBJECTIVES The significance of shifts from relatively highly technical medical projects toward simpler things is also worth noting. While a shift in either direction could actually happen, a person who takes the facilitative approach has to assume that some shift will occur as a new project develops to the point where it will be endorsed widely enough to make it acceptable and viable. To some RMP people in Washington, and a scattered few in the Regions, this was obvious from the outset, and by now it is a truism of RMP behavior, though not always recognized in Me's IV-9 Arthur D Little, Inc verbal understanding of itself. Shifts in objectives, it is worth pointing out, represent a rather special characteristic of the grants review process in RMP. The RMP medical school project originally con- ceived to measure physiological variables in patients in CCU's that turns into a continuing education course for nurses, and finally has to cope with physicians too, is a type example of what we mean by "shift in objectives." In a scientific project in medicine, the fact that the.project application has gone through a series of revisions need have little or no significance in terms of ultimate approval. Changes that aremade may heighten the chances of approval, but their primary significance is limited to the relationship between the reviewing committees and the applicant as they all try to cope with issues about scientific quality, relevance, potential payoff, and -3@udgments about professional competence. In RNP, however, the fact of change in a grant application may and usually does have a much more central significance, which is closely related to facilitation. What the shifts in objectives are and how they bear on the relationships among those who will be involved in a project carry great potential significance as an indication of the meaning and reality of "voluntary regional cooperation" for the project under con- sideration. Do the changes in objective achieve something that would make the project more intelligible or more useful to those with whom the applicant would be doing the project? The facilitator, of course, is always open to changes in objec- tive that enhance "voluntary regional cooperation" -- first, to allow the project to happen at all; second, to allow it to happen on a basis that all participants accept as alive and credible; and thus, third, to enable it to proceed on the basis of more genuinely shared objectives, the meeting of which will f@@t genuine needs of all, or most of, the participants. RMP reviewers at all levels of the process should pay explicit attention to these "shifts of objective" and should expect to be informed about their substance and significance. o Maine: Facilitation as a Central Requirement To become relevant is to take account of what you find. Dr. Manu Chattergee, the RMP coordinator in Maine, on first being appointed, saw himself in a program too much isolated from the practicing medical pro- fession. iiaine has no medical school. The RMP in Maine had been set in motion by other institutions, building on the network earlier established by the Bingham Associates Fund and the partly overlapping planning network established through the leadership of Dr. Fisher, the state Commissioner of Health and Welfare. The state medical and osteopathic organizations were certainly significant in the foundation of Maine's RMP -- but not dominant. However, the program coordinator viewed the physicians as the backbone of the medical care delivery system, although, like American phy- siciaris generally, those in Maine traditionally work very independently of IV-10 Arthur D Little, Inc one another, except in a handful of good-size hospitals, most of which are in Portland, Bangor, Lewiston, Waterville, and Augusta. If RNP were to have substantial impact on health care in Maine, the program coordinator felt that he would have to organize a body of MDs to support a program or build around a focus acceptable to them. In his view, the common denominator had.to be improvement of medical practice by upgrading communications and improving physicians' access to more specific and detailed information about diseases and about their patients' symptoms. Almost any physician has to favor the objectives of continuing medical education in principle. Those who would respond in practice could be expected to include a number also open to further working collaboration as it might develop rather naturally in local circumstances. However, constructing a continuing education and information exchange program rele- vant and acceptable to professional practicing physicians is a slow and rather delicate business almost anywhere. Persistent distrust of govern- ment programs is a factor that inhibits RMP, and the complex of forces that make it risky and unpleasant for any physician to have other people looking over his shoulders while he works is an added inhibition. So this approach bore,risks: the doctors might prove to be so unwilling to see practical advantages in working together that a period requited to enlist them in support of RMP would be hopelessly long. But another risk was still more troublesome. Working primarily among prac- ticing physicians would be perceived by those otherwise engaged in health planning and public health activities as being little more than a waste of time and money, leading to the creation of a rather useless duplicate bureaucracy that might simply reflect physicians' interests in the narrowest sense. Driving these "public health planners" into antag- onistic opposition was also a particularly dangerous risk in Maine, because the health planners and public health officials there represent a real and well-respected force, with a leading role in getting the state's RMP started in the first place. -At this writing progress is evident, but it is yet uncertain whether the Maine RNP will be able to bridge the gap constructively between the practicing profession and the public health and health planning interests, but it is clear that these agencies and professions represent the actual and immediately potential power available within the health and medicine scene in Maine. RMP must successfully bridge this gap -- i.e., must successfully work as a mediator and facilitator -- before it can obtain any broadly accepted role, such as leadership toward creating new manpower training programs, of new types, that will depend on the reality of the collaboration already developed. The issue of physician participation looms behind the conflict between "public planners" and "private practitioners," for the extent of physician involvement is still relatively limited. Will the physi- cians and the osteopathic physicians eventually be able to work with and judge one another directly instead of through mutual and rather defen- sive stereotypes? Will the high-technology hospital-oriented physicians and "the others" work beyond the issues that now somewhat divide them? Will the physicians who think that "conservative medical professional IV-11 Arthur D Little, Inc attitudes" constitute the major barrier to improved health care be able to explore those possibilities actively and openly with physicians of contrary opinion, or will they just wait for the "older generation" (whom they respect) to fade away? The meaning of "phy&ician participation" depends on how issues like these are built into the process of recruiting physicians into active involvement with RMP. Skillful facilitation will continue to be appropriate and, indeed, a major strategic tool for Maine's RMP in assisting the health care resources of the state. Resources there are scattered, and somewhat scant. Their collaborative use becomes an obvious method to attempt to improve their effectiveness. lr Part of the basis for judging RMP coordinators should be their skill in facilitation,, how well they use it, and what they choose to use it on. D. INCIDENTAL FACILITATION Confrontation between RMP and the rest of the world is only a secondary aspect of the confrontin- that RMP can helpfully bring about. RMP is more vitally interested in seeing that powerful institutions deal constructively with each other. Suggesting to a local United Fund manager how United Fund dollars could be used to fund a new medical and welfare agency service information clearing house exemplifies one way that RMP coordinators bring about changes. Enabling a state heart association to reallocate its budget and reshape its objectives and programs toward service-delivery projects is another activity in which some RNP people have engaged facilitatively, by allowing heart association staff committees and boards to face and deal with their own "internal" problems. RMP personnel, like any others, have a choice of engaging in either of these activities as ordinary meddlers, or joining a faction and engaging in a partisan political process. Their role in joining a faction would be quite indistinguishable from the role of other partisans. Even though there can be appropriateness in adopting a partisan stance, in the heart association example we have chosen, the price to be paid for avowed partisanship would probably have been high. In the actual incident on which this example is based, knowledge of the existence of other RMP projects and advertised RMP intentions in partly developed form helped turn the trick. RMP people themselves took no position with respect to the state heart association and how it handled its own money. But the fact that others might be taking up projects which until then the heart association had deferred (though admitting them to be important and well within their purview) became an example which the heart association board had to attend to. The political skill of the RMP program coordinator was very highly developed; he knew how and when to allow RMP to be used as an example. RMP could have chosen to reveal a lot less about its own internal processes to the heart association. Facilitation, in this instance, was based partly on the decision to reveal what was going on among various possible applicants for @ money. IV-12 Arthur D Little, Inc In the foregoing example, part of RMP's effectiveness depended in turn on knowing what was going on within the heart association. In itself, this is an important element in building the strategy of the facilitator. Obviously, you can't "keep the pot boiling" without being able to add fuel to the fire. Simply knowing what @-s going on is an essen- tial quality of the successful facilitator. It is a reason why core staffs have to include people from a variety of professional backgrounds, insti- tutions, and institutional affiliations and why the core staff has to be large enough to keep in touch with processes and activities that in them- selves may not involve any RNP funds or any explicit connection to the RMP charter in heart, cancer, and stroke. Effective RMPS, whether they are viewed facilitatively or otherwise, always turn out to be informed. Effective facilitators know how to "keep pots boiling." They are not agents of the status quo; they are not partisan advocates of particular "reforms"; they are people who can intervene without destroying or dominating in a delicate situation, and they can endure the prospect of being ejected from some of the situ- ations in which they attempt to intervene. Good facilitators are satis- fied to live with the lack of acclaim that goes with being only one among several influences that bring about change. Facilitators facilitate; they do not "cause," and they do not have a need to "control." The coordinator in the heart association example also knew how to present facts to people so they could not avoid facing them. This is, or should be, an objective behind every RMP planning study and collection of baseline data. But it takes both a perseverance and great skill (some would call it self-discipline or self-control) to present the relevant facts in a context that allows them to become acceptable and finally effective. If the facts are too well (too expensively) researched, the presenter may drive his audience into boredom or drown them from over- immersion or he may treat his data as so invulnerable that he invites attack by his attitude, which will be perceived as rigid and domineering. If his facts are too casually researched, exposure of their coarseness may leave the presenter in an indefensible position. There is a real advantage in presenting information cheaply won if one wishes to persist as a facilitator. He need have little personal feeling about casually gathered or second-hand information; when the attack comes, he can treat it as an attack on the facts and not on himself. He will be better able to perceive what is really going on, and much better able to continue "facilitating"; for example, allowing his critics or adversaries to stipulate what additional facts are to be gathered and how they are to be interpreted, thus taking a step toward committing these skeptics to some.set of more or less uncomfortable or unexpected facts. In the view of RMP as process-oriented facilitator, these skills are most clearly seen as essential,- but they are almost equally useful whatever the view of RMP. Regionalizer, medical school support program, assaulter on "killer" diseases, or whatever RMP is, many of the processes it invokes are Dolitical. The changes it helps produce are attitudinal and behavioral; the milieu in which it works is interstitial among pro- fessions, institutions, and communities. IV-13 Arthur D Little, Inc All RMP people should by now recognize that facilitation in RNP is imperative, though some still do not. Many more RMP core staff members, RMP staff members, and committee members should be selected and trained to be skilled in the arts of facilitation: Participation politics, Team-building, Issue confrontation,, Group process and "dynamics," Open-faced bargaining, Process-oriented consulting, Organization strategy and strategic planning, in s@l.ort, active and smooth administrative behavior. If these qualities are--not found in the program coordinator, they tend not to be found anywhere within the core staff. If they are not found in the program coordinator of an RNP, that @IP probably will not act very frequently as a facilitator. Committees, boards, and RAGs almost inevitably start out as watchdog operations. This means that few of the important members of these committees and boards see themselves facilitatively or have a basis in their own previous experience for thinking in 'these terms. Quite the contrary, they can usually look on one another only as representatives of professions, institutions, or communities appointed to protect their own interests. If a program does not already have a program coordinator who is properly skilled in these arts, it may have real difficulty in acquiring one. First, the members of the board may not even recognize what they are looking for if they have not identified faciliation as of key importance to the success of the job. Second,, such a person can often spot the difficulty or impossibility of working for a board in which few, if any, of'the members have much skill or interest in themselves acting positively as catalytic leaders or facilitators. Those boards which have been able to hire program coordinators who are strong in some of the skills under discussion have done so: (1) through the membership .on the board of some people who, at least, intuitively or explicitly recognize these factors and skills, (2) through offering the prospective program coordinator contingent guarantees -- medical school or other appointments that could strengthen his hand and underwrite his independence, or (3) sometimes by the accident of a person's availability coincident with an interest in participating actively in change. E. FACILITATION AND EQ STRUCTURE IN THE REGIONS Earlier mention of RMP structures -- the RAG, core staff, committees, the board -- alludes to facilitative uses of these structures. This section describes how changes in these structures can become facili- tative in themselves. The very fact of rather fluid, relatively easily revisable structures with relatively open membership characteristics, with legal requirements to include members from a variety of sources, IV-14 Arthur D 1-ittle, Inc is a facilitative tool well worth using. Furthermore, it is a tool for facilitation that a number of program coordinators do not recognize explicitly in this way. The prevailing, observable mood in which program coordinators approach regional reorganization of an RMP committee structure or a RAG or core staff is all too generally negative. Because things have not been happening fast enough or well enough, some other way of organizing is "suggested." The same problem can be looked at a lot more positively: what has been learned by the failures of a pre-existing organizational structure identifies or illuminates what needs to be facilitated. Probably what has been learned is at the root of the kind of reorganization that the program coordinator is asked to contemplate in any case, whether he is thinking explicitly in terms of facilitation or not. In the next several subsections we take up specific aspects of this question in more detail, making organization and reorganization facilitative. 1. Board Vis-a-Vis Program Coordinator and the RAG One aspect of facilitation and its organizational implications (already briefly mentioned) is the process of hiring a program coordinator. One issue in his hiring is the extent to which he is captive -- or viewed as captive -- of any of the institutions or professional forces that are under suspicion of wanting to take over RMP. His behavior on the job when he reports will, of course, be scrutinized as a continuing test of whether he is -- and can be -- his own man. This situation makes the question of contingent employment guarantees-(earlier mentioned) potentially a very touchy one. So far it has been best dealt with quite openly. In one region, for example, several medical schools recognized that identification with any of them would be the "kiss of death" for the program coordinator but, by the same token, that failure to achieve clear channels of communication between the most prestigious of these medical schools and the program coordinator would be equally fatal. The solution they worked out was to locate a health administrator already well-known to the health and medical establishment in the area, whose previous career was, in only the most incidental ways, associated with the interests of any of the medical schools involved. The coordinator-to-be was explicitly recognized as a figure in his own right. Appointment to the most prestigious of the medical schools involved would be considered almost as much an honor for the medical school as for the coordinator and would not imply "capture" of the new coordinator by that medical school. Almost automatically it received the informal, tacit consent of the other medical schools in the region. A delicate balance and a very nice (if slight) degree of interdependence among the sponsoring medical schools was brought about, and a degree of independence for the coordinator was achieved. In this instance,, some of the most crucially involved people, including the IV-15 Arthur D Little, Inc coordinator-appointee, insisted on addressing the issues of independence, captivity, and channels of communication explicitly and directly. The very process of hiring a coordinator became itself one that facilitated relationships among a number of health institutions and constituencies, starting with the medical schools. It did not solve any of the substantive problems in a very ticklish situation, but it was as clear a7,.a indication of a willingness to find ways to solve problems of substance as the situ- ation permitted. In the two years in which Arthur D. Little, Inc., and The Organization for Social and Technical Innovation have been observing the Regional @ledical Program, the identity of the coordinator ano his relationships with the board that hired him has led again and again to the same conclusion: a board obtains the coordinator it deserves. Board-coordinator relationships should accordingly be regarded as a progressive, boot-strapping proc@ss, the object being to develop the attitudes and functioning of the board so that it can strengthen the hand of the coordinator as a facilitator (as well as in any other ways that are judged specifically appropriate within a particular region). The board, in the process, will be brought to such a point that when the time comes to choose a new program coordinator, the board's capability of identifying and recruiting a top-quality candidate will have been enhanced. The national program director and his most senior associates are in a position to support program coordinators in improving the qualities of their own relationships with their boards and in many cases to influence the makeup of the board. Conversely, they are also in a good position to prod boards toward hiring the best available coordinators. Nor should the National Advisory Council and the Review Committee be overlooked, especially in site visits. 2. Staff-Committee Setups In a number of regions, there have been at least one or two major reorganizations of committees and task forces. In many of these reorganizations there seems to have been an implied hope that "this time is the last time... hopefully, we have achieved some finality in the structure of these committees." From the facilitative point of view, of course, there is no such thing as finality, although such a thorough- going reorganization as was, for example, accomplished a year and a half ago in Northlands may tend to defer the time when the next one is indi- cated. In that instance, a complete set of categorical committees and subject-area committees was displaced by so-called "functional" committees on health manpower, continuing education, and health care delivery. Such a restructuring suggests a very basic shift in the internal understanding of what the program is about, what RNP is attempting to accomplish, and toward a more active grappling with the real problems in health. IV-16 Arthur D Little, Inc Another historically common type of reorganization reflects the shift from the period when the work of RMP is to get RMP set up to a phase in which the issue is to use the set-up to do some work. So long as regional voluntary cooperative agreements were measured prin- cipally by names, professions, and locations listed on RAG, committee, task force, and project application mastheads, committee members were often passively chosen because they lived in particular places and had desired professional or institutional affiliations. When the job becomes recognized to be one of clarifying specific issues, developing credible agendas for the RAG, and addressing problems involving all the power forces that could materially affect the future of health care delivery within the region, committee and task force recruiting is looked at in a different and more active light. However, there are examples still remaining in 1970 where RMP core staff members are charged with identi- fying and recruiting for th@ committees needed to carry out sub-regionali- zation using criteria based mostly on place of residence and professional affiliation. Questions of power structure, ability to make committees function, and network building seem to be given only secondary attention. Those committees that a parently feel good about their roles .p and are respected for their contributions to the RMP in the regions share one obvious trait.in common: most of them received staff support to prepare agendas and schedule meetings, draft output documents, and help resolve differences of opinion among committee members. If no other facilitative technique was available or allowed, at least the staff pulled together available information and sketched out available alternatives. The committees set up officially by RMP and not provided with staff assistance are very likely to conclude that RMP really does not want them to function. Committee members in such situations often say they know -that money is in short supply and, of course, staff members already must have too much to do; but they ask how RQ expects them to do any significant work, and where are their resources to come from. Since facilitation is partly a way of getting other people to do work beyond your own capabilities, allotments of staff support for committees should be standard in RMP budgeting. The absence of this item should elicit questions both at the RAG and RMPS levels. 3. The Unique Role of the Regional Advisory Grou2 The common practice of having both boards and RAGs makes the RAGs subject to unique stresses. RAG members with whom we have spoken fall into two categories: either they are little involved and care, know, or will say nothing of the stresses on the RAG; or they are articulate and more or less wrapped up in RMP "politics." Their statements normally amount to the general assertion; "Somebody else has the power; the RAG doesn't." This is not a statement that should be taken at face value. No doubt in specific cases RAGs have been without power, but our experience suggests that this never lasts for long. IV-17 Arthur D Li ttle- Inc But other factors also lie behind this assertion. For instance: * Most regional advisory groups are large bodies of men and women who have little chance to get to know one another and to learn how to work together, unless they make very great efforts over very long periods of time and unless they meet more frequently than is common for RAGS, which is from two to four times per year. The typical RAG, of necessity, acts cumbersomely and inefficiently, and is in constant danger of losing momentum between meetings. * Under these circumstances, the set-up is ripe for the appearance and even the reality of manipulation. The review process that structures project application submissions through a series of committee approvals can'leave the RAG in a position in which its members have little choice but to approve whatever survives earlier review steps. Wheii this happens, the review process tends to look positively underhanded to RAG members, even though the serene complexity of the review process may have been concocted more out of frustration and fear of risk-taking than out of any lack of principle. * The dependence the RAG chairman almost surely has on the program coordinator and his core staff to prepare agendas and make materials ready for presentation further restricts the possibilities that RAG members have to make any initiative contributions. Indeed, virtually all RAG appointments, meetings, and actions are subject to the fear by its members that the Regional Advisory Group is little more than a rubber stamp. Furthermore, these fears are inten- sified by the hurt feelings of people whose applications have been turned down somewhere 'along the way. The unsuccessful applicants are bound to consider the possibility that their 'Long-sought projects were turned down because the RAG was dominated by cleverly invisible manipulators. In this view, the '@Establishment" always frustrates the attempts of "outsiders" to effect changes in the established order and to redress the balance of power long concentrated in a centralized clique. Both the coordinator and board (where the board is actually quite separate from the RAG), in recognizing and accepting these strongly nega- tive feelings, have a great opportunity to prove that RNP is, in fact, open to legitimate influences, that it does, in truth, offer a valuable sounding board for trying out new ideas for solving old problems, and that it does provide a safe forum in which to develop acceptable formula- tions and potentially actionable suggestions for handling old grievances. This is brought about by a combination of constructive listening and open- faced "leveling" with the disaffected persons. It takes a lot of patience on the part of the coordinator to listen long enough to people (notably present and prospective RAG members) with grievances to understand what the grievance is really about, beyond the level of sheer hurt feelings and a desire for self-aggrandizement. It takes a lot of skill on the part of core staff members to assist people IV-18 Arthur D Little, Inc with ideas to get them expressed in relatively nonthreatening forms. But what it mostly requires on the part of the program coordinator is looking at the RAG and learning to relate to it as a source of strength, rather than as a necessary evil. The RAG becomes a source of strength to the extent that RMP really is trying to behave facilitatively. As more and more RAG members discover that the coordinator and the board look on the RAG as a body whose active involvement is necessary to putting important issues in perspective and in actionable form, the RAG can become more and more an active supporter of RMP's facilitative role. It can then freely involve itself in some of the really tough issues of the medical care system: how money spent for medical care (and health care, more brToadly) is divided, how services are distributed, and how the mixture of services matches what people need. 'Every RAG,whose members have been interviewed in the course of this study, appEars to share most of the relevant facts and many of the opinions that are in general currency in American society today for airing and dealing with these health questions. Several of these bodies are definitely edging in the direction of confronting just these kinds of issues in terms appropriate to the specific region. Each is doing so within the constraints of Public Law 89-239 as interpreted locally, which produces understandably uneven progress in that direction among the 55 regions. We have just stated a profoundly sympathetic endorsement of RMP and the RAGS. If RMP can bring about the changes suggested here in attitudes, knowledge, and understanding, its justification as a facilitator will be simple. Our data come admittedly from the time before the filing of the current series of health insurance bills. Possibly facilitation is more difficult now if there has been a hardening in the position of organized medicine, in anticipation of the feder@il govern- ment's taking a hardened line. Our most convincing example is that of the New Jersey RAG. Over a period of twelve to eighteen months, in part while we watched, the mutual perceptions of physicians and social activists on the RAG changed enough to permit a near unanimous endorsement of New Jersey's RMP focus on urban ghetto health problems. The open exploration of the difficult health issues, when conducted with imagination, sympathy, and mutual respect, defines what needs to be facilitated and creates an atmosphere in which facilitation can be effectively employed. In both these respects, the Regional Advisory Groups, with their varied and scattered membership and including extended task force and committee involvement, can indicate to the coordinator when progress is possible and possibly what can be done. This view of a facilitative function of the Regional Advisory Groups will be found too hopeful by those who proclaim the intractability of self-interest and deny that powerful conflicting interests can be made to budge without very strong threats or the actual imposition of direct force, i.e,: "Threatened people respond only to threats." IV-19 Arthur D Little, Inc Almost all that RNP has to work with is the power of friendly persuasion. Even were RMP to grow to a national program spending $500 million per year, its relative size and actual direct power would still be very small compared to the $70 billion health system it is trying to influence. RMP can either learn to be significantly and steadily effective in the facilitative mode, or it must reconcile itself to a steady diminu- tion in its effectiveness as the health care systemgrows. Learning to. really use the RAG is not the only aspect of developing facilitative skill, but it is crucial and central to the whole task that RMP core staffs have in all the regions and the national staff has in support of the regions. 4. The Coordinator Based on the foregoing it is evident that the coordinator is a major fulcrum around whom processes of facilitation in a region can move. If there is to be motion as a result of the facilitative processes of the board, the RAG, the committees, and task forces, then the coordin- ator must use his core staff and administrative skills to see that there is followthrough on commitments. In the long run this will cloak him with an aura of authoritarianism that is incompatible with his role as a neutral and facilitator. He may also grow accustomed to the people to whom he must relate and, in the process, grow less sensitive to the opportunities of affecting their attitudes and behavior. The question of evaluating how well a coordinator does his job is raised. Consistent with the notion that his ability to facili- tate is crucial, there are at least two considerations applicable to most of the 20 regions from which our data are drawn: (1) in facilitation, as in most other aspects of management and administration, the ultimate power that an incumbent has is suicidal; i.e., a time -- when to resign is to facilitate is always a possibility; and (2) People sooner or later get stale on the job; they run out of new ideas; they no longer see any of the attainable changes as being improvements. Let us develop these administrative truisms into something more specific to RNP. In his way, the RMP coordinator fulfills many of the kinds of representative functions that a Presidential appointee in the upper levels of the national civil service fulfills. He gets intimately tied to specific sources of political power and professional interest; at the same time, he is generally accountable and must be responsive to everything that exists around him. In the same way, then, as the super-grade or cabinet appointee, the RMP coordinator may ultimately arrive at a situation in which principal progressive follow-on actions for RNP can become possible only if he departs. This involves lots more than the issue of personal competence. Expressed in the facilitative IV-20 Arthur D Little, Inc mode, constructive resignations must avoid heroic battles over small policy differences. To be really worth the melodrama, risks, and personal inconveniences involved in quitting the job, there have to be significant issues of role, thrust of program, and underlying philosophy that have come by the coordinator and major elements of the RMP constituency to be recognized as real, and not mere defensive justifications of narrow views. The other aspect to be examined is staleness. When is a coordinator stale? If the facilitative role is taken as central, then isolation within the RMP staff offices is an important sign. The coordinator who facilitates does not do it alone, and he does not do it through conversations with his staff. If he remains or becomes almost exclusively a staff director after the initial organizational phase, his awareness of activiti@ in the region is almost certain to be secondhand, and his region's limited knowledge of him is almost certain to result in an adverse opinion of him. A coordinator's relationship with his board is also a clue to an overrun term of office. A stale coordinator is one who has become the living embodiment of his and the board's shared ideology: he survives because he can be trusted to do or say nothing that is outside the limits of that ideology, not because he is too valuable to sacrifice. Alternatively, the stale coordinator is one who seems to have no particular views on anything except the importance of keeping his job -- which he does by adhering simply to the largest manageable power bloc within the board. Still another indicator of coordinator vigor or staleness is illustrated by what role the RAG coordinator attempts to build for his RAG. He who discovers more and more ways of involving the RAG members and its committees may be riding for a fall, but he is certainly not stale. RAG behavior is an important indicator of program coordinator quality as a facilitator. Size, diversity, and ease of entry into most of the Regional Advisory Groups makes the RAG a more potent vehicle for keeping PMP abreast of, or in advance of, its own medical world. The program coordin- ator can be the spearhead in facilitating significant processes in his region if, and only if, he is connected with emergent forces and issues. But these are complex and dynamic enough that it can be done only if the coordinator keeps RMP closely, continuously, and vitally involved with the issues. 5. The Facilitative Role of the Project We discussed shifts in project objectives as an important aspect of facilitation. These shifts have to take place in a specific setting, often that of an RMP project, or a grant application. Along with the direct day-to-day work of core staff members, RMP projects can be the basic tools of facilitation. Projects can themselves IV-21 Arthur D Little, Inc constitute RNP information networks. They can be used to create oppor- tunities for persons located in critically diverse positions in the health care system really to work together. They can become vehicles of atti- tudinal change, institutional change, or of changes in relationships among institutions -- and all within the setting of health care improve- ment. F. THE NATIONAL STAFF (REGIONAL MEDICAL PROGRAMS SERVICE), NATIONAL INFLUENCE, AND FACILITATION In a sense, RNP started as a medical school program. Signifi- cant numbers of coordinators and a fair sprinkling of powerful members of the national staff have been closely identified with American medical schools. In fact, RMP's first administrative assignment was, after all, to NIH. Given these circumstances and the fact that under the law almost all of the regions were most readily formed around medical schools and many universities were used as fiscal agents for RMP, the possibility of exercising national control over the program through the medical schools must have seemed very real. E'ven in the absence of any visible behavior to support this theory, it will be years before practicing physicians have completely given it up. Some members of other health professions and many consumers can be expected to hold to this view for a long time, as well. In the meantime, either the suspicion or the actuality serves to inhibit the trust that can be accorded RMP locally. The specter of the Federal Government dominating local medical practice through some kind of unholy alliance with academic medicine may seem ludicrous to medical school people and civil servants in 1970, given the relative lack of response the Federal Government is according the medical schools in their current financial plight. But this new development has had little impact as yet on people who are convinced that it is easy for the Washington staff to put pressure on regions through messages sent directly to medical school deans sitting on local RMP boards. In fact,.these people see the Washington staff as wanting to do just that. In those regions in which town/gown relationships continue to be strained, RMP can be effective neither facilitatively nor in any other way in the building of regional cooperative agreement so long as this issue is not dealt with frontally. In such regions, RMP is likely to be constrained to minimal action in the name of continuing education because it can be trusted to do nothing else. In any region in which medical school domination" remains as a suppressed or active issue, steps should and can be taken by the coordinator and the national staff to make sure that structural features of the local RMP positively negate the likelihood that the deans might be in a position to act as invisible agents of the Federal presence. These deans should not be in a position to "control" any Regional Medical Program, nor should they appear to be in a position to do so. IV-22 Arthur D Little, Inc Statements of the issue of medical school domination heard in the regions are quite complex and vary from direct and candid accusations to less direct complaints about the irrelevance of academic research. Complaints about "Federal control" often mask greater fears of "medical school domination" or vice versa. If an RMP is to be a genuinely facili- tative program whose grant mechanisms and local activities are to be trusted by people outside the medical schools, then both RMPS and responsible people in the region should press to erase the vestiges of "medical school domination" by discovering and developing a more appro- priate relationship between medical schools and the local RMP. An effective RMP is a glue binding all significant forces, amopg which must be numbered "academe." We are by no means preaching exclusion of the medical schools from RMP. What we advocate is genuine broadening of management that includes the medical schools and overcomes the problem of RMP captivity by sharing rather than merely replacing one kind of captivity by another. 1. RMPS Guidelines to the National Review Process Both guidelines and review processes need to be developed further if they are to serve truly facilitative ends. These mechanisms need not be particularly constraining and do not seem to be so regarded in practice, except in regions where the prevailing attitude is one of passivity: "You tell us what you want and we'll try to give it to you." The worst that can be said about the national staff in dealing with this attitude is that very few know how to say, "Your passive way of looking at your job makes a bad impression on me. After listening to you, I'm torn between telling you what I think you ought to do and telling you that the more truthful you become in stating what your region really needs, the more likely you are to give us what we really want. What I myself want (or might prescribe) for your region could be useful advice to you, but would just increase your passivity. Besides, giving prescrip- tive advice puts me in an impossible situation. As you are in the region, so am I at the national level -- only one agent in a complex process that's controlled by no one agent. I simply cannot deliver on the implied promise to get something for you if you do what I say. Telling you to do something about the real health care needs of your region won't accomplish a thing with you in your present mood; you will merely interpret that as Federal double-talk, or as an indication of Federal incompetence. The best I can really say is, 'You've asked the wrong question, so let's start over'." Members of the national staff with whom we have discussed these matters say they are not credible when they try to communicate these things. It is possible they would be more comprehensible and ultimately more believable if they articulated the strategies and IV-23 Arthur D Little, Inc processes through which people in the regions could go in order to identify, document, and present the region's story and to make the region's case for funds. This set of problems will probably gradually disappear. It certainly seems to have diminished some in the 1968 to 1970 period. Pro- gressive additional clarifications and broader agreements on the possi- bilities inherent in RMP will still further erode it. Agreement does not even have to center on the issue of facilitation; but to the extent that RMP should focus on facilitation, the appropriateness of national influ- ence being applied through guidelines and through help in devising and expressing appropriate local priorities and strategies will become further enhanced. When sufficient progress has been made in this direction, the generic advice of RMPS can be supported by actions that go beyond these words (a paraphrase of the guidelines): "It's less what you do than how you go about arriving at an auth@tic statement of regional needs, and flow the region intends to cope with them." The national staff, of course, is just as weak in facilitative skills and experience as is the typical region. Some well-staffed regions have more and better facilitative skills available than the national staff as a whole. This is not surprising. RMP was more rarely perceived in its NIH days as a primary facilitative program than it has come to be since 1968. Since those days, however, policy changes have been frequent, personnel changes have occurred with considerable regularity, and the external relationships of RMPS to health care planning agencies in the Federal system have also changed in significant ways. Furthermore, the central issues before the nation were not the same nor perceived from the same viewpoint as they were five, or even two, years ago. In the light of these changes, it is unfair to fault the national staff for being -slow to come to the point of general agreement that facili- tation is the most important activity of RMP. This is especially true because the selection of staff in the first place put little emphasis on the values, skills, or techniques of facilitation. Furthermore, RMPS is ,no less riddled than other government agencies with uncertainty about the credibility of its own behavior to those whom it needs to convince. it is difficult in this climate to expect the RMPS staff to adopt a facilitative interpretation of its role wholeheartedly. 2. Facilitation and Ideology An ideology may be either largely explicit or largely implicit in the minds and language of those who espouse it. Explicit or not, an ideology is almost sure to be perceived by those of different opinions as reflecting a self-serving theoretical formulation, which protects the interests of its proponents and places his opponents either in the wrong or in limbo. The detection of somebody else's ideology, accordingly, is usually a process in which one finds himself believing that he is uncovering rather narrow and selfish motives on the part of the proponent. IV-24 Arthur D Little, Inc This formulation makes it emotionally difficult for the observer to test the reality or relevance of the motives imputed to the proponent. It is usual, then, for ideology to get in the way of facilitation. As anyone will discover upon entering the RMP circuit, there are many medical and health care.ideologies and they have the effect of making their proponents appear small-minded to each other. From the point of view of the medical practitioner in community practice, relatively isolated from teaching hospitals and medical schools, the national medical strategy since the Flexner Report can be, and often is, seen as personally motivated ideology. That is, the belief that medical care might be significantly improved by uncoveringrprogressively better scientific foundations for medicine, and that, therefore, the teaching of doctors to be better scientists is the one "right" approach to better health care can appear to practitioners as merely a guise for making medical schools fat and happy. Worse than that, the continued support of the strategy in terms of a gap between attainable and realized qualities of health care becomes a criticism of all of the medical establish- ment outside the immediate purview of the teaching hospital and medical school. From the point of view of the dedicated professor of medicine (perhaps even more clearly so from the point of view of his chief resi- dents), the account that the local medical doctor gives of himself seems likewise to be so thin and unconvincing as to deserve the epithet - "Ideology." His claim to be conservatively avoiding interference with phenomena nobody understands very well, and his ready confession that some fraction of his patients get better, in spite of what he does, seems to the academic devotee to be partly a mask for the local doctor's incom- petence -- an excuse for his negligence in not using available science to try to find out what is wrong.-This is then extended to the imputed motive that the local doctor acquires an excessive fraction of the medical care dollar by delaying his consultations on the specious grounds that nobody else knows better than he what can be done for his patient. The community hospital view of health care can be stated as: "The hospital is the community center for health care, with convenient physicians' offices and ambulatory care available to all". People of a different health care ideology interpret this in a much more self- interested formulation: "Keep the doctors happy by giving them access to the gadgetry they want so they will admit lots of patients, so there will be enough more cash generated to pay for (and justify) higher salaries for the senior administrative staff." The Public Health Service view as perceived by most physicians in the private practice of medicine also turns into an ideology, justi- fying the "weakness" of those physicians who have chosen public service because they cannot stand the rough and tumble, blood, and long hours of private practice. In their overview, Public Health physicians have an ideology that may reduce private physicians to being money-grubbing repairmen for disorders that should largely have been prevented in the first place. IV-25 Arthur D Little- Inc There is no further need to elaborate. It must be obvious that an ideology is what somebody else believes which, from its opponents' point of view, is poorly founded in fact but deeply wedded to self-interest. Ideological views do exist and are real to those who hold them; to others they are more often seen as unrealistic, or distorted for rea- sons that are suspect. Every facilitator must appreciate this situation not only because it adds to the complexity of his task, but because he needs tools and techniques for dealing with it. Our object is to remark on what the regions and the RMPS can do toward minimizing the deleterious effect of ideologies. "Ideology and Facilitation in a Rural Health Care Project"* recites examples of several professionally centered "ideologies" and how they interacted in a particular setting. That report illustrates several of the points already made in this section, and places them in a more concrete setting. In discussing facilitation and ideology, the fact that facili- tators may have ideological hangups of their own, of which they are only partly conscious, should be mentioned. These are more likely to become important when a facilitative process is well underway than earlier in the process. When a task-oriented group has begun to reach conclu- sions and to make decisions, the clever facilitator will try to under- stand his own involvement well enough to be able to avoid the decisions that may take a quite surprising form. He must recognize quickly when his own standards of judgment are'at least partly ideologically based. There are many examples of this., In the experience of ADL and OSTI in interviewing peo;le in various RMPS, one of the most common p concerns is the relationship of RMP staffs with state and county medical society representatives. With distressing regularity these representatives have been perceived as absolute mastodons at the outset of their relation- ship with RMP. -Core staff members, and others as well, sometimes con- scientiously dedicate themselves to relating to Medical Society repre- sentatives in ways that support the preconceived "mastodon" image. Long .after it is clear that mastodons no longer exist, the very real contri- bution of these physician representatives of the medical society is limited by this constraining preconception. One final set of remarks that reveals the paradox and the agony of facilitation should be made. The facilitator who is going to be effective must act like an entrepreneur, a person able and willing to push ahead in the endless venture of trying to put together things in unaccustomed ways to demonstrate their superiority. However, the facilitator is not an entrepreneur; he does not "own" the situation. So, he must attempt to minimize his own ego-identification with any specific problem and its solution. His tasks are to identify where the Addendum 1 to this chapter, page IV@28. IV-26 Arthur D Little, Inc needed energy and capability for doing new kinds of work are located, to help other people build clear commitment to new courses of action, and to let the people on the scene do the job. Making RNP truly into such a facilitator,of course, is not without costs. It will require systematic training programs for national and core staff persons now on board. Changes in recruiting criteria will have to be adopted. Assiduous attention to explaining the role of facili- tator is required at all levels. Strategizing to weld the process of facilitation and programatic content is vital, too, if facilitation is to have any point. Assuming these efforts to make facilitation credible and real, facilitation can, indeed, become the central RMP%activity. IV-27 Arthur D Little, Inc ADDENDUM #1 IDEOLOGY AND FACILITATION IN A RURAL HEALTH CARE PROJECT This disguised case study is an attempt to illustrate the arduous and confusing path of the facilitator in an ideologically and politically complicated setting. What makes things complicated is that there is rarely only one ideology pitted against some other one ideology. Real-life situations in health care seem to have several of these "ideologies" implicit or explicit in them at any given time. The setting is in a relatively poor, largely rural part of the country which spread into tlW6 states. The area had been publicly labeled as impoverished and badly served, thus making the local health care professionals feel a bit guilty. Furthermore, it had been studied to death for years, thus angering the local people and making the members of the University medical faculties near the area feel guilty, also. A number of dedicated health professionals from two states brought together by.the promise of money from HEW were struggling to organize themselves to cope with the deficiency and maldistribution of health care resources. This case study covers six months of the life of the RMP project. The story begins with a new ad hoc committee (the Health Care Committee) composed of strangers, who at the outset knew one another only by reputation, if at all. The Committee was partly self-selected through the process of volunteering. However, its members were seen by one another and by outsiders as representatives of the institutions and places from which they came. Furthermore, they were "appointed" by state political authority, although some of the members were unaware (or did not believe) that this appointment had any significance. They were much more sensitive to a larger, also ad hoc group, from whose number they had been selected than they were to the state agencies. The larger group included state and local health officials, representatives of RMP and CHP, medical school deans, health and welfare agency repre- sentatives, Federal and regional economic development officials (including the staff of an Appalachian-type Regional Commission), leading private physicians in the area, and a few hospital administrators. The smaller committee itself was composed of: One pathologist who operated laboratories on contract in a fluctuating but always substantial number of community hospitals in the area; Two general practitioners, one from the largest ethnic minority in the largest town in the area, a man who tended to lead the medical staff of the hospital in which he practiced; the other, a good physician, well IV-28 Arthur D Little, Inc known in the state medical society, who practiced in a small town on the very edge of the area; Two medical school faculty members, both internists, one in the department of community medicine in his school; the other from the department of medicine in the other medical school; One psychologist-administrator who directed a community mental health clinic nearby, which was involved with several Federal poverty programs in the area. It seemed almost impossible for all those who had to be involved to come together on anything. The pathologist was viewed as primarily interested in increasing the number and capability of the labs he operated in the hospitals, and otherwise, in consolidating and extending his local leadership position. He was quite sensitive to the possibility that the medical schools might try to tell the local physicians how to practice medicine and that they would contribute nothing material to the area, although,he believed, their influence was needed. The two general prac- titioners were viewed as watchdogs for their own communities, primarily interested in the improvement of hospital facilities in tneir own communities. They had the least time for committee meetings and were the most guarded in voicing their ideological positions, except on the level of proclaiming that "what's good for the community hospital is good for health care." With respect to the academic physicians, the general practitioners were watchful but not as outspoken as was the pathologist about local doctors losing patients to the teaching hospitals through the criticisms of chief residents who said -- or at least implied to small town patients that local physicians were incompetent. The psychologist worked well with the physicians and had their respect, but he was viewed as being quite radical. He tended to see the physicians-as being willing to cope with a smaller segment of the problem than, in fact, they actually were; and though.he was very much in command of himself, he tended to be suspicious of the physicians' self-interest. The academic physicians were united on nothing: one favored hospital-based medicine as the needed solution; the other wanted more community-oriented programs without emphasis on hospital or other facili- ties. One wanted leadership on the Committee; the other came because somebody from Amsterdam had to come." Both tended to feel that medicine as practiced in the communities in the area was deficient, and were ideologically committed to academic medicine. All the committee members had some doubts about their abilities to cope constructively with all the others. All of them felt as though devising action programs for the community involved them in fairly deep water. When they began to interact to survey the health needs of the area, connect with community leadership, and concoct an action plan, dynamic difficulties emerged. The Professor of Community Medicine in the larger of the two medical schools and the circuit-riding pathologist, IV-29 Arthur D Little, Inc who was practicing in small community hospitals were perceived as powerful physicians. They had the greatest difficulty collaborating in any concrete ways without seriously antagonizing either the local mental health organizations (which were tied to suci-I consumer-poverty groups as existed) or raising the suspicions of the regional funding body (one of the regional Economic Development Commissions based on the model of the Appalachian Regional Commission). It took a number of meetings to get the private practice and academic physicians to agree even on what the deficiences were in health care in the area. To develop tolerance and mutual acceptance between the representatives of the two medical schools was also an issue. When achieved, any agreement among the committee members was slightly suspect by both the hospital administrators and the state and local public health officials in the larger group, on the ground that it must somehow simply serve the personal interests (or at least the professional interests) of the committee members themselves. A. ADDITIONAL COMPLICATIONS-: THE REGIONAL COMMISSION STAFF Worse still, the Regional Commission staff could not get much of a fix on-the "health care people," though their program was reasonably well cued into the political processes of the several states in its bailiwick in terms of such public works as highway development and such accustomed examples of economic development as creating and building new business enterprises. The Health Care Committee -- the smaller ad hoc voluntary group of able and dedicated, but mutually suspicious, people described earlier -- saw the Regional Commission, and most particularly its staff, as espousing an ideology of central Federal control over whatever projects were to be undertaken. But the Regional Commission staff was probably not in the grips of such an ideology; certainly its members were well aware that only through local initiative and local action could anything be accomplished locally. Furthermore, there were no Federal agencies, and no great amounts of tax money available to do very much "controlling." But the attitude of the Commission staff actually did represent a kind of paternalistic ideology. Without really knowing very much about the competence of the people with whom they were dealing, the Commission staff had concluded that: (1) These local people would never agree on what needed to be done (though the Committee, in fact, steadily approached agreement), (2) They lacked administrative competence and leadership to carry out any very extensive projects (though individually, several members of the Committee had pioneered and successfully managed good-size activities), (3) Conceptual models of clinics and health centers developed elsewhere and advocated by some among the staff members of the Regional Commission would be adaptable to local purposes and readily acceptable once people quit fighting IV-30 Arthur D Little, Inc and got down to work (though everybody recognized that programs brought in from outside would face very harsh opposition), (4) The Committee was not really local, but was dominated by outsiders.(though four of the six members lived in tne area itself and the two medical schools were the only ones close by). As long as the Commission staff looked on the local people both as incapable and as being inadequately representative, the Commission naturally could give the local action groups very little encouragement; and as long as the Commission staff was unwilling to waste its time on people judged to be incapable, there was no way that their perceptions of the local capabilities could change. There were, of course, consid- erations in addition to these ideological ones. The Commission staff was trying to create a strategy and a set of priorities for the entire region that included many aspects of life -- not just health. This required their developing a relevant and coherent picture of what was going on in the region and assessing what energy and power could be mobilized to address its economic development problems and hence justify a substantial Federal appropriations So there was a good deal of ambiguity, vagueness, and the kind of confusion that comes with working one's way into a problem on the part of both the Commission staff and the local health care committee. B. ANALYSIS OF THE FACILITATIVE EFFORT RMP played an important role in the project, making available its own communications network to help establish connections between the two medical schools, among the 11 hospitals, and among miscellaneous other agenices. RMP recognized that its own interest and "ideology" would be only one more complicating factor and that some outside energy would be required before anything could happen at all. RMP accordingly introduced outside consultant facilitators temporarily to provide neutral help so that the people involved might overcome as much of their mutual isolation and suspicion as possible and to make it a little bit easier for ideas (like shared participation in a CHP(B) Agency) to be perceived as more than just one more manipulation. The consultants were also charged with helping the Committee carry out a community survey of health needs, somewhat uniquely intended to get people in local communities directly in touch with the Committee to express their own sense of local needs, thus generating tfata and confrontations between providers and consumers useful in formulating an action plan. In this instance, the local RNP was quite aware of the sensi- tivity of the Regional Commission to RMP's interest. The Regional Commission staff believed that the whole idea of a health care planning process in the area had stemmed from their impetus, although it antedated their creation by 10 years or more. RMP diagnosed the Regional Commission staff as IV-31 Arthur D Little, Inc wanting badly to establish itself in the area, and anticipated a great fight for sponsorship if the work of the Health Care Committee amounted to anything. The Regional Commission staff believed pretty much the same thing about RMP, but in reverse. One additional complication was RNP's ability and willingness'to put more planning money into the process than the Commission staff could. Contact between the staffs of RMP and the Regional Commission was limited enough that each tended to view the capabilities of the other as they had been a few months earlier, at the last contact between the two. Each staff habitually tended to judge the other as being much less able to do effective work than was probably in fact the case. The result was that most contacts between members of the two staffs resulted in a certain amount of mutual "putting down.@' Unfortunately, the consultants had much more to do with the RMP staff than with the Regional Commission staff and, accordingly, were seen as much more closely related to the RMP staff. Since they were perceived as "being in the pocket" of the RMP, their usefulness in bridging the gap between the Regional Commission staff and the RMP was marginal most of the time and positively harmful on some occasions. The consultants in general attempted to increase the degree of mutual acceptance among people isolated from one another and/or accountable to competing organizations. Their attempt to build collaboration worked fairly well so long as the small group originally engaged in the work of the Health Care Committee had to have few contacts outside itself, but broke down in dealing with the Commission staff. In the Committee, private interests were so obvious that when one member began to become suspicious of another it was usually easy for the consultants to find a way to ask the suspicious one whether he was indeed beginning to doubt the motives of the other. This rather simple-minded device, applied with some discretion, usually cleared the air, a good enough relationship having been established among the members and the consultants that it was considered safe for the accused to demonstrate more clearly what his motives really were. Implicit in almost every group-shared ideology is the proposition that some powerful enemy is go ing to take over if "we don't defend our- .selves vigorously." In this case, RNP was perceived initially by some or all of the practicing physician participants in the Health Care Committee as just such a powerful enemy. That problem was dealt with simply through repeated and free exchanges among the pathologist, the professor of community medicine, and the principal RMP staff member involved. While none of these men started by completely trusting the other, very significant changes in their mutual attitude did occur during the course of the project. C. THE CnNr.FPTITAT C!nT TTn'ION There seemed to be relatively little that could be done to cut into the situation swiftly, since there were so many of the attitudes and theories "ideologies" -- in conflict with one another. All of these ideologies served as justifications for not trying to work with people in IV-32 Arthur D Little, Inc groups other than one's ovrn, because each ideology defined the other people as incompetent, greedy, and unwilling to collaborate. However, some of the local physicians were already interested in a continuing education organization (an "Academy of Medicine") started by the circuit-riding pathologist. Some of the agency representatives were interested in forming a local Comprehensive Health Planning group (a so-called (b) Agency). An imaginative proposal to enlarge the board of the Academy of Medicine and change its by-laws.so that the Academy itself could satisfy the requirements of the (b) Agency was a key to breaking down these particular barriers. It allowed all of the local groups to feel they had a piece of what was going on; it also involved the formal political structure of the states to a degree that made the Regional Commission staff trust the combined available competence signifi- cantly more than had been the case. So, progress began slowly to be made, after virtually every permctation had been tried out conceptually for structuring the Committee, the Academy, the (b) Agency, RMP, etc. All parties wanted to to something toward collaborating: search for a solution, therefore, went forward. The building of a common concept in which the unique elements are constantly emphasized usually provides considerable positive-enthu- siasm and creates a common identification. However, limitations on time and the difficulty of repeatedly bringing people together from great distances made it impossible to share enough of what was going on with a large enough number of people outside the Committee itself to permit maximum exploitation of the very real uniqueness of what these people were attempting to do: the system integration and problem-solving enabled by the structure they tried to create. D. INTERIM RESULTS Despite all these difficulties, the project is still alive and active. The partial mergers of the various structures (the anticipated beginnings,of a (b) Agency merged with the Academy of Medicine, in particular, and the Health Care Committee becoming an agency of the Academy) enforced a somewhat larger amount of communication. But a great deal still remains to be done in keeping the ideological barriers down between the various professions, communities, and government programs involved. The strategy has been to make a local staff available to the Health Care Committee, which is attempting to create specific health care projects to tie together the isolated interests and communities in order to improve access to primary care and to emergency medical care. E. SUMMARY: IDEOLOGY AND FACILITATION o The facilitator has to deal with confusion and complexity, because people espousing different viewpoints decide to ally (and to fight) in ways that are essentially unpredictable, once the situation is thawed out a little bit. Each successive re-alliance affects IV-33 Arthur D Little, Inc everybody else in the picture, and the facilitator had better try to find out how to cope with the situation. o There comes a point when building a common concept is appropriate to dealing with ideological differences, since the new concept may be capable of absorbing the old ones. But the new concept may still look suspiciously like a self-interested ploy to people who were not involved in developing it continually. Extending the circle of participation is the reasonably way of bridging this gap and is reasonably easy to accomplish in a fairly well organized situation with willing leadership. It is harder in an unorganized setting. o Ideologies usually are perceived by their proponents as defensive explanations; definitions of "their" territory and the basis for the claim they make. But outsiders perceive these claims as being threatening, aggressive, diid empire-building assertions against which defenses are needed. o In unstable or swiftly changing settings, this further implies that the facilitator has to deal with changing gaps between capabilities and intentions, which adds still another realm to the possibilities for mutqal put-downs and conflicts over territory that are really conflicts based on mutual misperception. a. When the facilitator is attempting to deal with people who do not share the same ideology and who are, in fact, in possession of conflicting ideologies, the facilitator must invent specific ways to get them to focus on the complexity of their own and one another's motivations. He must help them find ways to test possibilities, other than the self-interest they expect to find as a consequence of their ideological involvement. o Facilitation also involves a sense of pushing-ahead, of a very special kind of problem-solving orientation. In it, the diagnosis and pre- scription tend to be formulations that will allow people to feel and discover that they can participate in the process of solution and are not just being treated as parts of the problem. This is par- ticularly important in instances in which ideological differences are present; for as people come to feel that they have more control over the ongoing process, the fact that others share this control becomes less threatening and bothersome. IV-34 Arthur 1) Little, Inc ADDENDUM #2 HOSPITAL NETWORK IN WESTERN NORTH CAROLINA This addendum discusses changes in which RMP in North Carolina participated in the structure of hospital-based medicine in the Great Smokies Counties. The changes amount to a significant reduction in institutional isolation. Individual hospitals beg@n to shift away from the concept of developing all services appropriate to major urban community hospitals. They had become aware that several or all of the small hospitals (around 40-60 beds) could cooperate anct offer a more integrated set of clinical services which, in total, would exceed individual services. The story is rich with examples of facilitative behavior of a variety of kinds, and tends to show the entrepreneurial opportunism that so often characterizes facilitation in a social setting where power is widely scattered: getting problems to be solved demands more than a common recognition of an altruistic need; it also requires seeing how various individual interests can be melded into the common interest. A. STARTING CONDITIONS Eight hospitals having about 350 beds in 7 counties, and about 64 MDs (50 active) serving a population of around 85,000, were involved. While many of the hospitals were comparatively new products of the Hill- Burton program, some were not accredited and others were under a threat of disaccreditation by the Joint Committee (a threat capable of providing energy for action, when viewed perceptively). The hospitals and the doctors were relatively isolated from one another. County medical societies more or less corresponded to the medical staffs of individual hospitals. The hospitals were connected by reasonably good or excellent roads; under most conditions, it was possible to go from any one of the hospitals to any of the others in 2 to 3 hours. Adjoining hospitals could be reached in less than an hour. When John Hayes, then on the North Carolina RMP staff, first attempted to do something about the situation under RMP auspices, he probably had no fixed'idea of what the solutions to his problem would be, though he knew that he was working to meet the threat of disaccredi- tation and had some feeling, apparently rather early in the process, that there might be a way of accrediting all the hospitals jointly without any one of them fully qualifying for accreditation by itself. He very early shared his concept with others in such a way that it became, or was seen as, their own idea, too. Hayes, backed by Dr. Marc J. Musser, then the North Carolina RMP coordinator, had two objectives in mind: (1) to do something about the disaccreditation threat, and (2) to establish RMP in that part of the state. Because of the categorical restrictions on RMP, his project activities were 'somewhat constrained, but what he espoused could always be viewed broadly with respect to possible impact on the IV-35 Arthur 1) Little, Inc hospital and health care system as a whole. His view was never limited to projects merely being good things in themselves, justified in terms of the heart, cancer, and stroke program. Even though they were planned and carried out in such a way as to meet the categorical criteria, the broader goals were equal if not more significant. (The categorical stipu- lations can be a definition for the RMP program, or a fatal handicap. To the skilled facilitator they have been a constraint but often a useful one.) At least two other conditions helped to shape, organize, and impel the process. One was the concept of the State of Franklin, resurrected and propagated by Dr. Carl D. Killian, who had long,been concerned with unifying, educating, and activating the people of the western counties in a common course. OEO, Job Corps, Teacher Corps, Office of Education, Appalachian Regional Commission -- wherever a hope existed of finding grants in aid, Dr. Killian sought out the people in charge, both locally as well as those on the Washington scene. A descendant of the people who had first settled in the Smokies, Dr. Killian always kept before him the image of the shy, isolated, adolescent boys coming down out of the hollows in which they were born, somewhat gingerly, to sample secondary schools. To Killian, these boys were symbols of both the poverty and isolation endemic in rural Appalachia and the promise of doing something about these problems. For him, the ancient notion of a State of Franklin (formerly "Frankland" or free men) to represent the community of interest in the southern Appalachians retained its appeal after nearly two centuries. He saw it as a reminder of how much there was in their inheritance of which these people could be proud and how strong were the ties that bound them together. "Frankland" was not intended to isolate the people of the mountains further, but to preserve just a degree of clannishness and develop just an additional modicum of local pride, on the basis of which to generate self-help programs. RNP accepted the "State of Franklin" as being a mechanism operating to bind the people together in the interest of solving their own problems; accordingly, the State of Franklin represented real grist for the RMP mill. Each identified.and recognized the other as a potential source of strength; the leadership on both sides saw such identification of strength in pre- viously untapped places as a part of the role of any facilitator-leader. Another situation which bound people together in these counties pertained primarily to the physicians alone. Though a few of the older men were reputed still to go along the roads and trails on horseback, dispensing aspirin and sulfa drugs as they went, most of the doctors had been trained in a newer style. They felt good medicine required good hospitals. They accepted the concept of medical specialization, even though it seemed to them practically unattainable and economically some- what threatening. There were, nevertheless, a number of men who prac- ticed more or less as specialists, and several who had been board- certified in pediatrics, internal medicine, or even a sub-specialty like cardiology. Thus there was the actuality of local referral and some potential for expansion. IV-36 Arthur D Little, Inc What brought these men together, however, was their common practice of referring no patient to Asheville for specialized medical help if it was at all possible to avoid doing so. The problem was a simple one: for most of the people in these counties, Asheville was only an hour or two away by car; major roads converged on or passed near to Asheville. Once a patient had found his way to a physician in Asheville: the rural doctors knew they could lose him, since access to Asheville medicine was relatively easy. Instead, patients were referred to more distant centers, notably to Charlotte, 150 miles to the East beyond Asheville or to Winston-Salem, still more distant.' While it would be easy to overstate the importance of this checkerboard or hopscotch referral pattern, what it really communicates is the peculiar dilemma of a doctor in local, isolated practice for himself. In his own being, he is the major and sometimes the only medical care "resource" available to his community. What is -joing to be done, he does; he must accordingly be prepared to undertake a wide variety of tasks and to feel that lie has the competence basically required to do the job. He has little oppor- tunity to involve other health care personnel and facilities in the treatment of his patients; neither does he have much incentive to do so, unless he too can begin to scale the heady heights of a referral practice of his own. So the local doctors had at least weak reasons to band together: they could see the further development of medical centers in Asheville or other nearby places as a threat, and at least some of them could accept the theoretical virtues of continuing education to keep them "up-to-date." RMP leadership knew about these factors; they recognized them as impor- tant; they advocated no plan or planning process that violated the con- straints initially imposed by these conditions; RMP, for example, wasted no energy on building up relationships with either local or Asheville physicians. B. PROCESSES OF DEVELOPMENT Dr. Killian, John Hayes, and others helped Dr. Hugh Matthews and others who at first responded to visits from the Durham staff of RMP to organize a Multicounty Academy of Medicine. The Academy, which was to serve as the medical continuing-education aspect of the "State of Franklin," was open for membership to only five or six dozen people: the MDs who lived and practiced in these counties. Though the Academy became a dues-paying organization, it could obviously support very little staff work. Accordingly, the group looked to the RMP staff to help them with their planning and sometimes to lend a hand with such administrative chores as getting out agenda and minutes and setting up meetings. Perhaps more important still, the RMP staff members made it clear to the leaders of the Academy that they were there to ofker technical support and encouragement in accomplishing something which everybody foresaw as difficult; namely, developing a group of 60 physicians into a group strong enough to take a constructive role in planning how to meet the health care needs of the irea as a whole. The physicians were, however, IV-37 Arthur D Little, Inc perceived as the health care cadre for the area. Given this strategic judgment, proceeding without them would have been, to say the least, non-facilitative. The planning process of the Academy, as well as that of all the others involved, was both set back and accelerated by the emergent awareness that one of the strongest of the hospitals had proceeded in resources development, that it felt capable of becoming a regional center with which the others could affiliate, if at all, only as dependents. These were terms in which the others were not particularly interested. This circumstance had some effect in convincing the rep- resentatives of the other institutions, however, that they would have to strive more enthusiastically to overcome their differences, since further lapses into institutional isolation would sharply weaken the impact of their joint effort and -,reduce the advantage of collaborating across institutional lines at all. Specific RMP I)rojects also played a part in the developmental process. One project was in diabetes; another was to create functioning coronary care units and train nurses and others in the techniques of coronary monitoring and associated therapies. Important local sponsor- ship for the heart projects came from a dedicated cardiologist, Dr. Ralph Feichter -- one of the two cardiologists in the State of Franklin -- who formed a relationship with academic physicians at the Bowman- Gray Medical College in Winston-Salem. Bowman-Gray people would provide technical backup necessary to perfect local training curricula and provide consultation. This relationship had, of course, some real payoffs for Bowman- Gray, like other aspects of the processes we are describing. Voluntary relationships work only if they satisfy some of the immediate interests of the people and institutions taking part. In this case, Bowman-Gray, although it had begun operations in Winston-Salem a generation ago, was continuing its process of building a substantial local constituency. As the size and complexity of the medical school increased, so did its needs for outreach. But like most other medical schools, Bowman-Gray was able to take part in RMP projects only because these projects brought in some additional money to provide partial support to faculty and other staff members. For the cardiologist, Dr. Feichter, the other heart specialists, surgeons, neurologists, and others, Bowman-Gray offered some professional stimulation and a chance to make sure that its own approaches were up-to-date and as adequate as possible, as well as potentially extending and reinforcing the quantity and quality of the service it could offer in its own hospital setting. So this project alone could plausibly increase quality and access to care for any patients able to profit from treatment and supervision in a coronary care unit. These units were established in several hospitals. Nurses began to be trained on a rotating and shared basis. Coronary mortality dropped by about the usual 30% in the hospitals, and circulation of patients and doctors between the State of Franklin and Bowman-Gray increased somewhat. IV-38 Arthur 1) Little, Inc Not all the hospitals in the State of Franklin, however, were equipped with full-fledged, fully staffed, coronary care units. Tlle population did not call for such staffing; costs of training and equip- ment precluded it in any case. But plans had been made to upgrade training capabilities in certain other paramedical areas with the training sites expected to be hospitals other than the one already training nurses for service in the coronary care units. There was a drastic shortage of physical therapists, for example. There was also talk of sharply upgrading, extending, and formalizing the training of X-ray technicians. A'practical nursing program @ad already been insti- tuted; and plans had been made to upgrade it as well. At this stage the intention was to train people in the hospitals best able to do so, the assumption being that a good many of the hospitals would turn out to be uniquely superior in one field or another, but without so much duplication that competitive pressure would distort or delay creation of small in-house training programs of great potential benefit to all the hospitals in the area. At least some of these programs would probably be developed under the general supervision of or on the premises of the Western Carolina University (Dr. Killian's school), thus further significantly linking together economic, edu- cational, social welfare and health development programs for this North Carolina area of Appalachia. Dr. Killian and his associates had the reputation of being both imaginative and successful in finding multiple sources of funding for necessary programs which could lead legitimately to such funding. In developing these programs people were trying to apply a rule of thumb that seemed at once to minimize local competition and was intended to maximize the chances that the benefits of the programs would accrue locally rather than elsewhere. This rule of thumb was based on the observation that those who went 100 miles or more to be educated had already proved their mobility by that behavior. The likelihood of their returning to the communities from which they had sprung was already demonstrably low. The local intention was to create training schools close enough together that nobody would need to travel more than about 40 miles to reach the site of such a training program where periods of training more than a few days in length were to be involved. The RMP core staff dominated none of these activities, nor (except for coronary care unit training) were RMP dollars the principal source of support. But already in 1968 the leaders were repeatedly ready to testify that if RMP were to collapse immediately, it would already have served a vitally necessary function in getting people to begin to work together across community, institutional, and professional lines, in ways that promised both short-range and long-range benefits. The help most often mentioned was on the level of sheer human support and willingness to help. Almost all of the local health care leadership involved said that RMP involvement in the process had again and again meant that people discouraged to the point of almost giving up had received that kind of constructive advice and approval and the little bit of additional push it took to get them going again. On all these IV-39 Arthur D Little, Inc levels, which had little to do with "projects,t' RMP core staff members seemed to be effective: the names of Messrs. Musser, Hayes, and Holder were often mentioned and well regarded. The chief criticism of RMP was the unavailability of sufficient help in thinking through and drafting project applications. Hospital administrators and other rural members of the health establishment in Western North Carolina were aware that the effort involved in drafting a project proposal -- particularly for people inexperienced as "grantsmen" -- could easily cost as much money as the project would be funded. This combination of circumstances resulted in a counterpressure on North Carolina RMP to provide staff time, or help with grantsmanship. The men of the State of Franklin valued RMP at least as much for its other capabilities as they did for its power to provide project money. C. ENDS IN VIEW Early in 1969, it was possible to be fairly specific about which of the hospitals involved were expected to offer which shared services. 'If all of the hospitals taken together were to share joint accreditation certificates, this carried with it an increasing speci- ficity about internal referral patterns, and at least implied the beginnings of another rather clear possible objective. It looked as though an atmosphere could arise in which some of the basic allocation decisions could be made jointly among the institutions. Whether or how soon this possible goal would come to be a genuine aspiration could not at that time be determined, but enough had happened that anyone interested in specifying such a goal could also have begun to sketch out a more or less plausible strategy and process for achieving it. By June 1970, additional steps had been taken. A Hospital Commission had been proposed to be the ongoing coordinator of shared activities among the hospitals. By-laws had been drafted, and the eight Board of Trustee members had agreed to support a joint effort,-- first (probably), to administer and serve as fiscal agent for the continued funding of the coronary care units, and then to take on additional tasks, such as acquiring a skilled dietitian to devise and provide dietary in-service training for all the hospitals. Other common support services (purchasing, laundry, etc.) were contemplated. Even though the hospitals are relatively small and scattered, connections among administrators and boards had become strong enough to make such steps somewhat attractive, and long-range, tentative goals for further consolidation almost ready for discussion. In addition, Dr. Matthews had moved to Western Carolina University to assist in building a School for Allied Health Professions, primarily on the four-year B.S.-degree model, but pending development of a community college, including two-year programs as well. A training program for X-ray technicians, for example, seems considerably closer to realization. IV-40 Arthur D Little, Inc The attitude toward collaboration with Asheville appears to be shifting. With the designation of planning areas now formalized, the State of Franklin and the Central Highlands (Asheville) have both realized .that they do not want to give up their separate identities either. Further- more, concerns about domination or being swallowed up naturally persist. Asheville physicians want to be recognized. Their medical center has considerable sophistication. They believe it is one to which State of Franklin physicians can properly refer patients. The planned School of Allied Health Professions in Culowhee in the State of Franklin needs a municipal base from which to recruit students and in which to place graduates. So there seems to be emerging circumstances to promote more cooperation that will allow facilitative efforts to succeed in overcoming the old barrier between Asheville and rural physicians so long as the new circumstances are perceived in this way -- which they are. The Academy of Medicine appears to be a viable and useful agency, too. One of its current activities is to recruit new physicians for the State of Franklin. Some 2,000 brief letters of solicitation have brought in over 60 responses so far. Whether any new physicians materialize from this effort or not, it has had a favorable effect on local people, and it is a sign that the doctors are doing something actively to help with the manpower problem. In an increasingly real sense, the State of Franklin appears to be a way to cut across county lines: to allow health and other social service functions to-more readily organize themselves appropriate to the population distribution and their own scale requirements. In the words of Dr. Matthews, "There's less talk, but more action." The State of Franklin also represents an example of how, in the continuing processes that involve RMP, new possibilities emerge which can shape the basic purposes of RNP activity. What had started as a continuing education program in the coronary care field and an effort to stave off the closing of some small rural hospitals would have to proceed along its planned track, but new, additional, important objectives became visible in the process. Partly through RNP, these specific emerging objectives began to be incorporated into the process by local leaders, working both to concentrate the use of scattered local health care resources and to increase the amount of these resources. IV-41 Arthur D Little, Inc ir V. EVALUATION Arthur D Little, Inc I V. EVALUATION A. INTRODUCTION Evaluation is a normal part of intelligent individual and organizational behavior. An individual "evaluates" every time he stops to consider what he is doing, as a means of judging how well he is performing a certain act, or whether he can perform it better. The judgments can be reintegrated into the thought processes to effect changes in the behavioral patterns of the acts being evaluated. Diagramatically, the process might look like this: Action bV Individual (Work) (Implementation in Organization) Perception of Consequences Reformulation of Action bV Individual bV Individual 10 (Planning) (Ju'dging) lation bV Management (Evaluation bV and of Organization) of Organization) In organizations, evaluation may be considered as accomplishing three different purposes: o Justification: to defend what is planned or what has been done. The terms must be acceptable to higher levels of official accountability. "Justification" becomes a basis for reaching (1) agreement that what is being evaluated is worthy (or unworthy) of the support it is getting, or (2) a temporary truce in an ongoing argument about the worthiness of what is being evaluated. o Control: to obtain performance details ("monitoring") that management can use to make behavior conform to a standard. o Learning: to help the evaluated activity transcend itself, by developing new (or more explicit).goals, techniques, or strategies. "Learning" in this sense leads to the creation of new standards rather than conformity to old standards. Activities producing a more or less uniform, specific, and concrete line of products can emphasize the control aspects of evaluation V-1 Arthur D Little, Inc without risking the vitality of the operation. Activities that have broader functions, or whose products change swiftly, must emphasize the learning aspects of evaluation. If they do not, formal evaluation will effectively be ignored in the real workings of the organization, or the activity will turn into one with very specific and unchanging products, or it will lose its vitality. 1. Importance of Systems Bationale A systems rationale is developed in activities with narrow and specific goals as well as in those with broad and general goals. It stipulates objectives, the nature of the operations needed to achieve those objectives, and at least vaguely provides a methodological basis for evaluation. In terms of evaluation, one of the most crucial functions of the rationale is to provide a way of drawing boundaries around what is to be evaluated, and to decide how intensively the evaluation process should be pursued near these boundaries. But the rationale of the activity or "system" also does other things for the evaluator: o It tells what is relevant in a field of observation that contains more information than the evaluator can handle; o It guides him toward the several kinds of clients and constituencies concerned with the activity, and tells him something about their interests and needs (customers, employees, competitors, etc.); 9 It suggests performance criteria; o It helps to specify what is to be evaluated; o It suggests testing methods which, of course, may imply comparison with a fixed standard (for example, simple abrasion tests for rubber), or which may be very complex and quite vague in their connection with the activity evaluated (for example, the College Board examinations); and 9 It sometimes defines measures and standards of performance, which are numbers used as indices. For example, in the health care system, infant mortality rates often serve, though inadequately, as per- formance measures for the health system. The minimum mortality rate believed to be attainable then becomes the standard of per- formance. 2. The Rational Manager's Model of Formal Evaluation Managerial theory and bureaucratic practice tempt one to develop the systems rationale and its qualities into a highly logical structure. The result is a set of detailed, rational questions that allow the evaluator to test whether behavior of the activity evaluated v-2 Arthur D Little, Inc conforms to systems rationale. This process leads to a very particular kind of evaluation system in which the evaluators are supposed to be objective observers of the activity. Thus, evaluation becomes a process that includes: Questioning; Information-gathering; 0 Analysis (i.e., fitting information into a conceptual framework to generate a coherent description, comparison, and assessment against the standard provided by the system rationale); o Storage and assembly to allow the performance of many sub-units of the activity to br- compared and to allow comparisons through time; o Information transfer (i.e., making evaluation information available for people to use); and o Use: incorporating evaluation information into planning and policy-making. This evaluation process will not work unless everybody in the organization is acting rationally, as rationality is defined by the purposes for which the organization was founded. Such rational beings, therefore, will naturally allow the outside evaluator to compare what is actually happening with the job descriptions and mission directives that express their accountabilities. Evaluation of this kind is based on the notion of auditing, and it emphasizes the evaluator's role in "Justifying", that is, in testing the justification of the organization. In large and complicated organizations, these evaluative justifications may also be wanted for control purposes. But they are too voluminous to be used, and tend to be summarized in even more abstract form. This keeps the top managers from being overloaded witn information, but provides them with essentially unrevealing data on the basis of which to make management decisions. 3. The 110pen" System or "Discovered" System But most people seem unable to be "rational" all the time, and the organizations they work in actually operate as social systems with far broader purposes than the formal systems rationale would suggest. These purposes include qualities like survival, local inde- pendence, a more or less combative interest in territory, and an interest in maintaining stability (the status quo). This whole concept roughly relates to the distinction between formal and informal organi- zations that coexist in the same institutional structure. V-3 Arthur D Little, Inc We are, however, talking about something still deeper than informal organization. These same survival qualities can be the lively expression of an organization's attempt to deal with the perception of its members that its goals are changing, its tools in need of re-shaping to meet new problems, and its future undecided. In the midst of so much change, it means little to evaluate such an organization against the fixed standard of last year's systems -rationale. The problem is to evaluate its work in terms of the open". perpetually new goals and activities. If innovation is itself a goal, then the evaluation s system must work in such a way as to encourage new things to happen. Yet, an evaluation system based primarily on the model of the audit tends to discourage innovation. Auditing tends to make behavior conform to the systems rationale. The kind of evaluation system we want to develop is just as likely to be used to alter management behavior and the organization's goals as it is to alter the operations.of the organization to make them conform to a systems rationale. For example, we (Arthur D. Little, Inc. - OSTI) want our coexamination of RMP to change the goals of RMP to take account of the facilitative activity we have observed. The "auditor-evaluator" would take the view, on the other hand, that "facilitation" is at best an incidental aspect of RMP and that behavior and operations in the regions should be made to conform to the "obvious" purposes of RNP as stated in the law. And if these purposes turned out to be less precise than he would like, he would castigate the program for that shortcoming. This concept needs to be explained more completely, because it is crucial to developing and maintaining an evaluation system that is a learning system. The next sub-section deals with this issue. 4. Consequences of Coexistence of Rational-Closed and Real-Open Systems Most organized activities partake of some of the qualities-of both rational-closed and real-open systems. For example, agencies within the Executive Branch of the United States Government may be structured to deal with problems in an open-ended way, but they are subject to a great many pressures to rationalize. They cannot live long without being asked to produce a systems rationale that leads to organizational charters and hierarchical accountabilities. These always fail to embrace all the realities of human experience within rational categories. Sometimes it is purely a question of control and accountability. (The Executive Branch usually attempts to control.) Accountability is most easily maintained when one is tightly in control. Available canons of administration make control depend on rationalization. So the manner in which the closed system of the rational manager relates to the discovered, open social system becomes a matter of considerable interest. To make the closed system work requires some tacit recognition of what real people will tolerate. V-4 Arthur D Littk Inc To the extent that the rational purposes of the manager coincide with real interests of people in the system the co-existing systems of the rational manager and discovered reality overlap. Equally important, when the two systems coexist, they may be connected in the sense that they are more or less able to influence one another. The discovered system may be more or less effective in modifying systems rationale, and the rational manager may be more or less effective in subjecting the real, informal, social open system to acceptance of his system's rationale. But first, we will deal with other relationships to evaluation. Where the two systems have little overlap and little inter- action, evaluation is almost forced to take the form of retrospective justification. What can the evaluator do except to prod'uce statements believed neither by the producer nor by the consumer, which are generated ritualistically in response to formal demand? When this happens, rational managers produce justifying statements at regular intervals, expressed in the language of the systems rationale, and resources continue to flow into the system. Evaluation processes have no other output than justi- fication. They are used neither to modify the systems rationale nor to f-orce the real social system to conform to it. Where there is little overlap, but the rational manager seeks to impose systems rationale on the real, open social system, several things may happen: (1) The real social system may respond verbally, without other changes in behavior, by offering pro forma retrospective justification long on language, but short on substance, a process generally known as "conning."' The two systems operate substantially in parallel. (2) The real social system may respond to the controls that the rational manager seeks to impose by adapting to the evaluation measures he prescribes, but continuing to operate as much as possible as before. Measures of performance are always different from performance itself. For example, in an effort to control expenditures of the vocational rehabilitation system, Congress demanded to know how many "rehabilitations per year" the agency effected for a given investment. "Rehabilitations" were defined as job placements lasting three months or more. As a consequence, the vocational rehabilitation system began to "screen" its clientele for those most likely to graduate to job status leaving out those who were most in need and least able to qualify; to select low-level jobs for graduates so as to facilitate entry; systematically to avoid distinguishing between a "case" and a person, so that a graduate who had achieved job status, lost it and returned to training, could be counted as another "rehabili- tation"; and systematically to avoid follow-up of clients after three months. (3) The real system and the rational system may fight one another more or less openly until they reach a compromise. From the point of view of the real social system, this is paying a price. V-5 Arthur D Little, Inc Those in the system do some of what the rational manager wants in order to preserve considerable ability to satisfy the interests of the real social system. From the point of view of the rational manager, the real system is merely distorting system objectives in the direction of its own interests, but he has to put up with it to get any response at all. In none of these dissociated cases is there any interest in producing or using information that runs counter to the strategy of evaluation as justification. Where the systems are operating in parallel, but without much contact, there is common interest in avoiding information that threatens dissociation. In the other two cases, there is common interest in information that supports the system's rationale, since justification rests on the systems rationale and resource allocation rests on justification. The real system people are content to generate informa- tion that conceals how great the discrepancy is between the goals of the rational system and the behavior of the real system in order to protect the resource allocation they need to continue their doing more or less what they desire. However, where the whole activity is conceived as a learning system, then relationships between rational and real systems can be fundamentally different from those just sketched. The opportunity for learning is primarily in the real social system, which offers the most vital basis for reformulating systems objectives and redesigning systems theory. Discrepancies between the rational manager's system and the real system -- as perceived by its inhabitants -- become the basis for progressive modification of both the systems rationale and the real interests of individual participants, and for developing relationships between the total activity and its constituencies. It is critical that any discrepancies between systems rationale and the real system in an evaluation system intended to play an important role in intelligent management be recognized rather than buried. The evaluation system itself must become a vehicle for continuing interaction and mutual influence of the two systems. When oriented to learning, the ability of the evaluation system to support intelligent, direct interactions between the rational manager's system and the real social system becomes a central function and a central criterion of adequacy. While these con- siderations are important at all times, they become essential in a period of development or instability, when new kinds of activity must be devised to meet established objectives more effectively and significantly. 5. Leaming-Oriented Evaluation in Real Social Systems Hooked to Rational Systems When planning begins to incorporate a mutual modification of objectives and activities, evaluation oriented to learning embraces much more than mere measurement of the extent to which activities conform to specification, and includes such special features as: V-6 Arthur D Little Inc o The conceptual framework for evaluation is based on a description of the real social system as well as the rational manager's statement of the systems rationale, including a description of key actors and agencies, and their actual relationships, modes of interaction, and several interests. It must also include a description of the real (if informal) evaluation system as discovered -- the information that actors in the system in fact produce, are interested in producing, and how they use it. o An analysis of discrepancies and overlaps between the systems rationale and the behavior of the real system, taking into account the actors' differing perspectives. I o Strategies for responding to discrepancies between the real system and the rational manager's system. Mere analysis is not enough; I learning must be capable of application. These factors focus on gathering accurate information about the real system. The discrepancy between the rational system and the real system, or the response of the real system to the rational manager's efforts to control it, may mean that the rational manager is simply precluded from learning what is actually happening in the real social system. But the rational manager may be able to bargain for this informa- tion by exchanging information about resources and ongoing administrative changes to which he is privy for accurate information about what is really happening in the social system, Even more powerful, when central rational management gains some freedom to modify the systems rationale to take account of real local interests and activities, the basis for withholding or distorting information may disappear. The way may then be clear for central rational management and local people to bargain effectively and directly over changes in the systems rationale, local behavior modification, and information flow. As in all such cases, the bargaining will depend on establishing and maintaining good faith. Several additional consequences for the evaluation system flow from these considerations: o Information intended to modify behavior must flow upward to influence systems rationale, as well as downward, to bring the real social system into line with the pre-existing systems rationale. 9 The evaluation information that is gathered should be limited to amounts, complexities, and precisions determined by the capability and willingness of actors within the system to learn from it, as experienced in actual practice. Nobody in the system should be presented with more information than he can reasonably stretch himself to handle, nor should information be laid out in more precision or complexity than he can respond to. Analyses should not present actors with a greater breadth of alternatives than are real for them. As a corollary, the evaluation system should be able to detect the changing capability and willingness of V-7 Arthur D Little, Inc actors to use information, and should itself be capable of responsive modification in turn. o The evaluation process should be structured to accommodate to the different kinds of learning appropriate to different roles and levels within the system (rational managers, project pushers, evaluators,, planners, etc.). o The learning objective should also determine the.content, extent, duration, and accessibility of information in the evaluation system memory. This requirement places high priority on accessi- bility and retrieval capability on behalf of many different levels within the system in addition to that of the rational manager. o Since the learning derived-from evaluation may be applied to evaluation processes themselves, the conceptual framework for evaluation may itself be expected to change (sometimes rather rapidly), so information has to be gathered and formulated in ways that make it more or less equally usable in terms of a broad range of systems rationales. Priorities should be given to those bits of information that are likely to retain high relevance across a range of managers' rationales and real systems. 6. Cases With No Explicit Systems Rationale If the activity to be evaluated is itself recognized as so diverse, diffuse, swiftly changing, and open that no overall systems rationale is credible, then no explicit systems rationale may appear feasible. This situation may occur with respect to public problems urgently requiring solution, but for which there are no clear policy answers, where national willingness to devote resources to their solu- tion is high, though the credibility of proposed rational solutions may be low. Agencies may be funded to work on such problems, constrained only within very broad limits as to what their work should be like. In such a case, the implications for evaluation systems include: o The best possible definition of the problem becomes a necessity, as does an agenda of what is to be attacked. o Each region or sub-region (or other entity) saddled with a problem becomes a center of its own problem-solving process. The number and location will depend on the number of centers that prove capable of functioning under their own individually developed systems rationales. In this situation the distance between information and analysis is minimized, and responsibility for designing and conducting the evaluation process is very close to the actors who are accountable for the activities under evaluation. o In this case, central management's evaluation function is changed with respect to that of the regions. Central management may now impose on the localities criteria for the evaluation process, but V-8 Arthur D Little, Inc it is no longer in a position to impose criteria for substantive evaluation of concrete activities. For example, central manage- ment could still ask whether regional evaluation processes are differentiated in terms of justification, control, and learning, but the central evaluator would accord just as high marks to a region displaying one workable form of differentiation as to a region displaying another form. Only the region that did not explicitly attempt --- through its own evaluation processes to accomplish justification, control, and learning would be downgraded. Accordingly, the evaluation information flowing to Central from the local regions normally reflects the nature of the processes developed for raising and answering evaluative questions in the localities rather than the answers to specific tioiLs posed bv central management. o Central management afso takes on the role of building a network learning system, facilitating information-transfer from locality to locality, and encouraging specific local experiments. 7. Summary For purposes of conducting sensible evaluation, it makes a lot of difference whether the objectives of the activity are broad or narrow, changing or stable, vague or concrete. The implications of these differences are laid out in the preceding introductory section in terms of three functions that evaluation serves: justification, control, and learning. In a stable activity with narrow concrete objectives, evaluation can emphasize control. In a swiftly changing, more open activity in which the realities of the social system inherent in the activity must be supported, evaluation must emphasize learning. Both sorts of activities usually require justification. Organized government activities dealing with broad social problems generally call for the constructive use of both the rational manager's model and the open social system model. Part of what is learned in evaluation in such activities can be used to overcome the discrepancies between the two systems through modification of both. Done sensitively, this enables formal evaluation to be used to encourage the activity to be open to learning, and thus to be open to the acceptance of appropriate new goals. B. APPROACHES TO EVALUATION FOR THE REGIONAL MEDICAL PROGRAM 1. Assumptions About the Regional Medical Program To place the Regional Medical Program in the evaluation context developed in the previous section, the principal characteristics of an RKP should be recited. Some of these characteristics have been discussed and illustrated in earlier.chapters. In this section, they are taken as starting points or assumptions. V-9 Artliur D Little, Inc a. There is no single organization corresponding to RMP, which is a broad Federal program concerned with introducing changes of various kinds into a number of more or less interconnected systems of actors and agencies involved in health care. Within these systems, RMP attempts to play a variety of related roles with respect to other actors and agencies; but for the most part it cannot directly control them. RNP does not, therefore, concern itself with a single rational lisystem," in the sense used earlier, and its boundaries are vague and shifting. From the point of view of evaluation, this assertion has several implications. RNP's scope and turf do not have sharp boundaries. @ cannot be analyzed as though it were a unified organization, like the Veterans' Administration, for example; and while RMP has formulated broad objectives for itself, its fundamental activity in relation to these objectives ;ust be understood for the most part as "influencing" or "facilitating" rather than directly controlling. Thus, its "ends in view" can be analyzed. b. There is no single, established systems rationale either for the health care system as a whole or for RMP in particular. There are various rationales, held at various times and in various contexts by different actors in the system. (See also material under next sub- title: "Systems Rationale for RNP".) c. The larger health care system and the RMP are changeable. They are not in a stable state. The character and functions of these systems are themselves in process of constant change. Within them, the key actors are often unsure of their principal functions or of how best to carry them out, and they tend to shift behavior as they learn and as the system around them changes. d. From the point of view of knowledge and methodology, there are several sources of uncertainty for RNP, which go beyond the uncertain- ties that are characteristic of most -- perhaps all -- other broad social programs. For RMP, the problem of devising and applying per- formance criteria, measures, and standards is complicated by the fact that: o There are several levels of performance corresponding to levels of change in the health care system (change in the process of planning and interaction; change in the configuration and relationships of care-providing resources; change in people's access to care; change in people's health). o The baseline data corresponding to these levels of change are generally missing, or poorly understood, or identifiable only by hindsight. o We have only preliminar ,y theories to help us to predict or to establish relationships among these levels of change. V-10 Arthur D Little, Inc e. Nevertheless, as a Federal program, F4MP is locked into a structure of controls and demands for justification. At the national level these include regular reviews by the Congress, the Bureau of the Budget, and the Department of HEW. These demands for justification and for controls over the expenditure of funds are, of course, passed on to the regional program level. The problem of devising approaches to evaluation for RMP is essentially that of meeting what may well be conflicting requirements for learning, on the one hand, and for justification and control, on the other. The vagueness and changing nature of objectives, lack of program control over components to be influenced, and sources of methodological uncer- tainty all argue for a flexible, process-oriented approach to evaluation- as-learning, whereas the agents of rational administrative control tend to press for firm, quantitative measures of program impact. 2. Systems Rationale for RMP Like most broadly gauged Federal programs, the legislation establishing RMP represented a series qf compromises among the diverse interests of various concerned groups . The authorizing legislation is, therefore, a kind of mosaic of objectives, values, and constraints. Among the more important elements of this mosaic are: o Emphasis on the provision of means of improving the treatment of the three "categorical" diseases -- heart, cancer, and stroke; o Emphasis on the transmission of advanced techniques and knowledge relating to these diseases; 9 Emphasis on both the method of continuing education as a device for this transmission, and the major academic medical center as the principal source of expertise; o Emphasis on maintaining or improving the quality of medical care; o Concern with the region as the principal unit of activity; that is, concern that the program be a regional one, with regional centers of activity throughout the country; concern with recog- nition of a regional diversity of problems and resources; and concern with "regionalization" as a process of knitting together or building regional resources to realize the purposes of the Act; o Emphasis on the establishment of voluntary arrangements among regional institutions as the dominant mode of program activity; and, See Chapter II for history of this process. V-11 Arthur D little Inc o Specific warning against "interference in the interface between patient and doctor." The authorizing legislation made no attempt to rationalize these elements or to resolve potential conflicts among them. In fact, many of the key actors understood that, as the program matured, the specific meaning of the legislative provisions would be developed and clarified. It is not surprising, then, that there have been perceptible shifts over time in the dominant systems rationale for RMP, even though no element originally considered as the legislation evolved has altogether ceased to exert some influence. An evaluation scheme that is generally accepted as appropriate to one of the simplest and, accordingly, most easily rationalized inter- pretations of RMP is the center-periphery regionalizatim model based on the diffusion of technology and information that is assumed to be stored in the great medical centers. In this instance, it would appear desirable to judge the program initially -- at both national and regional levels -- by its effectiveness in reducing rates of mortality and morbidity for heart disease, cancer, stroke,, and related diseases. Individual projects are seen as means to these ends, and fall basically into the following Categories: deployment of new facilities (for example, coronary care units); establishment of new linkages between medical centers and peripheral care-providing centers (for example, exchange of personnel); the development of new working relationships (for example, changes in referral patterns); continuing education (for example, training of physicians and other medical personnel); and information dissemination (for example, DIAL Access). The major kinds of evaluative questions under this interpre- tation of the RNIP system are: a. @at are the kinds of baseline data and measure of performance by which the impact of diffusion projects on mortality and morbidity can be assessed? b. What is the relative effectiveness of the various technologies diffused in relation to cost, i.e., seen as a means of achieving reductions in rates of morbidity and mortality? c. What is the related effectiveness of the various methods of diffusion for particular technologies and for particular regional situations? (This question leads, in turn, to questions about the optimal "regions" for diffusion, the forms of greatest "diffusion impact" for a given investment of dollars and other resources, patterns of utilization of new facilities, and the like.) Other aspects of the activities within the center-periphery model of RNP -- for example, the management of new institutional arrange- ments at the regional level -- must be judged in terms of their effective- ness in enhancing the quality of care through more effective diffusion of technology, with the ultimate effect, of course, of reducing mortality and morbidity from the categorically identified diseases. V-12 Arthur D Little Inc The historical emergence of center-periphery regionalization for technical diffusion was, of course, more complex than we have so far indicated. Other themes influenced, interacted with, and to some extent confused that model; among these were issues involved with the decision to take RMP into NIH, the centrality or non-centrality of medical schools, the orientation or non-orientation of the program to physician providers as the primary constituency, and debates over restriction of the program to the categorical diseases. Although the concept of technical diffusion from centers to periphery continues to have supporters,* a well-articulated evaluation scheme to support it has yet to be developed, and we submit that it probably will not, so long as this model (the most realistic of the simple rationalization interpretations of PL 89-239 as amended) continues to fail to convince even a_,plurality of those with a stake in RMP that it is acceptable and credible. Beyond acknowledging that it would undoubtedly be possible to develop an evaluation scheme suitable to such a program and that a highly rationalized, project-oriented evaluation system would be appropriate to such a scheme, we want to spend no more time on the subject. At'least for the present, simple systems rationales for RMP exclude too much of the reality and tension we experienced in observing RNP in action in the regions. Accordingly, we advocate evaluating what is there and what is emergent, rather than the degree to which behavior conforms to any stereotyped, easily simple model that does not reflect reality. The point of view that emerges in reaction and constructive response to the satisfyingly simple, but unreal, interpretations is that 1W's dominant systems rationale is transformation of the prevailing system of medical care through voluntary cooperative agreements. In this interpretation: 9 RMP's central concern may be expressed through categorical diseases or with the diffusion of advanced medical technology, but RNP consciously concerns itself with overall improvement in quality of care and equity of access to care. o These sorts of improvements require changes in the structure and modes of interaction of care-providing institutions which no single agency controls, -- changes can generally be described as knitting together components of the system that are now fragmented, thus permitting more effective and rationalized planning and action. o These systems changes are necessary conditions for improvement in quality or equity of care, and must precede any significant improvement along these lines. *See the review article by T. Bodenheim, op.cit. But note that there are today no RMPs built essentially on this model. V-13 Arthur D Littk Inc In the past year, systems transformation* has begun to dominate among competing systems rationales for RMP (without, of course, completely displacing other views) at national as well as some regional levels. 3. The Implications of Systems Transformation for Evaluation Under a systems transformation model for RMP: o The primary unit for evaluation becomes the program, and since RMP is conceived as an essentially regional enterprise, this means the regional program. Although it is necessary to reach both "above" this level to the national program and "belo'w" it to the project, the regional program is primary. 9 The purposes of justification, control, and learning remain relevant, but within context they become: - How can we assess after-the-fact the impact of regional programs on the medical care system? - How at the three levels (at least) -- national, regional, pr oject -- can the necessary management controls best be exercised? - How can we facilitate learning about systems transformation, again at all three levels, but with emphasis on the regional program? o Every element of RMP takes on a dual aspect. Regarding project, regional program, and national program, we must ask about specific substantive effects on quality of care, about access to care, and ab.out systems transformation. Seen as systems transformation, RMP functions in two ways: (1) through the direct efforts of the regional coordinator (and those with whom he works) to knit together or otherwise influence elements of the medical care system in his region; and (2) through projects whose efforts effect substantive changes in the provision of care. o The processes of shaping and selecting projects become 'occasions to effect systems transformation. Further, the regional coordinator may seek to design clusters of projects so as to effect systems transformation. Every project and program, therefore, must be examined both for its direct effects on the provision of care and for its role in systems transformation. "RMP as process," "RMP as facilitator," and "RMP as opportunistic change agent" were expressions heard as early as 1967 and conveyed the underlying idea behind systems transformation before this rationale became as widely accepted as it now is. V-14 Arthur D little, Inc o Regional medical programs will share certain attributes: - Certain themes or dimensions of systems transformation; for example, the issue of whether there is "regional identity"; - Stages of systems transformation and the types of questions relevant to each stage;. - Levels or kinds of change taken as relevant from the point of view of program and of evaluation; - Criteria for systems transformation -- ways in which we tell, and measures we use to determine, whether and in what ways the system of medical care has been transformed; - Certain broad features of the evaluation system required by needs for justification, control, and learning in relation to systems transformation; - The "starting conditions" and the coordinator's diagnosis of them; - The issues of medical care taken as crucial; - The ends-in-view of the coordinator and other key actors for dealing with political processes and substantive issues; - The ends-in-view for delivery systems toward which the coordinator and other key actors work as they address themselves to particular issues; and - the basic strategies of systems transformation with which the coordinator and other key actors operate. The fact that attributes are shared means that both a normative and an analytical framework can be developed for examining systems trans- formation, cutting across all regions. The fact that these are so general and can be arranged in many patterns and with varying emphasis means that the character of each regional program has to be unique -- the starting conditions of the region, the array of resources, the problems to be attacked, the level of development, the regional strategy -- there may be as many of these as there are regions. From the point of view of evaluation, therefore, the content of regional programs should be expected to be different. There is no "model" of a regional program that is applicable to all regions, although a conceptual framework which will allow assessment of diverse regional models can certainly be developed. Evaluation must not only take into account this regional diversity, but it must also take into account the fact that regional programs are in critical ways open-ended, with particular, but con- stantly changing !!ends-in-view." V-15 Arthur D Little- Inc Regional programs undertake systems transformation by engaging the emerging issues of medical care in the region. These are only partly, if at all, within the coordinator's control; to be effective, he must use them and build on them. Evaluation must take account of the open- ended or existential character of regional activity; except within a very broad range, it cannot second-guess the issues to be encountered in a particular region at a particular time; and it must not impose on the region a model of sequential activities independent of the issues of medical care which in fact arise. C. EVALUATION AGENDA -- A CONCEPTUAL F ORK 1. Systems Transformation in RNP This section outlines a generic answer to the question: "What is to be evaluated in RNP?" It is based on two assumptions: (1) that the objective of RMP is to bring about changes in health status and health care by serving in the role of broker of voluntary cooperative arrange- ments and as a facilitative change agent; and (2) that RNP is expected to place emphasis on, but not work exclusively in, problem areas related to the chosen categorical diseases. In these terms, what is to be evaluated depends on two sets of considerations: (1) Issues of substance; i.e., what should be changed, how can change be accomplished, and to what ends; and (2) "Meta-criteria"* which concern the processes by which change can be brought about: the skills used to stimulate and guide (facilitate) the process. This section enumerates, classifies, and arranges these issues in terms of systems transformation; accordingly, it is a summary outline -- in skeletal form -- of evaluative processes which are appropriate to systems transformation brought about largely through voluntary means. The under- lying operational processes that have to be evaluated have already been described in more detail in the Chapters on Regionalization and Facilitation. Thus we must determine what contingencies the evaluator (central and regional coordinator) should heed as he seeks to assess the progress of regional programs in effecting transformation of the system of medical care. Regions will differ as to the particular goals they select for changes in health care, the particular strategies they employ, and the criteria they use for assessing changes in quality of care, and access to care and health. In this context, we are concerned with criteria and related questions which allow assessment of the program of regional systems transformation without "second-guessing" the particular content *Literally criteria on criteria. V-16 Arthur D little, Inc of regional answers to these questions. From the perspective Of systems transformation, both the "substance" and its consequences must be evalu- ated. But, until discovering and evaluating in discovered terms what has been attempted and its context, the evaluator is in no position to evaluate the content of FM activities per se intelligently. The initial elements to which the evaluator must address him- self are: 0 Starting conditions (what is to be changed?), o Ends-in-view (changed to what end?), and 9 Processes and techniques (how can change be accomplished?). Broad regional strategies'for systems transformation express directions for the process through which the region may be brought to move from its starting conditions (as they are conceived in a particular instance) to particular ends-in-view. Section C presents an overview of the evaluation agenda, and is potentially a guide for conducting regional evaluation (with empha- sis on learning aspects of evaluation). Finally, it is an outline of Section D, although the topics and questions are in slightly different order. 2. Evaluation Agenda a. Starting Conditions (Ulhat is to be Changed?) The evaluator must understand what is -- or was to have been changed. A given set of "starting conditions," which establishes a diagnosis, estimates the difficulties in the way, enumerates the actors, and tallies the resources, represents a form of baseline data describing what the evaluator perceives as useful when he commences. But "baseline data" as a term is commonly related to scientific methodology in which fairly rigorous distinctions-among independent variables, dependent variables, and specific constraints are attempted. Therefore, the term "starting conditions" has been used here to avoid any such rigorous connotations. What rM does is not a controlled scientific experiment, nor even clinical research; it is a social and political process, and it assists in conducting social and political processes. This forgives nobody of the responsi- bility of being accurate, rigorous, and specific in his descriptions. Nevertheless, social processes can take so many forms, can involve such varied elements, and can be stimulated (or slowed) in so many different ways that incorporating any notion of scientific precision in dealing with them is more misleading than helpful. What is useful is recognition that starting conditions have a profound influence on what it is possible to do in a social setting. In many ways, starting conditions determine what is to be evaluated and establish a basis V-17 Arthur D Little, Inc for Judging the accuracy and utility of the diagnosis, the quality of the planning strategy emerging from it, and the efficiency of the process of developing both description and strategy. Accordingly, the evaluator will have to inform himself of the starting conditions and, further, will have to compare his inter- pretation with the interpretations of others, notably those in charge of RMP activities being evaluated. (1) What is there: adequacy, accuracy, and actionable considera- tions of the interpretation made by people on the scene? What and who make up the health care scene? t At the level of health and health care, description and starting conditions, for example, will- include s Patterns of health "outcomes," a Patterns of access to delivered care, o Quality of care delivered, o Configurations of resources for delivering care, and o Preferences of consumers for how care is delivered. At the level of organizational and political relations among ele- ments of the health care system, description of starting conditions will include: o Regional identity (how the region defines itself), o' Patterns of inclusion or exclusion of particular geographic and institutional elements of the health care system, o Patterns of centralization and decentralization of relationships and activities, o Linkages, o Conflicts, in being or suppressed, and a Health issues. Each of these themes lends itself to related questions from the point-of-view of evaluation. Each,, in turn, may come to be the basis on which ends-in-view are grounded, and processes and techniques chosen. V-18 Arthur D Little, Inc (2) How RMP has dealt with the need to case the region: the efficiency with which the description is compiled, the degree to which compiling the description constructively involves other people in the activities of RMP, and the quality of judgments they have made about how much of the description to publish and how to publish it. b. Ends-in-View (Change to What End?) The evaluator will have to test the validity, credibility, appropriateness, feasibility, and significance of the objectives' I that will or should emerge as RMP develops. These "ends-in-viewit are the specific rearrangements sought in systems transformation; and they, too, have many qualities that are subject to evaluation. The emphasis, again, is, first, to discover what attempt has been made to identify these-qualities, and to deal with them. Evaluation of specific content makes sense only after it is clear and more or less agreed what had been attempted, and the context for attempting it: (1) Their responsiveness to starting conditions: explicitness, completeness of response. Do they make sense to people in the health care system in the region? (2) Their relationship to available health care system models (hospital-based, community-based, federally controlled, highly decentralized, privately controlled, and so forth). Is the degree of relationship between chosen ends-in-view and available health care models appropriate to RMP operation in the given region? (3) Their responsiveness to accepted social values and issues emergent in health care in the region or the nation: specificity, actionability, realism, achievability. (4) Their appropriateness, significance, sensitivity, and explicit- ness on the several levels of change on which RMP operates: (a) Changes in the process of planning and interaction within the health care system, (b) Changes in quality and configuration of care-providing resources, (c) Changes in character of care delivered, quantity of care available, and people's access to care, and (d) Changes in people's health. Again, the evaluator needs to reach a judgment on what he thinks of how well RMP has elucidated the ends in view: (5) Their thoroughness, breadth, imaginativeness, practicality, and acceptability. What-is RNP attempting to do in the course of establishing ends-in-view and how well have they gone about doing it? V-19 Arthur D Little- Inc (6) Their apparent impact on people in the region: do the stated ends-in-view serve to stimulate local people into con- structive actions or into opposition or lethargy? We emphasize that ends-in-view are literally just that: objectives so specifically connected with activities in process that their feasibility, and even their meaning, is partly defined by the pro- cesses and starting conditions themselves. They are close enough and concrete enough to be visible ("viewed"). Reducing mortality from lung cancer can be an objective, but it becomes an end-in-view when it is clear what is being done to accomplish it, how the process can be expected to develop, why it is expected to succeed, a'nd what sense it makes in terms of any (or some) larger view of the health care system. c. Processes (How Can Change Be Accomplished?) (1) Strategy How well is strategy worked out? (a:) Does a strategy e)cist? How explicit is it? (b) How well does it link starting conditions, ends-in-view, and resources available for systems transformation (appro- priateness, sensitivity, feasibility)? 9 The processes chosen, as compared to the processes actually available (regionalization, linking, facilitation). o The tactics chosen (projects, feasibility studies, symposia, task forces, merger-facilitations, network building) as compared to the tactics actually available. (c) Priorities, sequencing, and how these relate to starting conditions, especially including gaps in health care coverage: o What parts of the region must be taken into account? o What issues of quality of care and access to care must be confronted, and for which user groups? 9 Which key actors and institutions from the medical or health care system have to be taken into account? o Are conflicting forces being taken into account, both in terms of timely avoidance and timely confrontation? 9 What sequence of actions best fits the strategy (i.e., is genuinely consonant with starting conditions, processes used, ends-in-view)? V-20 Arthur D Little, Inc, 9 What priorities among actions are implied by the resolution of the foregoing questions? What process is used to validate and gain acceptance of the priorities? (d) What strategic options are actually open, seem to be open, and have been considered (negotiation, unlocking the system (collaboration), master planning, sanctioning, combination, and so forth)? (e) What themes of system transformation are incorporated 'r into the strategy (centralization vs. decentralization, regional identity, bilateral linkages, confrontation of conflicts)? (f) What process has been used to settle on a strategy, how was the process, and what has the result been in terms of conflict, cooperation and support? (2) How well has the strategy been carried out? (a) -Directedness, Focus, Speed, and Flexibility The evaluator must decide if the process is moving fast enough, in the direction intended, and sensitively in terms of developing changes and emerging issues: o Stages of Development Are attempts to spread understanding of the starting conditions leading to involvement? Is "involvement" leading to concrete planning that is appropriately sensitive to the need for involvement? Is "planning" becoming implementation? Is implementation uncovering new ends-in-view or corroborating the validity of earlier ones internal to RMP? Have changing outside conditions or issues been reflected in the work plan? In addition to noticing the general thrust of what is happening, the evaluator needs to decide where specific activities fit in terms of "stages": e.g., should a coronary care unit training program be judged primarily in terms of its success in involving people, or in generating planning data, or in satisfying an end-in-view duly-validated for implementation, or what? V-21 Arthur 1) Little, Inc * Speed of Movement, judged against starting conditions, and stated expectations. (b) Results Observable on Appropriate Levels of Change o Changes in attitude toward system transformation. The skill with which techniques are carried out (effectiveness, efficiency, risk minimization, public relations, effects on those directly involved), 9 Changes in the planning and interaction process within the health care system, T o Changes in quality and configuration of care-providing resources, 9 Changes in character of care delivered, quantity of care available, and people's access to care, o Changes in people's health, o Changes in methods of judging quality of care (applies especially to the third and fourth items immediately preceding). In bringing about change that is aimed toward system transformation, there is always apossibility that the quality of medical care will be reduced, not improved. There is also a possibility that the means by which quality has previously been judged will come to seem inadequate. Evaluation capabilities and standards change; issues thought to bear on quality also change. Accordingly, the RMP evaluator has to look at what is really being taken as the basis for deciding what quality is. We have so far been explicit about quality considerations only on the level of conducting RNP as a social process. But because things done in the name of RMP can have a direct impact on the quality of medical care and should have at least indirect impact on it, the evaluator has to be explicit about quality changes with himself and with those whose work is evaluated: 9What impact on "qualityllhas resulted from a project or other action taken by RMP? oWhat method of judging quality is being used (meeting specifications, meeting performance standa rds, meeting user requirements, auditing)? (c) Accountabilities Finally, the evaluator must make some overall judgments about what has been done. Does he think the responsibilities of the V-22 Arthur D Little, Inc people on the scene have been faced? Does their way of dealing with the issues (laid out in this section) add up to good per- formance? Is he willing to tell them what he thinks? Is he skillful enough that they will be able to accept what he really is trying to communicate in his judgment? 3. Conclusion Evaluation finally is a judgment process -- obviously when management control is a foremost consideration; equally clearly when justification is the issue. But the evaluator's own skill in evaluation and communicating the process and results of evaluation is most at stake when the learning aspect of evaluation is on the line. Traditionally, the day of judgment has been a day for trembling, not a day for learning. Section D. offers more detail on the process appropriate to evaluation designed to assure learning as well as justification and management control, and is suffused with the notion that evaluation in RMP has to support mutual learning on the part of the evaluator and the evaluated. V-23 Arthur 1) Little, Inc D. ELEMENTS OF AN EVALUATION SYSTEM FOR RMP: QUESTIONS, PROCESSES, USES IN EVALUATION-IN-LEARNING 1. Introduction An evaluation syst em has to be accepted as practical and has to be conducted by real persons as part of a cycle of planning, action, and appraisal of results. The design for an evaluation system must include a conceptual framework as well as specifications for the process by which that framework is applied to operations. All too often evaluation is regarded as a simple instrument of administration dealing only in justifi- cation or control. In this commonly accepted view, an evaluation is seen either as (a) the means by which a subordinate "proves" that he has done what was expected of him, or (b) the means by which a manager satisfies himself that his subordinate has-done what was expected of him. The classic model for this view is the industrial engineering system of labor performance standards, which serves both justification and control. This system breaks down if the manager is unsure of what he expects or keeps changing his mind about it.@ When he finds himself in either of these positions, his interest shifts from testing whether what he is doing is going as we'll as he expected, to learning whether what he is doing continues to make sense in the light of what he now knows. Pressed by the anxiety of his own uncertainty, he becomes preoccupied with the need to know enough more so that his uncertainty will diminish. At this point, he discovers the value of learning the real-life results of what he is doing, to satisfy himself that his objective is real enough to permit him to test the results of his actions against it. So one function of evaluation (testing) is to permit him to learn about the validity (reality) of his objective in the first place, and so evaluation becomes just as much a means of appraising his objective as it does of appraising the effectiveness of his actions in meeting that objective. The foregoing shows that in any large organization there will .be a need for evaluation schemes to match several levels of uncertainty; the wise manager will recognize the different needs and respond to them differently. In such a universe, one of the problems of designing evalua- tion systems is to find a way of meeting all the needs, ranging from reasonably straightforward justification and control to outright learning from scratch, with schemes that are at least mutually compatible. The complexity and volatility of social interests puts an enor- mous burden of uncertainty on Federal social-improvement programs. And since the available resources are always more limited than what is needed to meet social problems, the President and the Congress have the deep responsi- bility of requiring justification of their activities from all Federal program managers. This extends to every organization level within the individual programs. RMP is such a program. V-24 Arthur 1) l,ittle, Inc a. Implications of Need for Justification and Control By its nature, RMP is an experimental program dealing in new concepts and new methods of approach to bring about improvements in a deep- rooted health care system.* Uncertainty is the name of the game in RMP. What are the implications of the unavoidable requirement for justification? In RMP there are essentially three levels of operation: o Project (regional) o Program (regional) 0 System (national). There are also, less universally and clearly, levels of subregional and interregional activities. At each level, there are the distinct though interconnected evaluative functions of justification, control, and learning. The formal requirements of justification and control lock RMP into certain evaluative activities, such as preparation of yearly budget submissions. On the following page we show a diagram of the key evaluative events that are related to justification. Some assumptions underlying this diagram include: o The trend toward decentralization of review, as evidenced by the introduction of "anniversary review" and strongly encouraged by the team" task force, will continue. 0 Project requests will be handled essentially at the regional level; central RNPS and NAC activity will increas- ,ingly be limited to regional program review. If anniversary review is not extended, some other form of "decentralization" will be. o RMPS places a yearly budget submission requirement on tile regions along with the tliree-year budgeted program and project review required under anniversary review. o Projects are on a yearly funding cycle, with regional reviews at annual intervals, although the annual cycle is variable according to region. The evaluative functions by which Central exercises control over the regions will be embedded in the annual and anniversary review funding cycles, or they will take place outside of them on an informal basis. Regional control over project activity will also be tied to the To call it a "nonsystem", as so many do, is to cover up its deep-rootedness. The fact is that the many, superficially unattached segments of the system are welded together into almost impregnable relationships. It is a nonsystem only in the sense that there is no single authority running it and that its more-or-less frozen relationships are conceptually indefensible. V-25 Arthurl) I.ittle, Inc. RMP JUSTIFICATION DIAGRAM Evaluative Events and Activities National Region-Core Project 0 Yearly Budget Cycle 0 Support to Central 0 Support to Region Departmental, BOB, Congressional Review of Past Activities, Requests for Funding 0 Review of Regional 0 Preparation of Yearly 0 Support to Region Funding Requests Refunding Requests Anniversary Review Site Visits 0 Project Proposal Submissions 0 Review of Project 0 Project Refunding Submissions Requests RAG, Trustees - Progress Reporting Subcommittees (of Both) Reviews Site Visits Progress Reporting V-26 funding/justification cycle, although here variations in practice among the regions will continue. Additional detailing on justification evalua- tion is the subject of the final section of this chapter. b. Implications for Evaluation-as-Learning But justification and control, to a considerable degree, is accom- plished by the existing "evaluation system" of the national RMP program, which operates effectively in the review cycle. Evaluation-as-Learning, the subject of this section, is also present but is incidental, casual, and informal. We advocate emphasizing it, regularizing its practice, and legalizing the behavior required to do it. It now appears in: o Those site visits in which somebody pushes the cvonver- sation well past the level of the "show and tell" stage; o Technical site visits when the "technical" problems are put into the most valid and complete political and social context available; o Discussions in which irrepressible members of the national and regional staffs really get down to cases; and o Those reportedly rare instances in which regional representatives learn, at least at second hand, what the National Advisory Council has to say when it discusses what is going on in a region, and what its members believe the reasons for the course chosen by the region really are -- and those still more rare instances in which these views are fed back to the region well and soon enough to permit a sensible and direct dialogue to develop. With respect to the learning functions, both the organizational context for evaluation laid out in the introductory section of this chapter and the-specific description of RMP as engaged in systems transforma- tion put requirements on the evaluation system: o Evaluation should be a two-way process so that both systems rationales (program and project definitions, objectives and theories) and systems activities may modify one another; o The evaluative process must detect discrepancies between systems rationales and discovered systems, and tactics for responding to those discrepancies; o Project and program goals shift over time. That is often a sign of progress, and the evaluative process should help discover whether it is and, in appropriate cases, both reflect and encour- age it. This is another way of saying that evaluative activity at this level should be an integral element of planning (of program or project) rather than an audit. V-27 Arthur 1) Little, Inc. These requirements suggest the form of dialogue -- a continuing' process of inquiry in which two or more parties both raise and respond to questions. From the point of view of the learning function, then, the problem of designing evaluation systems is the problem of designing dialogues. Dialogues are relevant at several levels; but, given the importance of the regional program as a un-@t, the dialogue of greatest importance holds between Central and the Regions.. 2. Central-Regional Dialogue Why a dialogue? The dialogue allows questions, the purpose of which is both to elicit information and to influence future behavior, to flow in both directions between RMP-Central and the regional coordinator. The dialogue is inherently open-ended. It allows for the regional coordinator and his own discovered system, and for modifications of his systems rationale in response to those discoveries. In this section, we list guidelines or criteria for the kinds of questions to be raised in such a dialogue and, in some instances, One test of the success of a dialogue is illustrations of responses. that, on its basis, both coordinator and Central are enabled to form continuing, grounded judgments of regional program performance. A second test is that, as a byproduct of the dialogue, the coordinator becomes more proficient at designing and carrying out the process of systems transformation. A third test is that the national staff actually is enabled to create and develop progressively better "systems rationales". Guidelines for questions grow from the criteria established as regional programs are developed to address various levels of change (health of people, access to and quality of care, institutional config- urations, and planning processes). The guidelines are also based on a view of evaluation which identifies patterns of systems transformation in terms of stages of development: 0 Involvement (getting started, casing the region); o Planning and goal clarification (discovering feasible processes and choosing and testing specific ends in view); o Implementation (bringing about planned changes, and evaluating what happens in such a way as to permit feedback from the evaluation to generate, or influence, a new cycle of involvement and planning and implementa- tion. The main reason for identifying successive stages is to orient the evaluator. For example, a project in the involvement stage may legit- imately have involvement as a temporarily paramount objective. If it achieves participation and builds commitment., it may be a great success V-28 Arthur 1) Little, Inc When well explored and outlined in the dialogues, the diagnosis includes the data crucial to working out strategies of systems transformation, both those which define health issues and health needs and those which define the organizational and political character of the health care system: a What is the character of the principal health problems of the region? What is their distribution?. o What is the character of the present configuration of health care facilities and resources? What is the nature of the health care delivery systems that are dominant in the region? o What are the patterns of access to care among the principal population gri@ps? The foregoing questions ate aimed at establishing "starting conditions" at the level of health, access to care, and configuration of care- providing resources. The next set of questions is aimed at an under- standing of the "political" forces that can be used or that must be dealt with in any strategy for systems transformation., 9 Who are the key actors and powers within the health care system of the region and how do they relate to the power structure and politics of the region as a whole? o What is the nature of the linkages, the relationships, the patterns of referral, and the tensions and conflicts among these key actors? o What do the central actors perceive as the major issues of health care for the region -- whether these are identi- fied in disease-specific terms, in terms of access to care, quality of care, or in terms of costs, manpower, patterns of dominance and distribution, or other facets of the health care system? Responses to these questions contribute to regional diagnoses which provide the material for designing strategies of systems transformation for the region. Although the descriptions of Northlands RMP and Greater Delaware Valley RMP presented in a separate volume contain many observations that go beyond a diagnostic description of starting conditions, they include and represent examples of what we mean. The "starting conditions" description provided by Dr. Winston Miller, Director of Northlands RMP required about four hours of very interesting discussion, many points of which we learned more about, or inquired into, in later parts of our work in Northlands. V-30 Arthur 1.) Little, Inc. conflict, not all of which felt constructive to the people involved. Other regions (we think mistakenly), created (or were saddled with) data projects that were isolated, or potentially politically explosive, because their underlying assumptions were out of consonance with the realities experienced by health professionals or grated on vulnera- bilities or boundaries carefully defended. We have heard little that suggests much use has been made of these data, and we question whether enough has been learned from the experience of gathering it. Among more recently organized RMPS, however, we know of at least one example -- Northeast Ohio -- in which specific perceptions about I'starting conditions" were used to generate specific questions. Answering these questions guided the data-gathering process. the data gathered were presented directly to the working groups whose observa- tions of starting conditions generated the questions in the first place. This kind of process, if6 the extent Northeast Ohio RMP has been able to carry it out in practice, seems to be a thoughtful attempt to apply what has been learned in earlier experience. In addition, it represents one very natural way for RMP to gather and use data that respond specifically to the need to establish a moving baseline on starting conditions. The process became a dialogue integral to the central processes of planning and evaluation within the region itself. In short, we advocate an evaluative criterion about baseline data: "It is not what you know, judged against some external standard, but how useful are the things you have deliberately tried to learn, toward doing what you are doing." This extensive commentary on data gathering is inserted here because it illustrates one crucial reason for evaluative dialogue. Without dialogue the evaluator cannot establish the significance of what the regional core staff knows or does not know. In a simple questionnaire- the evaluator can ask how many physicians and what kinds there are in Minnesota and where and how they practice, but cannot decide whether the Northlands RMP has to know these things, however elementary they may seem to him, without arranging for Northlands people to tell him what they are doing with the information in their own strategic context. This means that a detailed questionnaire (ofte n designed to be machinable) will not accomplish what we are discussing here, because the structure of the questionnaire conceals a series of assumptions that themselves establish -- i.e., impute a context -- for the region without offering an adequate way of testing the reality of the context. The structure of the dialogue has to be designed to elicit context, not to assume any one such possible'context. So the structure of the dialogue is "structure about structure" -- i.e., meta-structure. When the national staff asks a question that evokes the response: "We can get that information for you, but we'don't keep it that way," one of the meanings to be tested is whether the context for such a question has any reality in the region. This is implicitly recognized by a qualifying remark often used by RMP staffs in conveying such (essentially evaluative). questions: "We don't necessarily believe you V-32 Arthur 1) l,ittle, Inc. starting conditions mean, what the people in the region intend by the ways they formulate them, and where they all seem to be heading, and 2) There are indeed a wide variety of starting conditions to be discovered: o Northlands: Minnesota is a prosperous, relatively homogenous society. Good medicine is practiced there and the profession is in relatively good repute with the local political-social establishment. As yet medicine and the other health professions are facing only tentative questions about the "relevance" of where subspecialization and bigger-better hospitals are headed. But something very real is brewing in the state legislature's effort to force a "Family Practice" Department on the distinguished specialists of the University medical faculty. Additional intimations exist in the reluctance and opposition of the Academy of General Practice to the way the medical faculty had first planned to teach family medicine. Many competent, skilled, devoted people work in hospitals and other health care institutions all over the state, all of whom tend to emulate, or somehow react or respond to, the presence of the internationally famous institutions: -- the Mayo Clinic, the University, and the American Rehabilitation Foundation. There is an apparent shortage of manpower willing and able to perform health care services on the level of ordinary care for ordinary conditions. Town-gown issues are real, but because "gown" some- how includes "Roch(--ster" as well as "The U", and because "everybody" was trained at "The U", the issues take a special form. Centralization of Mayo and decentralization of the University complicates their association, whenever joint commitments are required or contemplated. Good acute care general hospitals are plentiful, and coming to view one another as competitive whether they are or not. @n are trying to become referral centers both in attracting large consulting staffs of specialists and offering many high technology services. Generally, the Establishment -- medical and non-medical exhibits a tough minded, "show me" conservatism, tempered by a very active consensus and willingness to try out credible ways of improving the situation (e.g., 40% of Minnesota private physicians have tried group practice, and they and their patients like it well enough to continue it). V-34 Arthur f) Little, Inc. that there is turn inward, their rationale being little that can happen "unless you own it." Thus the tendency is rather stronger than average to turn RMP and its training dollars to the enhance- ment of existing institutions and departments. Rivalry conditions all attempts to regionalize or otherwise bring about constructive associations between people in the somewhat depressed cities of northeastern Pennsylvania and the rich metropolis of Philadelphia. New Jersey Almost all New Jersey informants agree about one factor: maj'br forces that bear on medical affairs in New Jersey emanate from New York and Philadelpha, since many powerful M.D.'s living in New Jersey spend their professional lives in major institutions across either the Hudson or the Delaware. Although external forces appear strong, internal forces do not appear to be strongly organized. The two medical schools in the State (now one) appear vulnerable, still too young to have a great deal of momentum, and too poor to rise above political requirements that may be imposed, legit- imately or otherwise, by the State. The Academy of Medicine only 20 years ago began to spread its influence to South Jersey; it has had little more than typical success in conducting courses in continuing education for physicians -- North or South. The state health department and the various social action departments seem strong, relative to the medical agencies in New Jersey. The hospitals have grown quickly with population and the switch to hospital-oriented medicine. They are described as paying little attention to outsiders, and even the larger, stronger ones tend to be only moder- ately involved with other hospitals; despite hospital association activities, hospital mergers and shared services seem relatively rare. However, a shared drive toward asserting state identity, and coping with urban poverty problems favors RMP. Politicians can support RMP because they need the support of proponents of all aspects of state identity. The medical schools can use V-36 Arthur 1) l,ittle, Inc. health associations. The variation in emphasis, pattern, and priorities among issues is enormous, as is the "non-medical" pattern (geography, demography, economic conditions, and the like). Beyond these variations, still more dynamic features have to be noted, such as, who is in a position to exert leadership and express vision, who will or can respond to the opportunity (or need) to deal with what set of issues, and how skilled are the available leaders? The specific combination that exists in a given region has to be learned by a "discovery" process. It cannot be inferred from any two or three facts that can be recorded in a questionnaire. It will not be described to anyone believed to be an outsider until grounds appear for trusting his discretion, at least minimally. "Dialogue" is the most-natural process that suggests itself. Some of it is already practiced. More dialogue could be very useful; and more explicit "formal" use of the process and its results would be helpful in supporting the regions and in developing sensible, viable concepts ("systems rationale") for what is going on nationally at any given time. A "national systems rationale" should be an in.tegration-of what is going on in the regions, and in the environ- ment of interest (starting conditions). 3. Strategies a. Strategies of "Involvement" When the evaluator turns directly to the subject of program strategy, he cannot, of course, forget what he has learned about the starting conditions of the region. How the strategy chosen reflects the condi- tions found and grows from those conditions is one of the fundamentals to be evaluated. The basic question is "How have you formulated preliminary strategies for systems transformation?" 0 Through what process? What is the substance of the strategy developed so far? - Why this far and no further -- or why so far in this direction? Often, the best way of getting at these issues in the dialogue is through discussions of substance: o Where are the outstanding strengths and weaknesses among key agencies and actors in the medical care system? 9 What are the patterns of alliance and conflict and how are these changing? V-39 Arthur 1) I.,ittle, Inc The relevant questions directed toward testing and refining the preliminary strategies vary with the content of these strategies. But there are certain common themes, which appear in "involvement" phases as well as later on: the adequacy of means proposed to the problems identified, responses to elements currently omitted, questions of scale and timing, ideas about the building or cumula- tive effects of the strategy, and responses to the problems or constraints which seem to underlie the strategy. There is a further set of questions related to another aspect of the situation common to many regions. The systems rationale for RMP overall has been in the process of change. Many coordinators,and their collaborators began operations over the first two years on the assumption that RMP was primarily oriented only to the categorical diseases, and primarily through the devices of continuing education, dissemination of research, training, and demonstrations. Now, with a shift in view toward systems transformation, they find themselves working at systems transformation not from scratch, or from a start- ing diagnosis, but from a cluster of projects already underway, and in a situation of limited funds available for new projects. Their problem is to take new perspectives on what they have, to convert existing-projects where possible into elements of strategies for systems transformation, at the seme time as they begin to design new projects or new core staff activities. For these coordinators, the train is already running when regional diagnoses and preliminary strategies have to be developed, and the evaluative dialogue not only has to recognize this fact, but it should also seek to discover how to embrace, modify, isolate, or terminate these projects acceptably. Where an RMP has been seen as merely an assemblage of unconnected projects, new projects may still be the only really acceptable next step. But they can also be explicitly judged by criteria that test their relationship to a regional program for systems transformation. o Western X RMP has taken the position that it is a clearing house for projects; it solicits and processes applications from elements all over the region. It is, therefore, a conglomerate of projects. How can it have a program strategy for systems transformation or anything else? But there is the sense of need to involve the two counties curr'ently disengaged from the program. The preliminary strategy has impacted on the starting conditions in a way that permits, encourages, and partly specifies a revision in approach. One county, medically under the leadership of a strong physician, has no involvement in the RMP program and 250,000 people live there. The county consensus is that "Metropolis always wins, and that is where the money is. " In spite of its apparent role as a "clearinghouse for projects", the RNP in western X turns out to be operating on a strategy which says: "Get every major actor and every county active in MP." V-42 Arthur 1) Little, Inc o Have the processes making for inclusion, discussed earlier, extended beyond formal membership in RMP activities, to formu- lation of ends-in-view and strategies for achieving them? o How are priorities formulated? Are priority issues being con- fronted explicitly at all? By whom? Do priority considerations enter explicitly into the deliberations and interactions of elements of the medical care system, or are they handled by the coordinator or core staff alone, or, ostensibly, really left to Washington? If there are conflicts among elements judged to be crucial to the region -- for example, conflicts between major hospitals and medical schools, between town and gown, between professional providers and representatives of users -- are these conflicts allowed and encouraged to enter into the formu- lation of priorities? Does the coordinator intend to attempt to build clusters of these elements into working groups, through explicit confronta- tion of these questions? If he is not doing this, is it a matter of deliberate intent? Is he working -- temporarily, or as a matter of continuing strategy -- on a model of compartmen- -talization,in which conflicts over priorities and ends-in-view are not allowed to come up, except within limited subsets of elements? Is he "subregionalizing" in this sense? If so, does it make sense to do so? Is conflict in ends-in-view being handled as a matter of "dividing up the pie" among competing actors, or is there also an attempt to relate such judgments to shared judgments about the urgency of health issues, or about the usefulness of issues as ways into systems transformation in the region? 9 How appropriate, acceptable, and feasible are the strategies being developed for achieving the ends-in-view adopted? For example, An outreach center, as a way of involving a major hospital and medical school in the problems of an adjacent rural- ghetto? Who will make it work? Who wants it? A joint coronary care project as a way of encouraging collaboration and rationalization of planning among a set of community hospitals? What will make it transcent its original focus? Questions about such strategies will focus on a number of dimensions: Adequacy of scale of the "solution" to the "problem", Feasibility of the methods proposed, V-45 Arthur 1) 1-ittle, Inc. Appropriateness of the strategy to objectives on multiple levels of the activity (e.g., substantive health impact, as well as systems transformation of ends-in-view; clari- fication of ends-in-view as well as involvement), Appropriateness of the strategy to the constraints and problems perceived to be underlying the issue. One of the questions is that of "teeth". Is the issue one that will yield best, or at all, to voluntary involvement on the part of the key actors concerned? Or does it require some forms of sanction and compulsion? This is a question of ideology, strategy, and legislative mandate for RMP, as, well as of propriety; possibly some other agency is more appropriate. Where the focus is on learn-l@ng, attention will go not only to questions of this kind but to questions about the ways in which the development of strategies is handled: Is there evidence of the active consideration of alternative ways of achieving the same ends-in-view? Does the deliberation over strategies carry with it consid- eration of effectiveness of the strategy in relation to the costs of carrying it out, and consideration of the cost/effec- tlveness characteristics of alternative strategies? Are there timetables for accomplishment? How realistic are they? Has there been consideration of ways of determining over time how effective strategies are in achieving ends-in-view? Tests for their achievement? Where the focus is successfully placed on learning, the impact of such questions will not be to "grade" the strategies at this zone in time where emphasis is on the development of specific ends-in- view, but to influence their development positively by "accelerating" and "enriching". C. Strategies of Implementation The implementation of strategies toward ends-in-view may take the form of core staff activity, the conduct of specific RNP projects, or the activities of committees or ad hoc groups, under the aegis of RMP. -The ends-in-view and the strategy may be specific enough to lend them- selves to only one of these kinds of activity, and to a well-defined unit of implementation, or they may lend themselves to a widespread cluster of activities. For example: V-46 Arthur 1) 1-ittle, Inc. o What is the relation of the regional coordinator and his staff to the activity? If it is not their activity, do they have, in relation to it, a continuing monitoring, learning-evaluative contact which allows mutual modification of the ends-in-view and the strategies by which the attempt at implementation is being made? 0 How compartmentalized is the activity? Is it connected to analogous activities in the region, or to activities which are parts of the same program strategy, so that both learning and concerted action may occur, where appropriate? o What is the relationship of these processes of implementation to the overall strategies of systems change held by the coordinator and/or his collaborators? Has the coordinator attempted to be explicit about them? Has,,there been an effort to relate them to particular strategies for achieving particular ends-in-view? For example, to connect a particular activity as a feature of a timaster plan"; to identify a particular negotiation as part of an overall strategy which seeks to involve key actors in a process of negotiation over their interests and conflicts in relation to the system of medical care. Is the coordinator able to use the experience of particular activities to learn from or to influence his overall strategies of systems change? There is one side of the question of impact which should be treated separately here, because it involves the impact of the process of implementation, which can reflect both on the formulation of particular ends-in-view and on the region's capabilities for carrying out further systems transformation activities. This is the process through which the definition of accepted ends-in-view may shift. o The connections established and reinforced in a particular activity may form the groundwork for new kinds of collaboration, e.g., the joint planning of a coronary care unit which leads to joint planning of a range of common facilities; the diagnostic screening project in a county previously cut off from the medical system of the region, which leads to a series of boundary- - - crossings. Are these things happening? Are there attempts to make them happen? 9 Learning from an implementation process can lead to changes which facilitate new processes, e.g., the cumbersomeness of a process of review and monitoring can lead to simplifications which make it easier and more attractive for others to enter the orbit of RMP activity. o Processes of implementation can display or enable development of "role models" which influence the character of new activities undertaken, e.g., the impact of Mr. James Musser as broker-facilitator on other key actors in the North Carolina region, or of Mr. Paul Ward in California, e.g., the influence of the few emerging medical care V-48 Arthur I.ittle, Inc corporations in California on similar, varying approaches to medical corporations. Questions about impact of implementation, then, also have to be addressed to the impact of the process of implementation itself. 4. Development of the Cycle: Clarification and Reformulation of Ends-in-View Regional programs develop iteratively, if at all. Cycle succeeds cycle, each growing from, but still resembling, its predecessor. A regional program, seen as systems transformation, moves through its cycle: casing the region, planning, and implementing; and then through another cycle widening and deepening its rings of activity. The evaluative questions of any one phase continue,-to be relevant; only, new sets of questions are also relevant to established activities, and to other sets of activities. The process of bringing new elements into RMP, for example, continues even as the ends-in-view emerging from earlier processes of inclusion begin to be carried out. New relationships come to the fore as people and institutions, formerly central, are encouraged to give way, to share their former centrality with newcomers. The most relevant new questions help uncover the directions of change in the scope and purchase of the whole program as it moves through successive interactions of the process. These questions are of several kinds: 0 Is the process increasing its scope? Is it increasing in the overall volume of activity, as measured, by actors involved, dollars mobilized, number of separate activities undertaken? Is there a widening range of parties involved in interaction and negotiation? Is the level of aggregation of the parties increasing? For example, is the interaction beginning to involve clusters of community hospitals rather than individual-community hospitals? Is the level of aggregation also.decreasing? For example, are individual physicians as well as medical society representatives coming to be actively involved in a way that extends the scope of the program? Is there an increase in the number of health issues engaged? Is there an increase in the coverage of the region represented by those issues and by the ends-in-view and activities generated? Within each phase, the map of the issues confronted and their location in the region should reveal changes of the following kind: V-49 Arthur1) I.ittle, Inc. Regional Location I ssu es A At Phase 1 Issues Regional Location A A -A A Phase 2 V-50 o Is the process increasing in depth and intensity? Is there an increase over time in the perceived importance, urgency, and ambition of the issues engaged and the ends-in-view formulated? Is there an increase in the connectedness and "clout" brought to bear on the issues engaged? Is the level of aggregation of the parties decreasing? Are individual physicians as well as medical society representatives coming to be involved in a way that deepens the program? An example of the development of ends-in-view and strategies in a regional program as it begins to go through a succession of cycles is described below. a. The K Region Dr. P., the coordinator, came from a program of continuing education in the one large medical school, a program of continuing education for GPs which, by his own present view, was not too successful. lie began by seeing the creation of RMP as an opportunity to expand his own educational program, and obtained a planning grant to create K-RMP. He visited local medical societies over the region and, with them, set up a program involving tumor registry, coronary care units, and continuing education. The boundaries of the region were established based on the expressions of interest of the parties approached who attended the meeting. As the program began to expand, its emphasis shifted away from the categorical approach. The RAG, which began with 30 physicians, began to change composition to include laymen. In view of the relative weakness of other institutions, including the state health department, K-RMP moved toward a controlling position for health planning for the state. In the beginning work with individual physicians and community hospitals had been emphasized, with education viewed as the easiest and least threatening way of entering. At the same time, the core staff became involved in project writing for individual hospitals; K-RMP has now withdrawn from CCU programs, except for continuing education. However, a similar effort based on the earlier experience (establishing facilities, loaning equipment to communities which could not afford to buy it) is now being carried out for respiratory programs. Dr. P. now realizes that the provision of continuing education for physicians and others is not enough in his region, which is poor in physicians and clear in its referral patterns, and which has only one medical school and not much institutional rivalry. Instead, he must provide a system of care with appropriate facilities within which the fruits of education can be realized. V-51 Arthur 1) I.ittle, Inc. 5. Systems Transformation Criteria and Their Application The RMPS-regional dialogue, as we have outlined it above, if it is successful, serves to promote regional learning in the process of systems transformation. But it also serves to provide an ongoing basis for assessment of the effectiveness of systems transformation on the part of RMPS and regional coordinators. It can do so because of the criteria for systems transformation which have been implicit in the questions outlined above. These are criteria for the conduct of systems transformation. They are separate from criteria for the substantive impact of regional programs on health care (changes in access, changes in quality, changes in the health of people, etc.), and separate from the criteria used in processes of monitoring and control. These criteria are, in effect, itmeta" in relation to the-substantive criteria. The meta-criteria to be employed by an evaluator require that there be substantive criteria, that they be appropriate to the varying strategies and ends-in-view adopted by the coordinator, and that they be formulated and used in certain ways. But this level of approach does specify the content of substantive criteria. Meta-criteria answer the need for ways of assessing the development of regional programs while they are in the process of development, providing a basis for influencing their future development, and still remaining consistent with the diversity of regional situations. The variety of regional situations forecloses the possibility of applying closed, comprehensive models of health systems to regions as a way of judging regional progress in systems transformation. The meta-criteria, applied through dialogues of the kind illustrated above, provide ways of assessing the performance of regional programs conceived as processes of systems transformation. They are applicable both to particular stages of development in the short term (measured, for example, in weeks or months) and to the movement of the overall cycle of development (measured in years). What follow are illustrative statements of some of these criteria, which we suggest be abstracted from a desirable evaluative dialogue, together with some of the intermediate "test" questions through which they may be applied to particular situations. a. Evaluating the Process of Casing the Rezion The coordinator should be capable of articulating a regionaz diagnosis which is credibZe, and which provides the basis for the formuZation of directions of systems transformation. The regional diagnosis should reflect the dimensions listed earlier. It should be based on a strategy for gathering and assessing information -- for example, statistical studies, interviews with providers and users, judgments given by key qctors in group sessions, observation of the V-53 Arthur I) I.ittle, Inc workings of the health care system, or any combination of these examples. It need not rest on any particular strategy, but it must find ways of incorporating views and attitudes of key providers as well as users of the health care system of the region (see "baseline data"). Proponents of the diagnosis should be capable of meeting challenges to the accuracy or relevance of their analysis. But the analysis need be neither exhaustive nor entirely accurate. It is of greater importance that it be capable of shifting in response to a challenge and that there be, in the inquiry undertaken by the coordinator, a continual source of challenge to be met. In particular, it is important that judgments about major issues of health need, quality of dare, and access to care, facilities, manpower, cost of care, and the political and organizational structure of the health care system, all be subject to the continual test of the multiple perspectives of key actors in the health care system. Where important conflicts of perspective arise, they should be confronted explicitly and actively. Where they cannot be resolved, these conflicts of view themselves become issues for continuing work and inquiry. Based on the regional diagnosis, the coordinator should have Formulated preliminary directions of strategy which refZect defensibZe judaments about crucial, substantive issues o health f care, issues reZating to the political and organizational structure of the health care system, and key actors and initiators of innovation in the health care system. While the coordinator should be capable of arguing for these directions of movement, on the basis of the regional diagnosis, these preliminary views about strategy should remain developmental in two senses: (1) They should take account of the issues they do not address, and there should be some thought given to the means by which these other issues may come to be addressed; and (2), in addition, they should be responsive to changes in the regional diagnosis which come to light in the course of RMP activity. b. Evaluating Involvement Processes The coordinator should find ways of including actors and eze?nents of the region's medicaZ care system identified as key in the regional dicgnosis; where some of these cannot be included at the outset, the problems about their inclusion should be explicitly confronted and strategies deveZoped for overcoming these probZems over time. "Inclusion" may be indicated by participation in a range of RMP-related activities, such as involvement in RNP committees, project work, or ventures initiated or supported by RMP. The difference between significant and @o forma inclusion must be resolved by tests that vary from case to case, some.of which have been suggested earlier. V-54 Arthur 1) I.ittle, Inc Factors to be appraised include: o Whether there has (or has not) been a real attempt to arrange for specific people to be included in RMP. (Was the labor union representative really invited to RAG meetings? Did he feel invited? Was there anything for him to do?) o How well the attempt is related to the coordinator's sense of starting conditions and his strategy and objectives (which depends on having learned those things first). o How explicit the coordinator can be about who is not to be included, and under what circumstances those persons would or should be included. o How much the coordinator and core staff learn about the process of including people from the experience of doing it. (If they had it to do again, would they do it another way? Are they increasingly imaginative and increasingly direct in their approaches to people?) o The impact on others of the coordinator's attempts at including people (clumsy or skilled, relevant or irrelevant, useful or useless, well planned and well understood or otherwise). C. Evaluating the Planning Process and the Process of Establishing Ends-in-View From interactions with key actors, ends-in-view shouZd have been estabZished, and they must confront at Zeast some of the key issues earZier identified as cruciaZ in the region. On the ZeveZ of substantive heaZth care, they must confront at Zeast some of the constant heaZth probZem themes, or emergent issues in heaZth care. Again, the coordinator should have addressed himself to the ways in which RMP may move to fuller inclusion of issues in its ends-in-view. The process used by the coordinator to clarify and state ends-in-view should have been: An explicit process, with its own psychological, dynamic, bureaucratic pattern, for achieving consensus and commitment. Worked out explicitly in advance to allow ample opportunity for contending factions to agree -- or to decide they want to continue to disagree, which might preclude adoption of a particular end-in-view, but would increase the likelihood of real acceptance for those which have survived. Accepted as a legitimate process by most or all of those importantly involved in it. V-55 Arthur 1) l,ittle, Inc Major themes of RMP activity should be developed and stated, and they should not be merely a reflection of what is common to ongoing activities, but a source of guidance for the generation of new activities. Questions of priorities among ends-in-view should have been confronted, through a process in which key actors in the region work on their conflicting interests not only on the level of ownership of RMP resources-, but on the level of substantive health issues and strategies. The coordinator should have explicitly confronted the question of the extent to which he is trying to build key actors into a working group, capable of planning together and setting priorities, as against allowing them to function in compartmentalized groups whose activities become connected only through the coordinator himself. His decisions here should reflect his developing regional diagnosis and his overall views of strategies of systems charfie. Strategies should be Formulated for achieving ends-in-view and should be matched to the ends-in-view and the constraints of the regional situation. Projects.and core-staff activities should be understandable as facets of these strategies. The choice of strategy should be defensible with respect to scale, timing, and appropriateness of method to the particular situation. The process of developing strategies should reflect consideration of alternative ways of moving toward the same end-in-view, the costs and likely effects of various approaches, and tests for achievement of ends-in-view. The inquiry into strategies should show movement toward increasing specificity and precision, along these lines, over time. d. Evaluation of Implementation Processes The process of implementation should be characterized by involvement of impzementers in the selection of ends-in-view and strategies for achieving them, and by a relationship of coordinator or core staff to impzementers, which permits continuing mutual modification of strategy and ends-in-view and of implementing activity. At this point, RMPS criteria for systems transformation in the region take the form of meta-criteria for the evaluation processes carried out in the region. oWithout specifying evaluative criteria to be used in assessing the impact of implementation on any of the levels of change, RMPS should require that such criteria be developed and that they be appropriate to the ends-in-view and strategies adopted. oThese criteria should not be limited to programmatic criteria (e.g., how many nurses trained, or how many calls received?), but should be primed to assess any change in health outcomes and V-56 Arthur 1) 1-ittle, Inc access to delivered care. Review of the definitions, test methods, and measures appropriate to the end-in-view and strategy involved should be made. 9 With respect to the process of evaluation, the evaluative framework should have been developed collaboratively between the regional center and the implementing agency. There should be an openness to modification, through the process of evaluation, both of the implementing activity and of the original choice of ends-in-view and strategy. This openness should be evidenced in the demonstrated capacity of evaluative activity to influence the planning of the implementing process, and in the evolution of the concept of ends-in-view and strategy durir;g the course of implementation; and the frequency and pattern of contact between core staff and implementing agency should be such as to make that kind of mutual influence feasible. o The evaluative processes adopted by the coordinator and core staff should be conducive to learning across subregional boundaries, so that those engaged in analogous activities (continuing education for GPs, for example) can learn from one another's experience, and those whose activities are elements of a larger strategy can interact in the light of that strategy. 6. The Developmental Cycle It is not reasonable to set uniform standards for the periods of time within which regions should have reached certain levels of maturity in their developmental cycles, just as it is not reasonable to apply uniform standards across regions to the time periods within which the various stages of development should be completed. On both levels, the time intervals will vary with regional conditions. The key factors here are not so much the size of the region as its complexity, its internal connectedness or disconnectedness, the number of conflicting or disconnected elements within it, and the seriousness of their conflicts or isolation from one another. Elements that affect the speed of motion include: Simplicity of the politics of the medical care system; few elements to be connected; few conflicts to be resolved. Relative weakness of other elements of the system, permitting RMP to function from the beginning in a dominant or unusually significant health planning role; Relatively high degree of connectedness among elements of the medical care system. It may be possible to establish a typology of RMP regions in terms of their potential for movement, similarities in strategy, and characteristic V-57 Arthur [) I ittle, Inc types of activities chosen to carry out the RMP program. There are, for example, many efforts to stimulate collaboration among community hospitals through their joint involvement in some program of approach to categorical disease; to establish outreach arms of major medical centers; to reach isolated subregions through programs using parapro- fessionals, continuing education, and the secondary support of specialists. Regions and subregions differ as to the constraints they put in the way of these kinds of activity, but they, too, can be grouped in terms of the seriousness of those constraints. Such a typology would not be structured so much to permit judgments of the effectiveness of one region against another as to provide guidelines both for RMPS and for regional coordinators bn the rates of movement it is reasonable to expect in a given region and for a given kind of activity. However, we did not feel it sensible to produce a typology on the basis of starting conditions alone, because these are too varied, as we pointed out near the beginning of this section of the chapter. Judgments about a region's progress in systems transformation may be made on the basis of its ability to meet performance criteria, within any given stage of development; its rate of movement from stage to stage, given the constraints under which it is operating; and the level of scope, depth, and learning evidenced by its overall cycle of development. In point of fact, most of the RMP regions are, in our judgment, still primarily involved in the problems of including key elements of the medical care system in @ activity and in the formulation of preliminary directions of movement and strategies. In spite of the number of operational projects, most regions are only beginning the work of fitting projects into strategies for achieving specific ends-in-view. Of those we have visited, most* are only now at the stage where the formulation of themes of RMP activity and the confrontation of questions of priority among ends-in-view are becoming feasible tasks. 7. Prerequisites for a Process of Evaluation Capable of Emphasizing Learning There remain questions about the particular vehicles through which the national-regional dialogue we have outlined for fostering learning in relation to systems transformation may be brought to effective reality. o The two parties to the dialogue must begin with some commitment to, and understanding of, the goals and methods of this kind of evaluative process. The requirements here relate both to the theory of the *Some exceptions: Intermountain, North Carolina, and certain subregions in California, Georgia, and New Jersey. V-58 Arthur 1) l,ittle, Inc evaluative process and the role of the dialogue within it, and to the particular skills and techniques involved in carrying it out. 9Although we have used simple words like "Central" or "RMPS" and ficoordinator," the parties to the dialogue will be complex. On the regional side, the dialogue will be carried on by groups of varying kinds, depending on the makeup of those involved in carrying initiative at the regional level. In one region, it may be a listrong man" coordinator, his key assistants, and, from time to time, others whom he may wish to bring along either to involve or to educate them. In another region, it may be the team the coordinator has been trying to assemble out of core staff, certain RAG members, and certain key actors in the medical care system cvf the region. oOn the side of the national staff, there is a key requirement for continuity of involvement in the dialogue with the Region over long periods of time -- ideally, over the life of the Region's development under RMP. The requirement for continuity becomes particularly critical, given the diversity and open-endedness of regional approaches to systems transformation; it is only from an intimate knowledge of the content of earlier stages of development that Central can be effective in dialogue with the Region. But, given the realities of life in both central and regional bureaucracies, continuity of this kind is to be achieved not through one man but through small groups whose members overlap in the course of time. From Central's point of view, the small group permits the inclusion of the varieties of competence required to carry out effective dialogue with the Region -- competence to question and respond on issues of substantive medical care and on issues of systems transformation, and skills in the evaluative process of the dialogue itself. oThe central-regional dialogue will have to be distinguished from funding decisions and, concurrently, to move away from the usual mode of central-regional contact, in which the Region displays its wares for Central, and Central and the Region then engage in a game of attack and defense. For the central-regional relation to be solely or primarily in this mode prohibits learning, in the senses outlined above, and makes it difficult, if not impossible, for Central even to gain information about regional activities. On the other hand, the dialogue requires that the RMPS staff be capable of being tough with the Region, raising issues hard enough to be heard, and challenging the Region in the light of findings and commitments which emerge from the dialogue over time. To make these things feasible, the roles involved have to be modelled, and the tone for such a dialogue has to be set, and concurrently, the funding-justification process has to be set apart and formally distinguished from the.central-regional dialogue. The dialogue will V-59 Arthur f) little, Inc surely feed judgments about regional funding into RMPS, but it should be formally and operationally separate from the funding process. Will such a distinction be feasible, given the tendency of the Region to view Central as monolithic and the Region's knowled ge that funding decisions will be made by Central? This problem is comparable to the problem of the regional evaluator in establishing his "helping" role, in spite of the fact that his findings will be influential in project-funding decisions; indeed, the problem is fundamental to any process of good management in which the manager seeks both to facilitate learning and to exercise control. The feasibility of the effort will depend ultimately on the good faith that Central and the Region are able to establish with one another, and on the extent to which the dialogue is found to facilitate learning. oThe dialogue requires a certain frequency of contact between Central and regional groups. Based on the rate of movement in most regions, once-a-year is not often enough. Within the interval of a year, too much happens and too many decisions are made which lock the Region into patterns of activity. Frequency of contact should be determined by the time required for the coordinator to take significant steps, or for the regional situation to shift in signi- ficant ways that mark important milestones in the stages of systems transformation. Intervals are likely to vary over the course of the Region's cycle of development. For example, contacts might be established around key events such as the first formulation of regional diagnosis, the establishment of themes of RMP activities, and the first effort at establishing priorities for specific ends- in-view, or the first phase of experience in implementing a specific strategy. Within the range of frequency indicated by "oftener than once a year," there should be provision for flexibility in establishing contact. Opportunity for flexibility increases if a representative of Central and the regional coordinator can maintain contact during intervals between meetings of Central and regional groups. o The central-regional dialogue offers another perspective on the role and conduct of regional site visits, and on the proposed process of anniversary review. The central-regional dialogue could become the main function of the site visit. The site visit team would then become Central's party to the dialogue. Such a concept would answer to some of the problems currently reflected in regional and central reactions to the conduct of site visits -- for example, the pattern of regional display and of attack-and-defense which make it difficult or impossible to find out what is really happening in the Region; lack of continuity in the site- visit team; lack of feedback to the Region;.or inability of the site- visit team to respond to the Region by clarifying or modifying Central's lisignals." There are also significant potentials of the site visit as a vehicle which the central-regional dialogue may help to tap: the opportunity for on-site contact with regional actors and agencies, V-60 Arthur D 1-ittle, Inc and the presence in the Region of persons regarded as peers by many of those undertaking regional activities. There is the further issue of the manpower requirements RMPS would experience if it took the conduct of central-regional dialogues with all of its regions more seriously. The site-visit team concept, in which outsiders are mobilized along with Central's personnel, could provide a crucial extension of Central's staff. But the concept would also require intensive efforts at internal training and team- building for the site-visit teams. With respect to the anniversary review, that event would have a very different significance if it were to function as the Yearly culmination of central-regional dialogue, rather than as an isolated contact which tends to appear, whatever the intent, as a funding-justification process. The site-visit team would then play a critical role in the Anniversary Review process, and the results of earlier phases of the central-regional dialogue would then provide the basis for the inquiry conducted and the judgments made in the course of the "anniversary review." V-61 Arthur 1) l.ittle, Inc E. JUSTIFICATION 1. Definition Justification encompasses a series of answers to deceptively simple questions: "Did you spend the money the way you said you would? Did you get something worthwhile for the money? How worthwhile was it? What do you want us to do now and why is that a good idea?" Justification can be considered as looking both back@ard and forward. When it is retrospective, it serves accounting-like functions. When it is prospective, it serves budgeting and planning functions. Justification is only as useful"as it is credible. Its credibility depends on the kinds of details it presents. This section presents our views on how to select details that will supply a credible Justification, given the nature of RMP. Can RMP be justified in terms of its impact on the health of people? If one tries to put the Regional Medical Program to this test, Justification becomes a series of pallid excuses. It is not at the level of people's health that RMP has so far had its major impact. It really is too soon to expect this kind of result from a program, the expenditures of which have only recently surpassed 0.1% of the total national health care budget. Furthermore, expenditures and forces affecting the health of people, in fact, go far beyond even the $60-odd billion attributed to Health Care. Given the multiple causes always operative where RMP is attempting to accomplish anything, identifying its impact is well recognized to be very difficult, especially on a region-wide basis. RMP is never alone in any field; that is, there is always a sense in which things are not considered the responsibility of RMP -- e.g., gathering really good and detailed baseline health statistics on populations. There are no agreed standards for guaging change in the health of people in any case, and most particularly not at the level of regions or other large populations. Nevertheless, there have been attempts made to provide this-kind of Justification. Those that try to isolate the effect of one variable, acting in a very complex situation, like the North Carolina RMP study of continuing education and Tri-State RMP's effort to establish the impact of coronary care projects, face, but hardly overcome, the difficulty of multiple causation. Those that try to offer Justification on the basis of opinion of "students" or "patients" acted on by a project run into another difficulty. For example, in Northlands we found physicians in continuing education courses often changed their views of the course after the immediate effects had worn off. Which opinion was more valid? Post- and pretesting to determine knowledge gained by training also his its well-known limitations: how is the new knowledge used and what difference does it make? Sophisticated attempts to deal with some of these issues are underway in a few places.. In the Intermountain Region, for example, there V-62 Arthur D Little lrx- is an effort to establish and specify changes in both care delivery and patient outcomes in such a way that these changes may be correlated with the short-range results of training offered in continuing education programs. So far, however, general efforts to justify RMP by trying to relate its activities to changes in the health of patients or populations have been either very expensive or not very convincing. One could, of course, conclude that the results mean RMP cannot be justified; our conclusion, however, is that the excuses listed in the preceding paragraphs are perfectly valid and that the problem lies in the time required for thi@ particular approach to justification. The approach nonetheless is worthy of some continued support: (1) It is expected by many constituents, and (2) parameters measured and suggesting ill health can draw groups together for action. 2. Systems Rationale for RMP Justification Justification implies comparison with a standard. There has to be sbme official rationale deemed suitable as the basis for comparison. In the case of RMP, we have argued that the system against which it should be compared is one perpetually under change and perpetually subject to rediscovery. RMP is a decentralized program based on voluntary regional cooperation. Although it operates at many organizational levels and has. three main foci at national, regional, and project levels, respectively, the principal focus is the Region; and the Region is the principal focal point to be used in working out a suitable overall justification of RMP. Of course, the temptation does exist to try to justify RMP regions in terms of the project structure they represent, and there are numerous analogies between project and program justification. But such attempts can be very weak, tending to be either summary compilations of individual project descriptions or rather sparsely supported prq forma assertions that appropriate legal and administrative procedural requirements have been met in the course of developing the projects. These approaches can be developed and supported by illustration and detail to the point where they supply useful justification. A list of briefly described projects, arranged by disease category and cross-classified by geographical location or institutional affiliation, can provide useful "accounting" results. One can tell something about where and how the money is spent from such a presentation, and test whether gaps and inequities exist. This kind of presentation may also demonstrate something about the actual priorities of a region, if it can be presumed that where and how a region spends its money bears some positive relationship to what it thinks it important to do. These are factors well worth considering, and they are elements of justification, but they are far from complete, since they do not reveal much about why the work was undertaken or what has been accomplished in any directly significant terms. V-63 Arthur1) I.ittle, Inc Furthermore, there are a great many possible accounting matrices. It is usually not clear in advance which matrix will prove to be the most useful. It is rarely very clear, after the fact, what the numbers mean, except in terms of "rules of thumb" about cost. For example, the fact that 100 physicians were trained in a short course at a total cost of $10,000 tells something useful only to one who is experienced in short- course costing, so he can compare alternative ways of providing the same service in terms of the $100 unit cost. Treating an RMP regional program (or even the national program) as a collection of projects is understandable, given the role projects played in getting RMP going @n the first place. This approach, however, does not offer a very complete or very convincing justification if one looks for internal coherence. It is true that one could define "program" to mean the sum total of whatever is going on. It is true that justification of such a program is relatively simple if it can be limited to the demonstration that all the projects have been suitably approved by all concerned and that they are all good projects. And even if one is hopeful for more than this, well-articulated reports on viable, constructive projects are at least a very good beginning. What we suggest beyond this is three steps: (1) creating rationalizations about the existing activities in various patterns until a sense of their coherence or lack of coherence is developed and shared; (2) developing some clear ideas about next steps to fill gaps and to create still greater coherence, based on the best among the rationalizations arrived at; (3) using the results of working out (1) and (2) as impulsion to develop a more profound strategy than the original rationali- zations. This process should result in developing a program-based systems rationale. If a region has gone to the trouble of specifying its "ends-in- view," of working out strategies and priorities for achieving these ends-in- view, and has been reasonably explicit about describing the conditions found at the beginning of the effort, these materials can provide a framework for justifying any of the specific activities of the RNP, including projects. Justification then becomes a process for relating specific ends-in-view to specific activities. If, for example, New Jersey RMP is trying to do something about perceived deficiencies in health care in urban ghetto areas and has formalized an objective for improving health care to the urban dweller of low income, it is obviously a stronger justification for an RMP cancer project in New Jersey if it not only satisfies the letter of PL 89-239 by being in an approved categorical disease area and looks as though it might improve the care of cancer sufferers, but also fits into a regional health and health care improvement program for the disadvantaged as well. The question of whether to condemn and cut off a cancer project run with only average attention to the poor represents another side of the priorities problem. This decision would depend on three related factors: (1) the intensity of the regional and national priority to help the poor; (2) the extent to which all of the agreed priorities of the region were simultaneously taken into account by existing activities (Does the program reflect agreed priorities?); (3) any adverse effects on the quality and productiveness of the project by virtue of its being shifted to the care of poor people. V-64 Arthur 1) I ittle, Inc. 3. Basic RMP Justifications via Historical Accounts The simplest form in which to express starting conditions, processes chosen, and ends-in-view is a historical narrative, which can evolve and explain itself as it goes, creating gradually its own content. Shifts in objective emerge from factors that were present before they became operative or from forces impinging suddenly from outside. Both can be recognized. Both can be pointed out as they appear, and when they are first noticed, and when somebody begins to act on the basis of knowing them. The historical narrative in a sense develops its own responsive form as it does, and to the extent that it does so is significantly different from the "accounting form" described in the preceding paragraphs. The very structure of the narrative and the assumptions present in it incorporate a good many of the values and express or imply many of the standards against which evaluation can proceed. The "accounting form" can express some such values, such as those having to dcr with equity of care, inclusiveness of ideas, and priorities. Let us take an example from a regional evaluative description of program design: "We had a couple of screening projects ongoing, which were seriously underrunning projected expenses. We studied the problems of approaching these diseases, exconining the possibilities of prevention. with respect to stroke., we have one route to prevention -- screen7,ng for hypertension. We got our 'screeners' to concentrate on the high- incidence areas, and linked this activity to training programs for physicians. We tried to make a program out of ongoing projects, by matching our perceptions of current need and our capability for handling the issues." Even in this brief passage three conditions intervening in the life of an ongoing project become the justification for a shift in objective: (1) Project underrunning its expenditures; (2) Discovery of high-incidence areas in which "the one route to prevention" of stroke -- screening for hypertension -- might be useful; (3) The need to create a program out of a series of projects formerly separated. The real structure of the process is the simple logic of combining the discovery of high-incidence areas and project underrun. This establishes a basis to justify a shift in objective which, however, is fully explicable only when the demand to make programmatic sense out of pre-existing projects is added. The narrative forces transcendence of the original accounting classification (project by project in this instance) and suggests a substitute categorization by investment in areas of high disease incidence. V-65 Arthurl)l-ittle,lnc The narrative, of course, does not itself constitute a complete justification, but many, if not all, of the questions that need answering to provide a justification are implicit in the narrative and the logic revealed by the structure of the narrative. For example, what plausibility was there in this new attack on disease in high-incidence areas? Were the screening projects underrunning because it early became obvious that they would not pay out except in areas of high incidence? On what kind of analysis was the shift to screen for hypertension based? On quite another .level, what institutional effect was caused by linking the screening activities and training program for physicians? Wha@ effect was anticipated? What planning and decision process was required to shift and recombine the pre-existing projects? ir By the time appropriate questions have been formulated on all four levels of possible change (planning processes, configurations of health care institutions, access and quality of care delivered, and health of people) the framework of a quite complete justification emerges. The justification growing out of a historical description thus has advantages: oBoth the evaluator and his audience can assess the work being judged from a variety of perspectives. Data and hints are present or implicit with respect to each of these perspectives. oThe evaluation takes account of change on or between all the levels mentioned, understood in terms of the historical narrative, expressed as examples of inter-relationships of plans and decisions and budgets and decisions. It encourages both the evaluator and his audience to take account of all the kinds of change that were attempted or brought about by the activity, rather than concentrating on only one or two possibilities. History alone is dull unless it expresses the difficulties overcome by clever management of the program elements so as to show (1) success and (2) useful techniques. People will use the perspective they choose in any case. The evaluator cannot control how people look at things, though he can encourage them to use his own perspective. But most people are more comfortable if reassured that-they have the data available on the basis of which to revert to their own perspective as well. The justification document thus conveys the message: "Here is what went on; here are the points at which crucial changes were attempted; here is where the changes were detected and here is what we can say about the magnitude of these changes. As to the individuaZ importance of these changes, baZanced against the costs and risks involved_, that is a question for individuaZ judgment. From our perspective we value the activity as about so good. You are free to place your value on it as you wiZZ."* *For a much longer example, see the addendum to this chapter, "North Carolina Comprehensive Stroke Program."- The passage in the text preceding deals with stroke projects other than.the North Carolina example laid out in the addendum. V-66 Arthur 1) l,ittle, Inc Where systems transformation is more explicitly the function of RMP, the explicit formulation of specific ends-in-view against which to progress is even more important than in the preceding project example. Formulating the ends-in-view of a stroke-s,creening program in an area of high incidence might be relatively obvious as to quality of care, access to care, and changes in the health of people. But if changes in institu- tional configuration or changes in inter-institutional planning processes are contemplated, these have to be made a lot more explicit than they were in the example above. The stylie of justification we are describing does depend on the explicitness of the ends-in-view of the program. In one region, for example, what we call an end-in-view came to be defined in terms of relationships among 13 rural and s'@mi-rural hospitals. The regional coordinator and certain of his key staff and committee members decided to strive for increasing levels of collaboration, interaction, and rationalization of planning among these hosptials, with an eye to strengthening primary, back-up,.and long-term care. They designed a series of related activities -- brokerage, data-gathering, training, all built around the establishment and distribution of intensive coronary care facilities -- as a subprogram aimed at this end-in-view. Justification of starting conditions and of program impact then depended on answers to questions such as these: o How much collaborative planning takes place among the hospitals, particularly with respect to - Definition of areas to be served by the hospitals? Definition of needs for expanded capacity? Definition of requirements for special facilities? Definition of division of labor in provision of special facilities? Definition of potential for joint purchasing? o How much collaborative (as opposed to competitive) interaction takes place with respect to - Cross-referral of patients in response to over- or undercapacity? Differentiated purchase of specialized equipment, and differentiated hiring of specialized personnel to operate it? o What are the effects of collaborative planning and interaction on - Configuration of care-providing resources; that is, the presence of distributed and shared special resources? Change in patterns of access of potential users to the various categories of specialized care, as measured by time and cost? Change in patterns of cross-referral among hospitals, among physicians adjacent to hospitals? Cost incurred by the hospitals through investment in and maintenance of special facilities, in relation to new performance capacity? Again, though we have been very sketchy about presenting the narrative, its structure and the ends-in-view stated provide the essence for the justification and provide the basis on which the more specific questions are generated. V-67 Arthur 1) 1-ittle, Inc. It has to be observed, of course, that the acceptability of the justification would rest with the audience sharing some, or a good many, of the objectives laid out or more tacitly assumed. However, the audience could also accept justification on the ground that the objectives were self-sanctioned within the region. This is where the @o forma assertions mentioned earlier come in. For example: "An examination of rosters of committee members will not only confirm that their qualifications rank with the best in the field, but they represent the broad spectrum of health interest, resources, geographic areas, and socioeconomic groups within the Region." In the framework we are espousing, this kind of sanctioning" assertion has real justification impact only if the involve- ment of all these groups and experts has led to the formulation of specific ends-in-view, which are, in turn, asserted in the framework for justifica- tion. 4. Quality of Care and RMP Justification The closer RMP comes to systems transformation, the more important is its sensitivity to quality of care. In connection with justification, this level of change assumes a special importance. Justi- fication of changes either contemplated or accomplished in the health care delivery field can have little political viability, unless it gives assurances that quality does not deteriorate. If RMP is to facilitate systems transformation, it will have to devote special energies to quality of care." We believe that RMPS must insist that all regional activities fully justify themselves in this respect. RMP's special interest derives from its close relations with providers of medical care and its mandate to concern itself with improvement in quality of care. Further, insofar as it focuses on systems transformation and thereby emphasizes broader and more equitable access to care, it has a vested interest in maintaining existing quality of care. Quality, of course, is not an easy thing to measure and it can refer to quite different things. Quality can be measured by provider characteristics (for example, the medical school of graduation, residencies held, etc.) in-,terms of the activities undertaken (conformity to best- practice standards is a good example) or to patient outcomes (as self@ perceived, judged by experts, or quantitatively measured). The process of peer-ranking combines elements of all three of the foregoing approaches to quality measurement, and allows its practitioners to avoid having to be explicit about what they mean by quality and the standards and criteria underlying their judgments. All of these approaches are well established. Each has been used repeatedly in the Regional Medical Programs. None has worked out perfectly. All are subject to further experimentation and development. Each has its place. It should be obvious that RMP activities that could directly or indirectly affect the quality of care should be justified in advance in quality tems." RMP people should specify what method of determining V-68 Arthur 1) l,ittle, Inc quality is currently in vogue in the realm in which they intend to act. Regional Medical Programs and workers they sponsor should not be held to more rigorous standards of measurement of quality than are others already working in the field; because they are attempting to change things, they may be subject to quite unreasonable requirements. In its own interest RMP has to be prepared to improve the state-of-the-art of determining "quality of care." Fortunately, assessing quality of care and change in quality of care is central not only for RMP but for all efforts to improve the medical care system. RMP's special need in this regard is therefore self-' justifying. RNP's special role may well consist in facilitating the development and use of the several strategies outlined above: o Actually testing patient outcomes and proposing quality standards relating to them (thd@North Carolina Stroke Program is one of many examples); 0 Facilitating sanction processes undertaken by professionals (national -contracts on best practice through the voluntary associations); 9 Designing and applying methods for the precise description of types of care and their correlation with patterns of patient outcome (the project shared between the Minnesota Medical Society and the Northlands RMP on evaluation of hospital care by physicians being one example). RMPS is in an excellent position to facilitate learning both by encouraging careful and varied approaches to assessment in individual regions and by connecting those involved in such assessments. Furthermore, where the issue is one of establishing relationships between the several strategies of assessment -- an issue which turns out to be tantamount to the problem of establishing relationships between levels of change -- there is a special role for RMPS. When we are able to show the impact of RMP activity at one level of change but not at others, what can we legitimately assume about the relationships between levels of change? What can we assume, for example, about the relationships between "improvement of quality," defined in terms of change in characteristics of personnel, and improvement of quality defined in terms of change in the pattern of care-providing activities? Can we assume that "improvement in quality of care," as reflected in changes in medical activity, will be followed by "improvement of quality of care" as reflected in patient outcomes? Obviously, statements such as these will be indefensible at high levels of generality. In still more specific forms, it may not be feasible to support them through inquiry undertaken within a single region, but it may be one of the principal roles of RMP- Central to help to formulate an inquiry into the validity of such connecting assumptions. V-69 Arthur 1) l,ittle, Inc 5. Justification and Baseli-ne Data In the beginning, everybody associated with RMP collected data. Everyone knew there was not much good data available. But early efforts have given way to the realization that wholesale collections of epidemiological information, resource distribution, and the like do little more than provide a sense of relevant activity. They do not lay a firm basis either for program planning or for program Justification. Data -collections set in motion when program directions were unclear have resulted in assembling masses of data that tend to remain unused. By far the most effective approach to baseline data and its collection has been that in which the search for data serves to clarify the selection of specific ends-in-view, the development of program strategy, and aids in clarifying and choosing priorities. All these purposes are frequently best served by rather7 "quick and shallow" data-collection efforts, using as much previously compiled information as possible, including the method we attribute to Dr. Morris Chelsky, formerly of the Greater Delaware Valley RMP; i.e., deliberately relying on information gathered in other areas and applying it to one's own area by analogy when no reason to expect crucial differences could be shown. 6. Priorities The justification framework has to include some statement of priorities and a description of the process used for reaching those priorities. The best justified priorities are those in which people believe what they are saying, and can document their list by offering specific indications of need and capability. Referring again to an example cited earlier in this section: that the New Jersey RMP could have reached a priority statement in 1969 that accorded first priority to anything other than the improvement of health care to poor people in cities would have been unbelievable to a majority of the New Jersey RAG and would have been very difficult to establish. Whatever the method used, the process has to be the same; i.e., gradually gaining the support and commitment for a given set of- priorities from the people who are knowledgeable and involved in the regional program. Rankings of priorities can be elicited using a variety of methods. One example comes from Florida RMP, where Dr. Robert Ausman, the Associate Coordinator, has suggested using a health care profile in which the elements he defines in the health care system are ranked or valued on a number of scales, with scores derived by adding the values assigned to these attributes (see accompanying chart, page V-71.) There is nothing particularly sacred about this way of presenting the elements, nor in this particular set of attributes. Indeed, if this method is to be used, two or three combinations of elements and at-tributes might be tried, just to compare what they might yield, and to clarify a variety of possible relationships among various perspectives on health care. For example, the "attributes" v-70 Arthur 1) l.ittle, Inc HEALTH CARE PROFILE 0 t; 0 > 0 cn co CD -W m m u a) CD ATTRIBUTES L) > 0 0 0 u F- ELEMENTS Primary Care System* Private Practitioner4 4 3 3 5 4 3 4 2 32 Ambulatory Care 2 4 2 2 3 2 3 4 1 23 Preventive Services 0 2 2 1 0 5 5 0 0 15 Rehabilitative Services Emergency Services System Transportation Care Institutional Care Intensive Care Intermediate Care Long-Term Care Minimal Care Special Diagnostic Services Special Therapy Services Pharmacy Services TOTALS V-71 might respond to a selection of stated or emergent national priorities: e Addresses the problem of medical manpower, o Reaches special target g roups -- e.g., children, the poor, pregnant mothers, o Taps funds in addition to those of RMP, or o Engages certain key organizations in the region. Another obvious method is to draft a statement which rank-orders various propositions, or at least classifies them in a way that could be subject to rank-ordering. One such example: o Supply, distribution, and education of health manpower: - Education and training to develop and increase the supply of new categories o-L allied health personnel, e.g., training of medical assistants; - Demonstrations of the use of incentives and other innovations to encourage relocation or reactivation of personnel ... o Institutional cooperation: - Demonstration of methods of making the facilities of the university and major regional and sub-regional hospitals more accessible to other institutions, e.g., especially by exporting services and skills from medical centers to community hospitals; - Demonstration of methods of doing institutional planning with special consideration for avoiding duplication of high cost facilities, e.g., joint planning for radiotherapy facilities. (From a memorandum of the Program Committee of the Tri-State Regional Medical Program, Harold W. Keairnes, M.D., May 5, 1970.) A more subtle priority statement devised for justification would take account of the internal elements, both political and economic. Availability of people to do a job, ability to pay, and staff time availability are real constraints, and their reality can be explained, documented, and taken into account in developing as real a priority statement as feasible. There are traps and limitations in all methods. Merely stating priorities is an insufficient justification, since there is nothing in such a statement to prevent its being used to limit, rather than to organize, one's efforts. Take the statement, "We cannot fund a project to train coronary nurses because our high priority is to improve ghetto care" (or vice versa). This makes it possible to avoid doing anything at all when the choice is between,actually being allowed (by local power and federal money) to carry out a training program and interminable delays V-72 Arthur I) 1-ittle, Inc. in being funded to carry out exclusively ghetto-oriented projects. In our terms, the real justification for RMP is activity directed toward systems transformation. We endorse a remark made by an astute RMP coordinator, "Today's planning is tomorrow's boredom, whereas today's activity is tomorrow's partial failure and therefore tomorrow's fresh exciting planning for systems change!" Whatever the method used to arrive at them, priorities developed in this spirit will ease the burden of justification and keep the gap as small as practicable among evaluations conducted in all three modes (justification, management control, and learning). 7. Problems and Burdens of the Evaluator Regardless of what view one takes of evaluation, the evaluator has some very special burdens on his shoulders. In the case of RMP (as in other government programs in which formal evaluation exists as a result of Congressional or Budget Bureau pressure), the evaluator may be unkindly compared to a superfluous sponge, absorbing money which would otherwise be better spent in some other way. Alternatively, he can be considered an irrelevancy, gathering data that have no practical application. In terms of the formal system theory of the "rational manager," the evaluator, when viewed from the top, may be considered as a monitor whose function is to determine whether projects meet their goals and timetables, are doing what they are supposed to do, are staying within their costs, and are worth the cost. But the real, live regional evaluator has no source of data except the people who do the work, so he has to get along with them, whether they are project personnel, core staff, or contractors. He cannot afford to be too pushy. As for the program coordinator, his problem is to manage an amorphous and sprawling set of activities involving a great number of people, most of whom are not on his payroll or otherwise directly respon- sible to him. Prom his point of view, if the evaluator can help either to protect his flanks or to somehow manage the whole show, then he can see some use in evaluation. Otherwise, the program coordinator usually is not much interested. From the regional evaluator's point of view, he can expect little in the way of consistent guidance from either the program coordinator or RMPS. This must be -- and will continue to be -- the case because the issues on which he wants guidance are so often the very issues on which RMP management either must not commit itself or is striving without success to resolve. The consequences? First of all, the regional evaluator cannot go very far toward occupying an exclusive role as an auditor or monitor in the conventional sense. The swift evolution of objectives and the obligation to develop objectives and strategies on local levels combine to assure that the really pushy monitor can and will be provided with safely V-73 Arthur 1) Little, Inc irrelevant information that management can safely, though perhaps exasperatedly, disregard. Some well intentioned and actually skillful attempts at pre-testing and post-testing the knowledge of people in continuing education courses sponsored by RMP fall into this category. The effectiveness of the cognitive educational process is safely removed from most of the important issues about health care and its delivery; but it is an accurate measurement, as far as it goes, toward describing one set of changes made by a project, and thus can be safely presented as justification. The evaluator can also function as a rather passive information transmitter, attempting to gather information but placing no judgtaents on it. This is a rather passive role for an evaluator, but it would be consistent with the view of RMP as a clearinghouse for projects. Again, this kind of information is useful in a justification mode, but the evaluator in the position of merely transmitting information is probably going to keep busy enough on irrelevant details to be quite unaware of the real work of project and core staff personnel, on the one hand, and of the purposes of the program coordinator and the national staff, on the other hand. If he works very hard but is passive in style, he will become the unconscious advocate of the project evaluated and reduce his value to management, o'r he wilL become the dupe of management, untrusted and unable to obtain cooperation from the workers. Thus, he becomes an insulating layer between management and others. The evaluator can also function as an information exchanger. In this role, or in carrying out a facilitative intermediary role between management and ilworkers," the evaluator can become steadily more influential. By virtue of his own willingness to really try to help solve problems of management, on one hand, and operational problems, on the other, he can make himself trusted enough to gradually gain an understanding both of what is really going on and why. This last role is clearly the most appropriate to systems transformation and to the view of program evaluation outlined in this entire chapter. It may also provide a firmer base for independent evaluative judgment, information transmission, and intelligent partici- pation in the process of justification. An evaluator's assumption of the role of helper depends on good faith all around and considerable skill on his part and those in project, regional, and national management in working closely together. If the regional evaluator is to function both as an influencer of prospective planning and evaluation on the part of project staff as well as of project development in the direction of regional program, then he needs the support of the coordinator and his colleagues who must work toward several objectives. o They must display commitment to a program planning and evaluation process in which the program is built around the formulation of specific ends-in-view and strategies for achieving them, and they must mold projects, core staff activity, and committee work to these ends. V-74 Arthur 1) l,ittle, Iiic. o They must support the regional evaluator in his helper role, emitting clear signals with respect to priority themes or program activity, and requirements on the project planning and evaluation process. Funding decisions on the regional level must complement these signals. The project review process must become relatively simple and timely (unlike the tortuous, multi-step and often overwhelmingly negative processes common to some regions), and must reflect ends-in-view, themes, and directions also espoused by the regional evaluator in his influence on project development. o They must display willingness to learn -- that is, to modify program goals, themes, and strategies -- on the basis of thediscovered system of the region, as perceptions about it emerge from project and core staff activity. Otherwise, they encourage a distorted 'Ipropose-dispose" game between project and region. a. Central's Role .Analogous conditions apply to the role of RMPS. Given the fact that Central's view of the role and function of RMP will continue to change, there is a need for clear signals relating to the changes of direction as they occur; similarly, with respect to priority themes and targets of activity, as these are formulated at the national level. Central must take its function in the central-regional dialogue seriously, playing in relation to the regional coordinator a role very much like the regional evaluator's role in relation to project staff. Program funding decisions should reflect earlier signals as to program priorities. Central should impose performance criteria for the region's evaluative system on the regional coordinator rather than identify itself with particular evaluative instruments to be employed in all regions. Central should make explicit the basic assumptions -- particularly, the "connecting assumptions" mentioned earlier -- on which regional planning and evaluative activity depend, and make these the subjects of centrally supported inquiry. Central should take on a "network learning role," encouraging regions to share and make explicit their experience with types of projects and types of approaches to ends-in-view for systems transformation, just as the regional coordinator must do for analogous core and project activities in his region. Clearly, these requirements on the regional evaluator, coordinator, and Central are more easily stated than met. Program and project evaluation remains a tical process in which there are potential dangers to be experienced in making explicit the discovered system of the region, goals and ends-in-view, and strategies. There are also unavoidable zones of uncertainty, due to obscure and changing contexts beyond the control of participants, which stand in the way of clear "signals" and responses. Nevertheless, the demands of justification and learning may be more nearly V-75 Arthur1) l.it tie, Inc, met if project, regional, and national participants press to the limit their ability to meet these requirements. The basis of working together in the mode suggested is mutual trust and good faith. This means the ability to come to common judgments about those materials that need to be treated informally and kept for the time being in oral rather than written form, and those modes of presentation that are appropriate to more@or-less public scrutiny. What this amounts to is our judgment that the evaluator who is incapable of taking an active and constructive role as a part of a management team is also incapable of performing the kind of evaluative function that we are advocating. V-76 Arthur 1) I.ittle, Inc ADDENDUM 1 (Italicized type in the following Addendum indicates commentary inserted in the descriptive-evaluative narrative, to categorize what went on in the project described in terms of the view of evaluation presented in this report.) NORTH CAROLINA COMPREHENSIVE STROKE PROGRAM An RMP project can entail voluntary cooperative agreements across community and professional lines as well as among institutions. The North Carolina Comprehensive Stroke Program is such an arrangement, and one that is particularly interesting because of the light it throws on the process of disseminating medical knowledge. Because the end-in-view was to bring knowledge, technique, and competence already used in the medical center setting to rural areas remote from such competence, it won ready acceptance as a suitable categorical project within the RMP framework in early 1968. We treat it as an exconpZe of an RMP project that has lived and flourished long enough to be subject to several levels of evaluation: o As a project with internal, preconceived goals, intricately imbedded in "starting conditions," intricately responsive to "ends-in-view"; o As a project, characteristic of a number of RMP projects, significantly innovative as an administrative mechanism: a social and political invention achieving medical impact through these inventions and thus achieving the "diffusion of knowzedgell- o As an aspect of an emergent regional program, involved in systems transformation through an expZicit regionalizing strategy capable of producing changes in institutional relationships and patterns. A. STARTING CONDITIONS The stroke program was relatively early among funded projects because of an agreement among the three North Carolina medical schools to emphasize stroke projects at Bowman Gray. The three medical schools in North Carolina, as major RMP backers, made some attempts to "divide the turf" in an effort to minimize competition among the medical schools in RMP terms. But this was not easy to do in any very clear way. In the first place, part of the commitment that the University of North Carolina Medical School at Chapel Hill made in 1950 in arguing for its transformation from a two-year to a four-.year school was one of outreach. As the only state medical school in North Carolina, the Medical School of the University felt that this commitment necessitated and justified its presence in many, if not all, parts of the state. Furthermore, it was a large school and its graduates had long been dedicated to practice in North Carolina. Both V-77 Arthur 1) 1-ittle, Inc. Duke and Bowman Gray were willing and anxious to participate in the program of outreach, too. But the fact that there were three schools and that they were all somewhat different in their interests made a "categorical RMP" division of territory rather easy. In the early stages of RMP the University of North Carolina tended to specialize in heart diseases; Duke contended with cancer; and Bowman Gray emphasized stroke. This, of course, meant that priorities for preparing project applications in the several schools were to some extent predetermined; perhaps more importantly, it meant that each school had an incentive to bring in early projects in these categorical areas as agreed. The project under scrutiny here was prepared by and through members of the Department of Neurology at Bowman Gray. B. OTHER STARTING CONDITIONS The concept of the Com@tehensive Stroke Program arose partly from the experiences-of the Cerebrovascular Trainee Program at Bowman Gray (federally funded) which was designed to improve the skills of practicing physicians in neurological techniques and treatment of diseases. These physicians would spend two or more weeks in the Neurology Department during which time they would engage in didactic and practical exercises which included working as house officers, attending conferences, and making rounds. At the time of this report, approximately 15 physicians from 10 communities in North Carolina had participated in this program,* thus creating some additional new linkages between physicians in the Department of Neurology and community practitioners. The experience convinced the neurologists at Bowman Gray that effective continuing education could be accomplished in community outreach programs. But the training program was only one of the themes in the work of the Neurology Department at Bowman Gray that contributed to developing the stroke project. In 1964 the Department had established a mass screening study of cerebral vascular disease. In 1965 it initiated a stroke rehabilitation program under the joint sponsorship of the Forsyth County Heart Association and the Bowman Gray School of Medicine. In 1966 its Cerebral Vascular Research Unit was established to conduct clinical investigation of cerebral vascular diseases, around the theory that improved diagnosis and treatment would result from a coordinated multidisciplinary approach. The Department was already well connected with the American Heart Association and was becoming closely associated with the North Carolina Heart Association. Two members of the Department, Dr. James F. Toole, Chairman, and Dr. Richard Janeway, had contributed to the special national task force report, "The Medical Basis for Comprehensive Community Stroke Programs," published in June 1968. Many North Carolina stroke victims were not necessarily receiving the most appropriate after-care. Although accurate diagnosis and treatment immediately following stroke could be very important, their value would * As of June 1970, 25 physicians'from 12 of the 50 states, the Philippines, and Canada had also received training under this program. V-78 Arthur 1) 1-ittle, Inc often be diminished unless rehabilitation and long-term management were made available. Stroke victims were more often than necessary left in bed, unless they chose to rehabilitate themselves, learning for themselves how to live with whatever permanent residual handicap resulted from the attack. Furthermore, North Carolina is one of those southeastern states in which stroke appears to be much more prevalent than elsewhere in the United States, and is known to be relatively short on neurologists and others with special concern and training in stroke diagnosis and management. Those who had these special skills and interests were for the most part concentrated- in medical centers in larger cities physically close to only about a quarter of the state's population. The project sponsors commenced with rather clear and interlocking objectives in mind: 1. To identify stroke-prone patients and prevent disability by instituting appropriate prophylaxis; 2. To try to make better lives available to stroke victims through a process that would broaden and deepen their existing network of North Carolina practicing physicians in communities not necessarily close to Winston-Salem (where the total population was insufficient to provide them the number and variety of teaching cases they needed); and 3. To train interdisciplinary teams, based on local community hospitals, to provide the actual patient screening, management, and care. The project took account of: o Tensions and agreements inherent in the very foundations of RMP in North Carolina; oThe pre-existing reZationships, friendships, and resources of the Department of NeuroZogy; oThe growth and deveLopment plans of the Bowman Gray Department of NeuroZogy, including its research capabiZities and its interest in the clinical identification of the stroke-prone; oKnown stroke epidemioZogy; oExistence of the "sanctioned" guidelines in the "MedicaL Basis for Comprehensive Community Stroke Progrcons" report; and oThe "energy" inherent in outreach for - RMP, A reZativeZy new and energetic department in a medical school which saw itseZf as new to the area, The neurologists themselves, V-79 Arthur 1) 1-ittle, Inc Local people, The North Carolina Heart Association, because stroke was a relatively common fconiZy tragedy, about which people might be persuaded to want to do something. initially, as we shaZZ see, the project tended to overlook: o The amount of difficuZty inherent in stirring up or identifying sufficient community interest to make something happen ZocaZZy; o The amount of development necessary to perfect a workable acNinistrative modeZ; and tended to overemphasize: o The value of the relationships it had established with Local physicians through the CerebrovascuZar @ainee Program. On balance, as formally proposed, the project had developed into a sophisticated response to, and integration of, an impressive number of startinj conditions. It all added up to what could be (and evidently was) concidored a promising attempt to cope with a significant problem in an -ii@72ovat@ve way where the innovations were welZ within the scope of RMP. C. PROJECT PROCESS AND CONCEPT In collaboration with the faculty of the Neurology Department of the Bowman Gray School of Medicine (notably, Dr. James F. Toole, Chairman, and Dr. William McKinney), Miss Lydia Holley of the North Carolina Regional Medical Program and the School of Public Health of the University of North Carolina, and Dr. Warren V. Huber (Chief of Neurology, Central Office of the Veterans Administration, Washington, D.C.), the formal statement of the objective of the program was defined: "To provide the right care, in the right place, at the right time for all patients with a stroke or with the potential for a stroke." A central agency, the staff of the Comprehensive Stroke Program (B. Lionel Truscott, M.D., Director), was chosen to implement the program. The objective was further defined by the stroke staff in collaboration with the three medical centers, State Board of Public Health, and others; viz., to offer the actual or potential stroke patient the opportunity of receiving comprehensive and continued care in his community. This was to be done in each community by (1) training local stroke teams; (2) conducting in-service education for physicians and nurses; (3) providing guidelines of stroke management; and (4) developing a follow-up mechanism of treatment by a public health nurse. In most cases, stroke teams were expected to be drawn from within the normal definition of those included in the health care professions and V-80 Arthur 1) little,, Inc. paraprofessions -- physicians, nurses, and physical therapists, in particular, but in specific communities possibly augmented by people of different back- grounds and radically different kinds of access to the community and to rehabilitative processes. For example, members of the local fire department in the stroke management team could be of great assistance by maintaining an emergency or ambulance service. A skilled fireman or ambulance attendant could direct a stroke victim to the most appropriate medical facility. He could also, in some cases, administer useful interim or emergency care. Others who were considered potentially useful were homemakers trained t(> adapt the processes of normal, everyday living to suit the capabilities of someone handicapped by a stroke; e.g., to notice where ramps or wider doorways or signaling systems could help. Even beyond this, such a person, as part of a team, could mobilize carpentry or other trades or professions necessary to make changes in the house and grounds of a stroke victim. Often, small expenditures could make it easier for both the patient and his family to resume a kind of life as near as possible to what they had earlier led. The plan was to stimulate creation of the initial stroke management teams by calling on a physician who had been exposed to the work of the Neurology Department in the earlier program. These men would be asked to help identify suitable local candidates who could together make up such'a team and who gave strong indications of wanting to undergo the training needed to serve their communities in this capacity. When the teams were assembled, they were to be brought to the Comprehensive Stroke Program Center located in Winston-Salem for training in stroke diagnosis, available therapies, and rehabilitative management. These functional subjects were to be taught in different ways to members of the several different occupations involved, but each team was also going to be trained to work together. Team-building was designed into the curriculum with primary emphasis on practical experiences in devising and taking concrete steps in medical and rehabilitative management of stroke victims. The program seemed to be novel enough that its sponsors expected they could interest early graduates to come back to the medical center after six months or so for a second go-round. This plan would tend to improve the curriculum by making the experience of the early graduates available to the central staff and to further sharpen and improve the skills of the teams which could now be presumed to act as integrative agents-of stroke management. While many of the early ideas were validated, the hope of persuading trainees from the earlier program actually to become local stroke coordinators did not work out. Of 18 stroke coordinators appointed by May 1970, only one is such a trainee. But the general principle involved did prove to be a good one: a new working network is essentially composed of old friends. Before a Community Stroke Program could be launched, it was necessary to find local MDs who wanted to do something about stroke management in their communities. In most cases these men were well known to the Bowman Gray project sponsors and staff members or to the leaders of the North Carolina Heart Association (one of whose staff members has been seconded to be the Assistant Director of the Comprehensive Stroke Program). V-81 Arthur 1) Little, Inc Once local physician interest has been aroused, it has proved very easy to persuade other members of the local health care establishment to participate; indeed, in a number of cases, nurses, hospital adminis- trators, and others have been interested and willing even before any local physicians arose to assume the kind of leadership required to implement the program in a particular community. The process of team-building and community acceptance then leads to a crucial step: formation of an ad hoc steering committee, including all those individuals whose coordi@t d@efforts are essential -- physician, nurse, physical therapist, administrator, public health nurse, representatives of public health and welfare departments, nursin4 home and extended care facility personnel, social worker, medical records librarian, and the.like. When these people meet with project personnel from Bowman Gray to test the program in their community, the steering committee proceeds to set up the necessary local organization. Sub-committees on discharge planning, area resource development, public education, and in- service education are then appointed. The in-service education committee constitutes the stroke team itself hospital physician, nurse, physical therapist, public health nurse which attends an intensive four-day basic training course in diagnosis and treatment of stroke. On their return to the community, the stroke team conducts in-service education and patient-family education, using prepared materials provided by the stroke project itself. Each community also appoints a part-time executive secretary to coordinate continuing patient treatment; for example, seeing to it that the (hospital) discharge planning meeting is scheduled for each patient and that the proper professionals accumulate the information necessary to answer the questions that will and should be raised in this discharge conference. The secretary also ensures that hospitalization and follow-up data are gathered, recorded, and made available both locally and to project headquarters personnel in Winston-Salem. The project staff conducts the analysis of these data and does the follow-up necessary to find out what's going on in those cases, the reports of which seem unusual, and to suggest changes if this seems necessary. Standards are based on previously sanctioned practices. For example, Dr. Truscott, the Project Director, was careful to make sure that various medical practices and procedures involved were consistent with the guidelines in the "Medical Basis for Comprehensive Community Stroke Programs" report mentioned earlier. Experience indicated that systematic follow-up is very important. Although an annual workshop was included as a way of obtaining feedback from the local stroke teams, occasional visits and specific inquiries about unusual conditions are the only methods found successful so far in maintaining the standards of planning and practice that were built into the project. V-82 Arthur1) [.it tle, Inc. In May 1970, there were North Carolina Community Stroke Programs in 18 hospitals and 7 nursing homes, involving 15 counties with a population of over 725,000. Before beginning the program, each hospital prepares abstracts describing 10 stroke patients then or recently in its care. These abstracts furnish some basis for later comparisons. The project center has over 100 of these abstracts and will obtain more. Additional comparisons are possible because some stroke patients in the participating hospitals are not in the program, usually because they are in the care of non-participating physicians. The project center has access to data dew-ribing the conditions of over 100 of these patients, as well as the conditions of 50 additional patients who were originally in the program but have sinc'@ dropped out. Analysis of these data and comparison with data on participating patients is underway, and results are expected before the end of 1970. Forms have so far been received reporting data on the first 300 participating patients, many of whom have been in the program long enough to furnish considerable insight into its progress on the level of patient care. The training course has so far been offered to about 100 physicians, 300 registered nurses, 100 public health nurses, and 200 practical nurses, as well as some 200 others. The project faced and adapted well to the need for some initial changes in its mode of operation: o The work, management, and training of the stroke tecons themselves was not known in advance, but had to be developed from experience. There was a real attempt to learn from this experience; for ex=pze, by bringing earLy graduates back to the Center for two-way refresher and review sessions. o The concept of physician Leadership over the stroke tecon proved somewhat difficult to implement. WiZZing physicians proved hard to locate. Techniques for laying out the program to a suitable Local audience in a way that did not antagonize the Local physicians were relatively easy to deveZop, but impZied a willingness to wait for weeks or months while non-physician "proto-te=s" persuaded some Local physician that becoming coordinator would not overcommit him. o Project sponsors and managers were active and energetic in trying to deveZop suitable community contacts within the medicaZ profession, but apparently'were unable to be effective among compZete strangers. They deserve high marks, however, for imaginatively seeking out old friends and "working" their pre-existing networks of associates and acquaintances. o In all respects -6ested, the project personnel appear to have been sensitive to the political preconditions for making the new project go: -- The real initiative Locally had to be administrative, and was left to a hopef4ZZy active administrator: the stroke executive secretary. V-83 Arthur D Little, Inc. AZZ significant "sanctioning" questions were dealt with: physician Zeadershzp, best practice standards, community commitment. The project personnel were not themselves trying to furnish sanctions: they were on the line as educators, technicians., managers and consuztants; not as guarantors of the propriety of what was done. But these guarantees were expZic@tZy made accessible to all concerned in the process of Zaying out the project and gaining first national (Review Committee and NAC) and then community acceptance. The project anticipated the need for detailed evaluation by trying to gather appropriate data about patients treated, both for its own research purposes and ultimate justification and learning. Monitoring local prograu7s is touchy, but possible, through occasional consultations and the scrutiny of incoming data on individual patients. Monitoring is thought to be crucial by project leadership, since the number o participants is f large and team membership and background is diverse. (An on-site evaluation wouZd3 of course, also coZtect impressions 'rom local stroke teams on the value of these monitorial- consultative visits and other aspects of the program from their perspective.) D. INTERIkl ACCOMPLISHMENT8 Experience with the first stroke management teams and their patients suggests that this process is valuable. The first 150 patients were released from the hospital about four days earlier than was true on the average of other North Carolina stroke patients. It also seemed clear that these teams, even though they had been given very limited training, had become quite perceptive about diagnosing stroke-like conditions and sorting out patients. Although not all the physicians in the community hospitals have elected to participate in the stroke program, it is clear that the existence of the stroke team does influence other MDs either to request training for themselVE'S or to begin referring to the physicians in the stroke team or otherwise significantly utilizing the procedures thus made available, through the skills of the non-physician members of the team. (These speculations will presumably soon be reduced to numbers: how many physicians? how frequently does this sort of thing show up?) In several of the community hospitals involved in the project, discharge planning and rehabilitation follow-up has been applied to other diseases in addition to stroke. Obviously, there is nothing unique about discharge and rehabilitation planning for stroke, and it occurs rather naturally to people in the hospital setting that these techniques have broad utility, of which they are beginning to avail themselves. V-84 Arthur 1) l.ittle, Inc. Once in-service training has been completed the only significant local expense is the salary of the stroke executive secretary. It is probable that in most communities -- Dr. Truscott estimates at least half -- local resources would be used to make this project self-sustaining if RMP were no longer able to finance it. The expensive part of the program is the maintenance of the central project staff, which is more than a full-time operation for the Director* and is a full-time occupation for the Assistant Director, the nurse coordinator, and the physical therapist. This skilled staff handles organizing the local community, the in-service training, the follow-up, the evaluation of records generated, and the annual workshops. The total project cost is about $172,000 a year, an amount of money equalled by the estimated annual reduction in hospitalization costs for 1000 stroke victims in North Carolina. This is based on their hospital stays being reduced on the average by four days (as reported). The 1000 stroke victims represent two thirds of the estimated number of people who will suffer strokes in the counties already being covered by the first 18 community programs. The project is a vigorous attempt to change the quality of care available to stroke victims through a continuing education program designed to meet the conditions found in small community hospitals. Both the training process and the patient management techniques involved require an absolute minimum of physician time. Those who receive most of the training and do most of the work are either allied health professionals or the lay coordinator, the so-called stroke executive secretary. Quality of care is tested by collecting and analyzing data on each patient treated, as well as by testing the procedures used against the project standards. So the project is explicitly answering questions about quality of care being delivered by its participants. It is also sensitive to changes in access to care brought about by training (until 1970) nearly 900 persons. Project evaluation is less focused on discovering or encouraging changes in institutional processes, beyond those implied by creating a community stroke team. For example, the team may or may not insist on applying discharge planning to patients with conditions other than stroke, and whether they do or do not will not be questioned by project central staff. Dr. Truscott plans for the nature and support of the program to continue to L-volve: toward emphasis on prevention, an almost untouched subject in most communities, and eventually toward decreases in RMP-funding of the training of community stroke teams. He aspires to establishing additional teams; in fact, by the end of the third year of the project, he intends that 25% of the population of the state will have access to RMP-trained stroke teams; more would be desirable in his view. What's been learned so far is still partly to be derived or verified from analysis of the detailed patient and institutional data compiled. Significant epidemiological findings may well be forthcoming. But the main achievement of the project has been to discover how to transmit * A second physician to assist the director in his coordinating and teaching function is necessary. V-85 Arthur D 1-ittle, Inc therapeutic and patient management techniques to community teams, and how to obtain and use feedback from those teams on a mutually acceptable basis. The project works across town-gown and interprofessional barriers in such a way as to reduce their deleterious effects, or to erase them. It accommodates the complex attitudes and requirements that physicians place on continuing education. It depends on community initiative -- but can use its own initiative to identify and develop avenues for local leadership to express itself. It also meets local needs in a meaningful way: while sensitive to the requirements for and possibilities inherent in sophisticated diagnostic tests, its emphasis is on simple techniques not requiring heavy capital investment, and accessible to common-sense application. The project is an innovative exercise in administration that ties people together in a new pattern, and enables them to learn and apply new and useful skills. Interim evaluation is possible on several Levels: o The effect of the project on the "heazth of people" is not yet ascertainable, but many of the data needed to make educated judgments appear to be in hand. Tt will be known if delivered care is in some sense better if more patients survives if they survive as vegetables (or are to some specified extent rehabiZitated)3 if recurrence seems to be affected. Whether the existing data will answer all these questions immediately, or within a year or two, it is too early to tell. o The project has worked out a defensible rationale to justify itself in part even in the somewhat unlikely event it proves to be medically inefficacious: reduction in hospital bed-days appears to defray project costs. o There has been little explicit attention so far to stroke prevention, but if the project continues at its existing LeveZ. data are now available to enable easy access to the adult children of the stroke patients participating, to offer a population of pres@ed interest, for screenings and for possible preventive management. This information was obtained efficiently as a part of the initial patient abstract. One issue is whether this extension of the project would enable evaluation of the utility of attempts at "stroke prevention." v The project has not tried to push the use of its techniques (discharge planning, community outreach in rehabilitation) in non-stroke cases, but has been sensitive to the fact that this might change "access to care" -or "quaZity of care" for some patients by cataZyzing changes in institutional behavior. (We wonder if North CaroZina RMP core staff members have followed or evaluated this process.) o There is apparent intention in transcending town-gown issues through medical school outreach, and an interesting example of contemporary V-86 Arthur 1) l.ittle, Inc. continuing education of a health care team. Have adequate exploration and utilization of the learning that has been possible on this level really occurred? To -answer this question would require additional discussion with RMP core staff members as well as those interested in continuing education at Bowman Gray and project personnel. Further justification in this vein may well be required. 0 In a number of specific ways, the project has adapted itself to changing objectives and emergent conditions. How it reZates to the North CaroZina RMP proarwn at this stage we would investigate aZso, if we were actuaZZy'ev@Zuating project and progrcon, rather than trying to use the observations we have made of the project as an exconpZe of what we mean by evaluation on the "various ZeveZs-,"'and in terms of both learning and justification. (Note that our information on the North CaroZina program as a whole is for the most part nearly two years old. But our -information on the stroke project is a good deal more nearly current. Hence we will venture no judgment an how the project fits in with other current RMP activities in North CaroLina, although this is also a significant issue for evaluations V-87 Arthur D Little, Inc v VI. RELATIONSHIPS Arthur 1) Little, Inc. VI. RELATIONSHIPS: NATIONAL-REGIONAL, INTER-REGIONAL, AND EXTERNAL A. INTRODUCTION 1. General Purpose of the National Staff In the work statement which accompanies the Arthur D. Little, Inc.,- Organization for Social and Technical Innovation contract with the Regional Medical Programs Service, the contractor is enjoined to prepare a report on Regional Medical Programs (P@) as an experiment in "creative federalism," with "particular attention to the role of the national staff vis-a-vis the regional programs." The work statement also suggests that the role of the national staff is to stimulate action by serving as a resource responsive to regional needs. In our inquiries we have tested this conception and found it fully consistent with what is possible for the national staff and what is needed by Regional Medical Programs. This chapter explores what "being a resource" means and discusses other, alternative perspectives and roles for the national staff in its relations with the regions. in earlier chapters we have already said a great deal about existent or desirable national-regional relationships, especially in connection with evaluation processes embodying a "regional-central dialogue." Similar con- siderations apply to other'aspects of the relationships between RNPS and the regions, since in almost every respect these relationships are tinctured by the same considerations: RMP is "regionalized"; that is, the regions have been established, are organized, and are forthcoming to varying degrees. Program accountability and meaningful program integrity can exist at the regional level, but at the national level not so concretely. The national level RMP staff has little direct access to the health system; it must work through the regions or through other national organizations, associa- tions and programs in the health field. The national level RNP function is to facilitate the work of the regional organizations. The regional level function is to interact with the health care system, and through such interaction, facilitate system transformation. Though RNP is a grants program, it is also a change-oriented faci- litator of social and political processes. This role sharply limits the possibility of the national staff acting as a controlling agency. Because RMP is necessarily a program of learning and discovery, the headquarters staff is restricted in the amount of authoritative "teaching" it does and expertise it can impose safely and success-fully on regions, particularly with respect to those regional RNP activities which interact with the health system. "Helping" has to be the principal role of the national staff and, at that, "helping" carried out only with the consent of the "helped." Anything less than helping is likely to equip the national staff vi-i Arthur D f,ittle, Inc with too little information to be directly effective. Anything more than helping is likely to work against voluntary cooperative regionalization. Since 1968, the processes called "creative federalism" have been converted and absorbed into a different conceptual framework, explicitly billed as decentralization and the "new federalism." The concept of re- gionalized decentralization, accordingly seems fairly stable. Administra- tors and individuals of differing political persuasions have been or are pursuing the same general goal. Moreover, it is shared by some legislators, the most prominent of them perhaps being Senator Muskie, as well as by the Executive Branch. Regional Medical Programs, of course, have long since been decentralized to a very considerable degree, as the report of the FAST Task Force seems to Acknowledge.* The combination of two years of additional experience in the regions since our contract began and the continued pressure to decentralize decision-making in federal health programs makes clearer both the content and mode of the relationship between the national staff headquarters and the regions. The dilemmas inherent in "serving as a resource responsive to regional needs" are even more apparent. So are the requirements. TI ie job of the national staff is to formulate and transmit broad national priorities and other aspects of policy development and guidance to the regions, to pass clear signals about these priorities and policies to the regions, and to support the determination ofthe regions to formulate their own work and carry it out within the broad limits of the law. This implies national staff capability to aid the regional program coordinators and their core staffs in taking initiative to create program strategies con- sistent with national priorities, to encourage regional development of priorities, and through consultation to develop and carry out genuinely regional and regionalizing programs. It also implies national staff parti- cipation in identifying and specifying local "accountabilities" for per- formance, mistakes, and program development. This does not mean that the RMPS staff has to decide what these accountabilities are. The national staff must be assured in the course of its observations that there is a meaningful local acceptance of the concepts of accountability that is sub- @stantive as well as fiscal and that there is a shared local view on what those accountabilities mean. Genuine accountability lodged at the regional level is crucial for the further development of RMP. This concept will be discussed more definitively later in this chapter. 2. Three Perspectives on RNP Administration and Their Influence on the Relationships between RNPS and the Regions (RNP as Grants Program, Centrally Administered Agency, or Change Agent) Even when commonly acknowledged characteristics of RNP are assumed, there are still several different ways of looking at RNP which basically FAST Task Force Recommendations and Request for Implementation Plan on Regional Medical Programs, July 14, 1970, pp. 3-6 VI-2 Arthur D Little, Inc affect what one can expect from relationships between the national staff and local regions. We will assume localism, FM engagement in systems transformation, regional dependence on doing its internal regional work to a great extent through time donated by people otherwise employed, and the "categorical" restrictions and the proscriptions on "interference." Although these assumptions still allow immense latitude, profound f.iif- ferences will exist depending upon whether RMP is a "grants" program, a "federal" program, or a "change agent." This set of distinctions, we be- lieve, is an appropriate way to clarify some of the basic dilemmas that have heretofore conditioned the approach of the national staff to its own internal organization, its functions, and its external relationships. a. PIT as a Grants Program If RMP were purely a grants program, its basic national staff function would be to assemble financial information and administer the review process to effect equitable distribution of program funds. This view would be highly consistent with an "arms-length" relationship between the national and regional staffs; it would also be consistent with treating the local staffs (or shadow staffs in medical centers) as branch agencies in the grants process. While grants review and manage- ment considerations do make up one major aspect of RMP, to treat it as the basic organizing perspective on the work and structure of the national staff violates the realities of regional core staff engage- ment in many other activities in addition to grant stimulation, re- v.Tew, and processing. Past references to individual regions as "little NIHS" expressed the implication of RNP being a grants program in a rather well understood mode. Subsequent inability to deliver project money in appreciable quantity as well as the regional dis- covery that other kinds of activities were necessary and had pay-off associated with them have made continuing effects of viewing RMP as a grants program more noticeable at national than at regional levels. Recent developments, including the beginnings of an @'anniversary review" and the study of the FAST Task Force, point to a national role of h(@lping (and stimulating) the regions to develop their own program strategies. This role for the national staff would transcend the broadest interpretation of a "grant" program as usually understood. It would have to be based on appreciating the social and political realities in the regions in a way which is not relevant to the traditional NIH mode of project review, as well as performing procedural review as grants program administration must always do. The implication of this broader scope for RMPS-regional relation- ships is that the grants review function within RNPS has almost certainly been given too much importance in the past year or so. b. RNP as a Federal Program Let us turn to the stereo-type of the federally administered health VI-3 Arthur D Little, Inc program. There is a sense'in which RIT viewed as a grants program, as just summarized, is consistent with PIFP viewed as a nationally administered federal health program. This is the view thattreats the regional programs as branches in the grants review and grants management cycle. Driven to an extreme, this view would make the network of national and regional programs something vaguely analogous to the Post Office, in which priorities, standards, procedures, and policies are all centrally determined. What would happen under this theory would necessarily reflect these national standards and in- junctions. We do not advocate this view in the extreme form just statedland neither does anybody else. But some of the attempts by the national division staff we have observed to place regions in categories, to apply PPBS to the regions, and-to describe the structure of the regions as though all RAGs do (or should, or might) have identical roles tend in this direction, as if it were valid to view RMP as a centrally administered program. The irony is that such proposals are made even by people who do not believe in the centralization of RMP. Further revealing this view about federal leadership and central control is the assumption held by some that RNP is self-contained. Claims and accusations ma(le from this perspective tend to ignore the important reality (though admitting the form) of part-time loartici- pation in RMP by significant people in the regions who are very little under the influence of RMP. Limitation of local right to exert control could rather easily make it impossible for RIIP to accomplish anything at all, given its small size, limited funds, and lack of sanctions. Federal leadership and control assumes that the "essence" of RMP is known in advance of what RNP does, and neglects the profoundly positive aspect of the remark attributed to Dr. Robert Marston, then Director of Regional Medical Programs, on the approval of the first series of operational grants: "Yesterday I couldn't have told you what the Regional Medical Program was. Today I can tell you. It's what these four programs are set up to do." Had Marston believed that RMP could or should be centrally administered, his willingness to accept what the regions produced would have seemed misplaced We sympathize with the impatience felt by many --including many members of the national staff -- at the lack of directedness. But we have yet to see a plan for directing the program more tightly that does not "cut off its own nose" in the course of trying to make things more coherent. Indeed, the viable plans we know about seem to us to work on quite a different basis. Governing through guidelines pushes the national staff role toward being that of assuring coherence rather than enforcing control-and imparting direction. As we interpret the guidelines, they impart little positive direction to RNP. That is not their function. Similarly, vi-4 Arthur[)I ittle, Inc. our own proposals in the preceding chapter to promote influence on the program at "meta" levels, while they include other devices in addition to national guidelines, still are intended to work on the basis of interaction leading to regenerative feedback. Direction arises from the needs and activities of the regions, and is clari- fied and given substantial priority meaning in the course of the dialogue between national and regional levels. But the direction does not stem from national power sources. Feelings about health care delivery are too sensitive and interests, are too significantly vested for any plan for federal RMP leadership (read as domination) in local priority setting, let alone for Oethiled standard setting, or still more for specific decision-making to be viable. The classic bureaucratic tradition has some validity even in 1970, but we do not see how to make it work-in RMP, iintil bureaucracy transcends itself and becomes something different. Its proponents in RMP, whether commissioned officers in the Public Health Service or otherwise veterans of federal health programs, no doubt believe they have already transcended the limits of tradi- tional bureaucracy. C. RMP as a Change Agent The third role that can be seen for RNP is that of change agent. Anyone who has read this report through from its beginning is al- ready aware that we believe the Regional Medical Program has to be viewed from this third perspective, which partially embraces the other two views ("grant" program and "federally administered" program.) RMP does, indeed, require central administration, if for no other reason than to ensure the orderly and legal expendi- ture of the funds entrusted to it. It was set up from the begin- ning to enable these funds to flow to the regions primarily through the grants mechanism. But our main emphasis is on its special qualities as a program capable of charting and helping others to use paths toward systems transformation in health care service delivery, utilizing its categorical disease setting and its original thrust toward "continuing education," not only as some early proponents urged, but also within the breadth the law enables and emergent issues require. As in earlier chapters, and also in this one, our intention has been to analyze the relationships we know in terms of change-agentry and to extrapolate in the direction of how to make them still more compatible with this view. For some of the remaining long-service members of the national staff, our view may well entail or anticipate a pronounced change in their own perspectives. The definition of Me as a federal health grant program in particular carries with it a wealth of PHS and NIH precedents and experience, more of which, we are suggesting, need to be transcended than have yet been. But the idea that RMP is a federal program which must vilti- mately fall into a familiar mold to which all federal programs ultimately solidify is also one we have encountered and would resist. We disagree when it is expressed as a cynical concession to reality, when it is a hope for VI-5 Arthur D Little, Inc making a rather disorderly process shape up and start "producing" health improvement, or when it is a fear for the future of a program in which many of its participants have put a good deal of faith. We do not wish to b'e interpreted as denying the ability or the wisdom of-those who have adopted either of the views, even though we here suggest that those views should be superseded more completely than they have been to date. Different experience begets differing responses for dealing with the political process. But we believe that what security there is to be had from treating RMP like "another government health program" or "another grant program" is more than offset by the risk to the real value of RNP. If the risks involved in leaving this security behind are painful, better suffer pain than succomb to the passivity that leads to atrophy. Better a losing battle fought in the interest of coping with real problems than not to fight, even while pretending to fight, and thus to lose by.default. We are trying to clarify our view that the role and relationships of the national staff of RMP can and must be built in response to the tmique, facilitative, system-transformation role of RMP. If it trips to function otherwise, the national staff will most likely either, (1) keep itself largely irrelevant, or (2) undermine the better regional core staffs in their unending efforts to maintain balance, establish contact and keep in touch with their complex constituencies, and enable the health and medical power blocs I.n their regions to confront health problems rather than each other. We believe the regional core staffs already constitute a valuable resource. We see the primary national staff task to be one of enhancing this resource and helping to develop the networks spreading around them in the more active regional programs. This, to us, offers great promise for dealing with American health care problems. B. RELATIONSHIPS BETWEEN NATIONAL AND REGIONAL LEVELS This section is a direct attempt to answer one of the principal questions asked in the work statement incorporated in the ADL-OSTI contract with RNP; namely, what relationships do and should exist between the national staff and regional levels of the Regional Medical Program? This question has had special significance from the earliest days of RMP. The possibility of diversity among the regions was always recognized. With a relatively small staff in Washington, the problem of communicating with (and in some manner 'loverseeing") the 55 regions would be a real issue, even if the programs all were to be identical. As the program began to assume its current form, it became increasingly obvious that a staff of a few dozen people could not "control" over 50 regional programs each spending from one to several million dollars annually, if they really turned out to do some- what different things for somewhat different reasons. Moreover, in addition to the differences in administrative perspec- tive discussed in the preceding subsection, there were a variety of per- spectives on the substance or basic purposes of the program. The content and nature of the national-regional relationships could be seen to vary VI-6 Arthur 1) little, Inc. considerably depending on one's perspective. (More detail about this subject is provided later.) An example will illustrate the point. If RMP were exclusively a program devoted to the propagation of center-periphery regional structures in the regions, then the real meaning of the relation- ship between the national staff and the regional staff could be determined by that purpose. The relationships would have as their central purpose to see to it that this kind of regionalization happened, to protect center- periphery regionalizing processes wheL.e they seemed Lo be working reasonably well, and to facilitate the spread of expertise on how to regionalize. If, on the other hand, RMP were seen entirely and purelv as warfare against categorical diseases, the relationship between the national staff and the local regional programs would exist primarily to help fight @hose diseases and to gather and allocate money to that end. If either ot these two pur- poses were to be taken in pure form, it would set the content of the REPS- regional relationships on a q4ite unique track and could significantly affect the mode of that relationship as well. A national staff de-voted to center-periphery regionalizing would include, desirably, specialists experienced in bringing about this kind of regionalizdtion. Its evaluators would concentrate on measuring the extent and specifying or helping others to learn to specify the value of particular center-periphery efforts. It would draw heavily on political science and economics to conceptualize, carry out, and evaluate these efforts. Its political role might range from very open involvement to strictly invisible activities in the regions, but it would probably do a lot of alarm-ringing in some appropriate way when center-periphery regionalization was being seriously challenged anywhere. In contrast, a war against categorical diseases organized on self-determined regional fronts would call logically for a battery of national experts in these diseases as well as people closely connected with the voluntary and other associations devoted to attacking these diseases. Politics would be involved; however, a higher proportion would be medical, academic and association politics than in the center-periphery case. Technical medical and delivery-system expertise would be more important to national-regional relationships, and those who were categorical experts of various kinds would no doubt play very important roles in carrying out these relationships. This chapter will not argue that any one of about six or seven perspectives on the program is, or should be, taken in pure form, but it will assert and illustrate that there are now in existence several dif- ferent emphases and different combinations of emphasis. The consequence of these several perspectives in tandem turns out already to have affected the nature and content of present relationships by making them subject to a good deal less clarity than seems appropriate, especially to the proponents of the simpler, more focused perspectives. Equally important, shifts in perspective change the judgments one makes on the quality of relationships. A "good" relationship given one perspective on the program may be irrelevant or even injurious in terms of other DersDectives. VI-7 Arthur D Little, Inc Although we started out by expecting to find that differences between regions arose from differences in geography, demography, and health care resources in different localities, what we have actually concluded is implicit in the preceding paragraphs: The real differences in relationship and the need for varying relationships between the national and regional programs stems from the differences in perception about the capabilities and purposes of the program as a whole or in a particular region. These .purposes can, let it be emphasized, exist in almost any combination in the minds of persons or groups surveying the Regional Mecicai Programs. Let us simply list those we have encountered most frequently. @IP is: a A medical school-support program delivering service toward ifregionalization" in continuing education (and possibly enhancing the power of the academic medical centers); o A program in categorical disease warfare (either in its own right or as cover for a more political set of objectives); o A technical diffusion program (hopefully justifying earlier expenditures in research by applying research results); 9 A program in center-periphery regionalization (embodying either the regionalization ideas endemic in PHS for at least 20 years or the academic centers with community satellites of early "DeBakey" proposal); o A confused abberration of American political maneuvers (that left RNP with no clear sanctions and belonging clearly to no apparent sponsors); o A social experiment in regional autonomy and voluntary co- operation toward self-defined regionalization (system trans- formation); or 9 A providers' "go-between" -- a broker, convener, and faci- litator (toward the process of system transformation). The last two, either by happy accident or imaginative design, have finally resulted in producing the only sort of program that could function usefully in a decentralized health care delivery svstem. All of these views are endemic wherever RMP is discussed. It is theoretically possible that those that are uppermost in a particular region may be most accepted in some instances because of genuine local peculiarities of geography, economics, sociology, or medical resources that make the local views of RMP rather "inevitable." We know of no such examples, however. All the regions we hAve visited show indications of shifting views, either continuously or abruptly changing. These shifts sometimes go to the heart of program definition and identity. Permanence of style, or emphasis on a single self-perceived function, seems remote VI-8 Arthur 1) I.ittle, Inc. from RMP experience in most of the livelier regions. Much more frequently, a regional program in RNP arises as a consequence of interaction among people who originally held one or two of these seven views in more or less pure form -- but not all of whom shared the same views. Where a clearer perspective has emerged, the process of interaction, the relative strengths and successes of various of.the RNP local leaders, and local confidence in handling local relationships with Washington seem to be what really determined the regional perspective on RNP. The process starts with people taking rather simplistic views on what RMP is about (usually one or not more than a combination of two of the first four in the list above.) It typically progresses toward a point in which survivors begin to converge on a view that irelds several of the original perspectives and takes account also of emergent issues on the health scene. The first melding, particularly common in 1967 and 1968, among hospital administrators and-physicians, and even among the RAG mem- bers and some core staff personnel, was that the program made no sense. it was a boondoggle or a mistake. It was easy to see how peopie could reach this conclusion fromthe fact that nobody emerged triumphant and was acknowledged the victor after the legislation had passed, from the clash of views that resulted in frustrating confusion about what to do and priorities for doing it, and from the inability of anyone to pin one of the simple definitions for RNP to any large segment of the program for more than a few months. In short, most of the frustration has been ex- pressed by people who are understandably in a hurry and who feel they have good solutions at hand for most major healthcare problems. During the period in which we have been working with RMP, we have noted growing sympathy and understanding of a more positive view both within and outside the program. As developed earlier in Chapters II and III, History and Regionalization, respectively, there has accordingly been a more generalized willingness to view the RNP either as an experiment in regionally autonomous voluntary cooperation or as a kind of broker- facilitator. These two views are admittedly rather similar. Both are capable of embracing most or all of the original simplistic views, as Chapter IV, Facilitation, and the argument on systems transformation were intended to make clear. Both views have other advantages in addition. These views recognize FM 's political and economic insignificance with respect to the total investment and resources in the American health care scene; and they both provide a set of tasks for ME to do that are almost unique to RMP.* 1. Specification of Areas in which Differences in Relationship are Implied by Differences in Perspective on Program Purposes Some of the "pure" interpretations of RNP are almost impossible to embody organizationally without having authority concentrated at the * They do, however, tend to force a convergence or other form of close working relationship with Comprehensive Health Planning and Health Ser- vices R&D. These relationships, however, are taken up in Section D of this chapter. VT-9 Arthur[) l,ittle, Inc. national staff level. Others are quite difficult to carry out without having a very large degree of autononr .7 and authority in fact lodged at regional levels, or perhaps even more locally. (Given existing circum- stances, including the law, we suggest that the latter alternatives have to be chosen.) Center-periphery regionalization, in particular, is not a credible function for RMP unless backed by central power that is very un- likely to be concentrated in RMP. The expertise, the language used, and the ki@ds of people appro- priate to conducting the relationship between national and regional levels vary greatly, depending on which of the "pure" interpretatione- of @fP one is attempting to carry out. (Even a combination of functions still im- plies a mix: the issue is how well the current RNPS staff fits the chosen mixture, and, if it does not, then how to modify it.) We are leading up to the conclusion that significant restructuring and rather extensive retraining is probably appropriate. The relative amounts and kinds of evaluation, management control, consultative support, financial auditing, and other.functional aspects of the RMPS-regional relationships vary with the interpretation of RNP's role. Some.of this is discussed in Chapter V (Evaluation). Accordingly, many basic management parameters bearing on rela- tionships are affected by the role or combination of roles that RNP nationally and regionally is attempting to carry out: The skills expected of principal communicators in both the national and regional staffs; The balance in internal relationships among the several branches of RMPS; Internal organization of RMPS; Basic procedures of RNPSZ Nature and content of national management information system; and The relative sizes of core staffs. More about all of these aspects is laid out in Section E, (Recommendations). More detail about the implications of the several perspectives on RMP is to be found in Appendix C to this report. Although our impression is that RMPS staff has made progress toward achieving a common interpretation of RNP for itself, some of the conflicts and contradictions of 1968 still persist. Some still believe that RNP reality and purpose correspond only to one or two of the simpler inter- pretations. The difficulty is not unawareness of difference but some unwillingness to concede validity to views different from one's own and to push for a formulation that adequately embraces all the necessary views. VI-lo Artlit! r Lest there be any doubt, the ADL-OSTI conclusion is that all of these interpretations (medical school support program, a program in cate- gorical disease warfare, a technical diffusion program, a program in center- periphery regionalization, political mistake, a broker-facilitator program in process of system transformation, self-defining regionalizer of the provider system) (-I.o have a certain limited validity in terms of national need, legislative enactment, and sheer historical evolution. But the in- compatibility among the most monistic of these interpretations makes it difficult to carry out any of the single purpose interpretations in pure form so long as proponents exist for any of the others. Real-life needs go beyond the limitations of any of these individual perspectives. The urgency of attempting sensible integration of the fragmented health care system results in priorities for the Ilealth Services and Mental Health Ad- ministration that render most viable the view of RMP as an autonomous, self- defining regionalizer or as a facilitator-broker, both working to realize voluntary health system transformation. Accordingly, we will devote the next subsection of this chapter to examining the adequacy of the relation- ships between RNPS and the regions in terms of system transformation and brokerage,-it being once more emphasized that these views can embrace the earlier interpretations and that both views allow a way of positively dealing with the confusion and frustration that follow the notion that RMP is a mistake or a boondoggle. 2. ADL-OSTI Observations on "Relationships" During the time we have been observing RNP, the informal working relationships between a good many of the regions we visited (or checked on at intervals) and the central staff members we have met in Washington have noticeably improved. The Washington staff and regional staffs have become more accustomed to the jobs they do, and are more aware of their counter- parts and their problems in other places. On the whole, these staff mem- bers appear to have more confidence in one another and a greater willingness to share problems and work out common solutions. Evidently their experience with one another tends to be positive. We interpret this condition as an indication of improved ability to function. It would be wrong, however, to imply that revolutionary- changes have occurred and that the millenium has arrived. Many complaints continue to exist on both sides, and if they are voiced less frequently or with less intensity in 1970 than in 1968, part of the reason is mere acceptance of the familiar "status quo." Some people have given up on the possibility that improvement might eventually come about, while others have actually experienced significant improvement; both have good reasons for complaining less. Our conclusion is that national-regional working relations are bet- ter than they were in 1968, but not as much better as the decrease in com- plaining would suggest. Many of the mutual perceptions of the relationships between the RMPS national staff and the regional core staffs are based on the review process. This is natural because the review process obviously constitutes a major and ubiquitous connection between the two PM levels. Specific issues VI-ii Arthur 1) 1-ittle, Inc vary a good deal, but those that have been commonly and persistently ex- pressed will characterize the nature of these relationships. a. Relationships from the Viewpoint of Regional Core Staff and Other Regional Personnel 0 "Review takes such a long time." Once an official committee or a responsible official has reached a conclusion on an application, whatever happens after that point seems to these decision-makers to represent unnecessary delay. From any of the local viewpoints, accordingly, eliminating delfay is impossible until applicants, reviewers, commentators, and vali- dator-allocators are reduced to one person or one very small group. Any work done ostensibly to speed up the process may not be per- ceived as capable of serving that end. Procedures are constantly under revision in the regions and at national levels. However, from the point of view of the individual grant applicant, every succes- sive change which "is for his convenience" seems to have the effect of increasing the time it takes for him to get approval -- or disapproval -- of his application. Potentially massive changes in the review process instituted nationally, such as anniversary review and the recommendations of the FAST Task Force to cut out national level project review, probably will not have the effect desired any more than have past schedule changes and refinements. Certainly they do nothing to simplify the regional review processes themselves, where one-half to two-thirds of the time is consumed. And at regional levels, revisions in the review process have, in fact, been governed by two countervailing impulsions that were much strongerthan the impulsion to reduce delays. These impulsions are described in detail below: (1) , The first and most important of these impulsions is the need of ever more people to "get into the act." As RMP develops further, we can expect this impulsion to continue to be strong at all levels. As RMP gains in recognition and impact, more kinds of experts, more people with power, and more sorts of interests will naturally press to achieve a point of in- fluence in RNP. The review process, whether nationally or locally, offers a myriad of possibilities simply because it bears on the allocation of much of the money through which RMP does its work. This process of allowing influences to come to bear on RNP processes from outside is essential if RNP is to be credible to those who are outside its immediate family. RMP must be open to expression of these influences, especially if RMP is to serve in brokerage and catalytic roles where its organizational neutrality and willingness to attempt to integrate disparate interests are conditions of accomplish- ment and not just conditions of servitude. Even if the long time-constant for the review process is lengthened still more to allow for constructive influences to be brought to bear, we feel that it should be allowed to happen. We are, of VI-3-2 Arthur 1) Little, Inc course, suggesting that the review process has to be pri- marily local; otherwise it will not be visible. With respect to the review, the national staff role would be to test whether the local review is genuinely visible to all the local "pro- viders," and it is also visible and comprehensible enough to non-providers that its workings are acceptable to them as well. We think it would be refreshing for everyone in the game to have a chance to prove good faith (which each provider-group believes it has) and allow real testing of whether good faith will be allowed to get in the way of self-interest. We are also presupposing that RNP regions will have the courage to open up the tough issues in the delivery of healthcare services, in access to health care, and in manpower. The national staff can also play a consultative role in helping them to face these-problems; this would also have a bearing on the content and mode of national-regional relationships. Accepting delays in the review process is not the only option. It would be possible also to build the local RMPs to a point of sufficient internal coherence that the RAGs and their com- mittee structures could include, integrate, and express the external influences with enough openness, competence, and confidence to allow at least simplifications in the local review process. Where one prime requirement is credibility, the more that is actually done where numerous RAG members can see it, feel it, and make it happen, the more credible RNP will be to them. The review process by its past nature tends to be invisible and closely held; the more obvious and more basic are the issues it deals with and the more openly it deals with them, the more credible will be the assertion of integrity in the review process. In terms of the new dispensation, in the year of the FAST Task Force and the beginnings of the "anniversary review," what are the implications of the preceding paragraphs for RNPS? RNPS could put itself in a position to press for and ease- simplification of regional review processes. In the same way the FAST Task Force has specified ways to do this nationally, so can informed members of the national staff make reasonably practical suggestions to individual regions about their processes without themselves becoming reviewers. It is in the interest of RNPS to join this battle. The "anniversary review" is ambiguous and limited in its application, at least in the form expressed in the early summer 1970. Without understanding and assistance, the regions cannot be expected to face the difficult local de- cisions on projects, priorities, and programs any more squarely than they have in the past. "Anniversary review" and the possibility of acquiring a "development component" VI-13 Arthur D Little, Inc only enlarge the problem. Furthermore, the temptation to use anniversary review" as an excuse for delays and vacillation is considerable. RMPS will be the "whipping boy" in absentia for a good deal of this, whatever happens. In our view it makes good sense for strong RMPS staff people to ask to be invited by strong coordinators to come out and help simplify local review processes so that they reflect 1970 conditions, and so that they become as well adapted as possible to emerging local program strategies and the most genuine attempts possible to face basic issues. The increasing presence of national staff members will be interpreted in many ways in the regions; at least some mature local people will be able to appreciate skilled intervention in this kind of simplification process, and the national staff will gain insights invaluable to the refinement, development, and further simplification of national (including anniversary) review procedures. The first attempts at "anniversary review" site visits should be indicative of the manner in which this can be done, although we are by no means suggesting that the consultations we urge can be carried out in the course of site visits alone. To be sure, the national staff has to be clear about their actions when they visit the regions. They have to know how to keep their behavior consistent with their twin objectives of keeping as many program decisions as local as possible and of reducing rigidities in review processes. They are, after all, dealing with situations in which everybody is watching everybody else somewhat suspiciously. In regions where we examined the review procedure closely, mutual suspicion had been the major reason why local review processes had become to complex and slow: the review procedure is an attempt to ensure fair treatment by sharing responsibility widely through a serial review process of many steps. The national review process seems to involve a quite specific set of internal RNPS relationships that should also be noted. This is the division of labor within RNPS between those who have heretofore organized and executed the review process and those who have previously provided feedback to the regions on the progress and results of review. The Grants Review Branch usually knows what is wrong with an application, but hereto- fore members of another branch (people called liaison officers for most of the life of our contract), with less detailed knowledge, communicated review results officially to the regions. As "an- niversary review" develops and the preoccupation with project review diminishes or disappears inthe Division, the national staff will have a good opportunity to work their way through this problem. Our own organization proposals for dealing with it are to be found in Section E (Recommendations). (2) The second pervasive and increasing pressure that complicates and delays the review process concerned accountability for past VI-14 Arthurf) I.ittle, Inc actions deemed to be mistakes. As a federal program takes shape and hindsight shows some early experiments to have been erroneous, some of those responsible for having approved the experiments in the first place become cautious. This kind of caution makes people more hesitant about approving new requests, especially ones that seem strange. People can ab- stractly agree that failure results from a high fraction of experiments and that learning through failure is essential to progress, but someone is nevertheless likely to be left "holding the bag" when failures are identified. The available remedy here is not less experimentation but more clarity about who is accountable for what. A primary task of the national staff is to maximize the accountability of r'egional core staffs, regional advisory groups, or boards. Discharging this task more actively can speed and simplify the review process,, with or without an anniversary review" and with or without"developmental com- ponents." Pressing accountability downward would strengthen local motivation for coping actively with local failures, hopefully by converting them into something more useful than they had become. We suggest explicit guidelines that charged the national review levels with assigning (or recognizing) accountability, and that require the national level to specify these accountabilities to the grantee. The object is not a guarantee of specific performance from the grantee, but merely his undertaking a conscious obligation to shape the ongoing expenditures under the grant continuously, so as to carry out credibly the strategy outlined by the grantee in his application. If national review is to be responsive to local (regional) conditions and initiative, then it should attempt to judge a series of "consisten- cies" at the local (regional) level. These would include: Are the objectives and local strategy.consistent with local needs? Is the local strategy consistent with local objectives Are local strategies and objectives consistent with national priorities? Are specific tactical ventures (projects or other activities) consistent with strategy and objectives? Are expenditures projected for specified program elements con- sistent with local priorities which, in turn, reflect local objectives? Are the various activities of RNP within a region clearly con- sistent with the scale of activities elsewhere in other regions, or is there an obvious and credible justification for their inconsistency? VI-15 Arthur 1) Little, Inc Are local risk and feasibility estimates consistent with credibility? In functional terms what we are advocating is a simplification of the review process (through "anniversary review" or otherwise) by integrating review with a regional planning and evaluation process. In this combination, the regional planning and evaluation process would have to furnish the comparative base with which.the degree to which the consistencies mentioned were consciously sought and actually attained could be judged. The procedural objective of the essentially non-judgmental central-regional dialogue characterized in Chapter V (Evaluation) is -- on the national level -- to produce documentation to allow prospective and/or retrospective judgments on these con- sistencies. Furthermore, maximizing real accountability at the re- gional level will help to make-explicit the national objective of decentralization. The content of the last several paragraphs is, we believe, con- sistent with the language of the instructions handed out to the regions in May 1970, whenthe "anniversary review" process was im- plemented. In fact, on the conceptual level, it is not so inno- tive that it should cause any great difficulties. The problem lies in commencing, adapting detailed procedures, reorienting and re- training numerous RNPS staff members, and obtaining agreement between the regions and national levels on a cycle and style of review and evaluation that is as yet only partly explicit and, to only a small extent, yet undertaken. (More on procedures for carrying out the central-regional dialogue is presented in Section E below.) 0 "They (the nationaZ staff) don't teZZ us much of anything." Over the whole two-year period of our contract we have heard this fairly consistent complaint from the regional core staff, RAG members, task force members, and prospective project applicants. Translated it could be conceived as an attack on the competence of the national staff, and some of the complainants assuredly meant it to be so construed. It could also reflect the changing scene in and around RNP. It is frequently true that policy is changing, and almost any positive statement could be proven wrong in enough ways that relying on individual opinion as guidance to action could be wasteful for the region. By the same token, the risk of in- ducing mistaken activity can limit the types and amount of informa- tion given to regional people by members of the national staff. Another interpretation imputes passivity to the regional people who voice the complaint: rather than finding out, figuring out, and acting for themselves, the regional people sometimes prefer to blame the national staff and thus avoid responsibility for acting themselves. Finally, we note that internal differences among branches and individuals within the national staff further compounds the VI-16 Arthur f) I.ittle, Inc impression of confusion, disagreement, inability to respond to questions, or failure to anticipate local need for information. There are elements of truth in all these interpretations. Further- more, the pattern varies from time to time and place to place. But fortunately this issue is also one that has come into somewhat im- proved perspective with time. With closer acquaintance, people on national and regional levels have come to understand one another somewhat better. Communication now occurs in a context founded on historical experience. Expectations are partly based on past per- formance. Some of those employees and committee members judged least capable of clear communication have been eliminated. But much more still has to be done to resolve one basic issue of communication that goes to the very heart of the relationship between national and local level@. The issue has already been stated: if the initiative is to be local and regions are to have a significant measure of local autonomy, the national staff cannot have -- and should not be expected to have -- substantive and detailed answers about priorities adaptable to.regional applications.* The national staff.can be expected to have good "process" and technical information of many kinds: connections with, or informa- tion about, other federal and private programs which should be explored by people within regions, suggestions on how to go about arriving at a practical answer to questions about resources and feasibilities raised by people in the regions, better information on what is going on in other regions, advice and assistance in recruiting a core staff, consultative assistance for the region that needs such assistance to begin setting its own priorities, sound information on established standards of care (and other practice standards), information on what -- or how to arrive at -- what is established (or "best") practice, and assistance in management development and with consultant training. Determining what the national staff should do for the regions is a difficult and basic issue. Intelligent, experienced, and active people on the national staff cannot help but become more expert as they carry out their jobs. To ask them to serve consultatively rather than authoritatively is to place a special burden on them. To train more of the national staff to carry out this kind of con- sultative-evaluative-planning-support role is to place a very heavy burden on the national director of RMP. It can be better done if both Health Services and @iental Health Administration and the National Advisory Council for RNP agree, and strongly endorse, the primacy of this kind of assistance tole for the national staff. We see the national staff as being a consultative staff, not a general staff. The officials nearest to being "line executives" or "line officers" are the Regional Program Coordinators, which rationale again leads us to accountability. But "what looks good in Washington today?" will always be of regional interest. Clear statements of HSMHA and RMPS priorities will continue to have many uses in the regions without their becoming either commands or governing rules for funding success. VI-17 Arthur 1) Little- Inc Our argument is that regional autonomy can take place and will work only if the national staff achieves a continuously clarifying view of what regional autonomy means and insists on allowing the regions to behave autonomously. This means accountability, which in turn means that the regions will not be "told" much by the national staff. In turn, this implies the development of a "helping" role for the national staff from which their local relationships may flow, consistent with regional autonomy. 0 "They not anZy don't teZZ us, they try to hide things." People in our culture have an understandable reluctance to wash dirty linen in public. Our experience as consultants tells us that the linen usually looks a lot dirtier to the people closest to it than to others. Revealing that it is dirty often has no negative effect; concealing it may have quite.a negative impact on those who are closest, but literally remain unknowing, though necessarily aware that something is peculiar. Our experience tells us that al- most nobody is really skilled at concealment without resorting to behavior subject to a lot of misinterpretation. For example, what is done to keep some minor controversy quiet, usually in the name of reducing vulnerability to external attack, is interpreted by those from whom the information is concealed as some kind of plot against them. That is, what is really intended as a defensive opera- tion gets interpreted as being offensive and aggressive. The manifestations are mysterious or hidden, but the intention to con- ceal is rather apparent. The moral of this philosophy is that the RMPS staff should adopt a policy of communicating freely, clearly, and explicitly on matters affecting the regional people and the plans they are trying to im- plement. To say that "the matter is under study" is better than to say that "I don't know anything about it." Of course, to say that lithere's a disagreement going on between Grants Management and Planning And Evaluation. We don't know what the resolution will be yet. All we can say about the issue is would be far superior. If the person raising the question may be affected by the outcome of the disagreement, he then has a chance to furnish potentially constructive and specific input that may swing the decision in his favor -- and may also have a much more general significance in forcing more decisions to take more account of more of reality as experienced in the field. We have seen more and more openness of this kind on the part of an increasing fraction of the RMPS staff. In our view, those persons w,;o cannot take the risks involved in working out sensible ways of discussing internal disagreements should not be given assignments requiring such discussions with regional people in the first place. They are just not ready for that kind of responsibility. VI-18 Arthur 1) 1-ittle, Inc. 0 ?'The RAIPS people @ a Zong way off. The typical RMP coordinator feels quite distant from Washington much of the time; though again, it is clear that whatever the short- comings of some of the previous annual and topical national meetings between RMPS and RMP 'staffs may have been, the Airlie House conferences and some of their other, smaller meetings were beneficial. They have shown both RNPS and regional coordinators that they were in closer, agreement in some ways than they had previously recognized, or at least could tell in the context of larger meetings and earlier ne- gotiations. The mere passage of time is also helping with this problem, relationally, psychologically, and in a sense, even in terms of overcoming the effects of physical distance. However, the distance between the regional program coordinators or directors and the RMP@ senior staff persists; some gap no doubt will always remain, since institutional interests on the two levels are not necessarily nor perpetually identical, and geography will also continue to have its own effect. We have found most people quite sensitive to this gap. No doubt, this can be attributed to the fact that RNPS is a Federal program attempting to operate in an environment previously hostile, contemptuous, or at least iso- lated from government programs of any kind. It should be clear, and we hope it is, that program coordinators have almost had to play this game both ways. Regions obtain their money through Federal processes. The regional constituencies, however, are made up almost exclusively of'hon-Federal" people. Most citizens who are not directly employed by the Federal Government and many who are can unite in sharing the stereotype that the Federal Government and its program are inefficient, cumbersome, and often unimaginatively administered. Agreeing on this topic can serve as a convenient horse on which to load real disagree- ments that can be sent galloping out of the discussion. All this is so much a part of the Governmental scene, the American culture, and the medical-political subculture that we would not dream of bothering to say, "Change it." It is a psychological factor in RMPS-regional relationships that is best controlled by the regional discovery that at least some of those Federal bureaucrats are competent, sensible, friendly human beings. The best way to permit this discovery is to have the "Feds" come out as visitors and get intimately involved in the details of those activities that some of the people in the regions are trying to accomplish. b. Relationships from the Viewpoint of RNPS 0 "They give us 'dog and pony' shows." There are complex reasons why national staff members visit the regions so infrequently. These reasons go beyond the mere fact that traveling is exhausting, takes lots of time, and costs money. One of these reasons is expressed in terms of the "dog and pony show" VI-19 Arthur D Little, Inc treatment accorded frequently to "visiting Feds," who feel there is little sense in going to the field to find out what is going on if you are then exposed only to careful contrivances and whisked from place to place on a hectic basis. "Dog and pony shows" are more likely to happen, of course, when a number of site visitors appear on the scene together. Under those circumstances, or where there is a real urgency to cover ground both substantively and geographically, "dog and pony shows" serve a real purpose. People on the national staff certainly appreciate this problem and said that they can get closer to the "grass roots" when they visit the regions informally singly or in two's rather than in larger groups. The main point in this discussion isthe reciprocal effect of distrust between the division and regional staffs. As we have been saying, the regional people still feel--that they are not being cued into the Washington scene; similarly, the "dog and pony show" is seen as a device to infonn the "visiting Feds" on innocuous regional activities and no more. We would advise regional coordinators, when site visitor@ are identified, to call their best friends and find out what these particular site visitors would really like to know. If there is any possibility they might like to really understand what is going on, then they should do their utmost to informthem of their activities. 0 "We're Feds-, so they don't dare teZZ us things." One basic way of overcoming the regions' reluctance to be candid with the national staff is to connect it to the purse strings. Mature members of the national staff simply do not expect regional people to reveal activities in their regions if, in their estimation, it might represent a risk to their regional budgets. We have been told any number of times that national staff members cannot serve effectively as consultants to the regions, since, supposedly, they will always be considered as spies whose reports to headquarters significantly affect funding. We have never been able to understand why those members of the national staff who have used this explanation believe that they are dealing with a single-edged sword. After all, from the regional viewpoint there is always a chance that friends on the national staff will uncover more money for dealing with problems that are demonstrably more severe than are found in most other places. We can be even more peremptory. The regional core staff member who is incapable of leveling with a member of the national staff is not going to be a terribly useful or dependable "facilitator." We believe there are few valid excuses for holding back. If the national staff wants to find out what is going on in the region, it should be able to do so. The process of discovery can be ex- hausting, but acquiring a sense of what is really going on does have a fairly consistent effect; the more you know, the more people can tell you. VI-20 Arthur 1) Little, Inc. C. OBSERVATIONS ON RELATIONSHIPS BETWEEN REGIONS In general, relationships between regions tend to be good. Where geographical overlap t)etw@eil regions includes an area where jurisdic- tional or other problems already exist, the overlap is likely to be a problem for RMP. We asked to study two adjacent but overlapping regions, specifi- cally to find out more about this type of relationship. We chose the New Jersey and the Greater Del,.iware Valley Regions. Southern New Jersey is a typical case of a population whose interests, affections, dnd loyalties flow in a variety of directions. Relatively close to Philadelphia, rela- tively remote from northern New Jersey, these people quite naturally look to Philadelphia as a center,,- in many ways, including to some extent, medi- cine. But laws, regulations, certifications, and the like, flow from New Jersey. The pattern simply is not clear; the agreement between the two R@iPs to overlap in a selected series of counties was a sensible recogni- tion of realities, and it appears to work reasonably well, despite dif- ficulties and a past history of tension. Earlier in this report, we said some things about the situation in Memphis where the "centers" of political, economic, medical, and other sorts of activity are considerably more widely separated than they are on the eastern seaboard. Outreach from Memphis extends into five different states. People who live midway between Memphis and some other center are assumed to be willing and able to turn in at least two directions. Our im- pression is that free choice and bartering are everyday experiences on the borders of the Memphis-Midsouth Region, at least as far as middle-class people are concerned. Equal accessibility to two centers is taken as part of the normal coin-in--trade of those who live midway, whether in medicine, marketing, or education. The six or seven RMP regions involved accept this condition,and it is simply not much of a problem. It looks as though the regions are well able to handle their border problems with one another with minimal nudging from the national staff or members of the National Ad- visory Council. We have more concern with the relative weakness of interregional organization among the coordinators. With the good example of the south- east (and possibly the southwest) coordinators as the exceptions, the relative failure of coordinators in other sections of the country to ac- complish as much when they get together seems worthy of mention. The southeastern regions share some significant qualities that may have made it easier to work together. But in view of some of their significant differences, their association has been very valuable. Observation of their coordinators' meetings and the information emanating from "counter- part" meetings (in which the functional specialists in the core staffs of the southeastern regions meet) showed their intrinsic value. In these meetings, common problems are discussed, argued, and, for the most part, solved quite well. What one region is about to confront, others often already have experienced. Coordinators counsel one another on the in- tentions and policies of the national RMP. From our observation, for VI-21 Arthur D Lit tle- Inc- example, we have seen a simplistic and extreme view, once expressed, worked through, sometimes quite directly, sometimes only after an inter- val has transpired, until the coordinator responsible for the remark in the first place realizes the complexities, either disregarded or unknown, when he presented his formulation. Although not always successful, clearly interregional meetings of the southeastern regional type have continuing potential as training sessions if leading coordinators will treat them SO. There may always be some reluctance at the national level to en- courage the formation of these sectional or interregional associations, since they could become potential alliances or pressure groups. @What we have seen suggests that it is more in the interest of RNP to have co- ordinators working together with one another than otherwise. The competi- tion among regions (and their coordinators) is real, and is based on fundamental factors, Lncluding the scar'city of funds. Creating mutual support networks among coordinators seems to us to have very great poten- tial reward. In the first place, to the extent RMP remains as a continuous "learning" program rather than a routine production program, the learning will be both more rapid, more rapidly digested, and really used if shared by regional peers. In the second place, where RNP must function politically on both local and national levels, maximum opportunities for coordinators to work together are important. It enables better constituency-building efforts, and more and more frequent opportunities to present the RNP case to outsiders who have to be included in its activities. We suggest that these advantages outweigh whatever risk there is that coordinators might form sectional power blocs capable of threatening the whole, or of looking like "improper" associations from a distance. Are technical expertise and functional information shared inter- regionally? We have spot-checked interregional diffusion a number of times. In this regard, as in most others, the development since 1968 is very substantial. People in specialized roles in core staffs are increas- ingly confident that they are aware of activities in other regions. We have never systematically tried to verify how various bits of knowledge and various approaches to problems have actually spread from region to region, and we do not know whether there have been instances of simul- taneous invention. What we have observed is a tendency on the part of core staff members to deny the importance ofthe national staff in keeping them informed on overall RNP activities and policies. Even in 1970, to discover a core staff member who believed he got much help from the national staff in locating people in other regions who had specific in- formation with which core staff members were concerned would be considered a rarity. In terms of our own knowledge of some of the principal members of the national staff, many of whom are well informed about the activities in technical fields in many sites within and without RNP, this phenomenon appears to be an example of network development of a rather familiar kind. By now the concerned and active members of the national staff and the con- cerned members of local core staffs both know who the "interesting" people are in the specialties in which they work. Most of the core staff VI-22 Arthur 1) l,ittle, Inc. members who have been in their present jobs for a year or two cannot be helped much by the national staff whose knowledge extends to past acti- V44ties and the relative worth of certain individuals, and rarely to what problems the core staffs should next address. National sta@f members can contribute something special by helping to identify new leaders as they emerge within the core staffs and elsewhere and calling these people to the attention of their regional core staff counterparts. Short of a reorganization of the national staff that would vastly increase the intimacy and comprehensiveness of the knowledge of the regions currently available in Washington, this would seem to be about the best help that national staff members could provide to aid in what is essentially a direct interregional process as things now stand. D. A BRIEF NOTE ON RELATIONSHIPS BETWEEN RMP AND "THE WORLD" Although not explicitly asked to deal with RNP's external relation- ships, we would like to volunteer a few remarks, since allusion to the sub- ject are shot through this report. We see the program behaving in the regions as a facilitator; naturally we also view the appropriate external relationships of the national programs as those that help to carry out facilitative functions. Given that view, we see the national staff and National Advisory Council trying to account to the Administration and to Congress for the necessity of RMP serving as a facilitator, discovering ways to justify this facilitative behavior, arranging to discover what is the best practice in certain specific bits of medical practice that get much attention in RNP projects, building political networks to support system transformation efforts, and actively collaborating with other Govern- ment and private programs. We will discuss briefly two facets of the immediate external relationships of FM as viewed as a national program: (1) What RNP can do for HSMHA, and (2) RMP in terms of CHP and Health Services R&D. 1. What RMP Can Do for HSMHA RNP can connect the Health Services and Mental Health Administra- tion (HSMHA) to local.data about providers and how well they are grappling with their problems. These data are otherwise much less accessible to HSMHA through its other programs. RNP can reach local establishment people, including the local private practice medical establishments. Furthermore, while other HSMHA programs connect this agency to public health people, especially at the State House level (but also in local and educational levels) and to that part ofthe medical educational establishment that concerns itself with mental health and community medicine, RNP can open up HSMHA channels into the heartland of academic medicine, way beyond the limits of Federal and other Government health and medical programs. What this amounts to is simple. RNP can connect HSMHA to the largest available selection of locations where actual decisions on the short- and long-range resource allocations in the so-called health care VI-23 Arthur 1) Little, Inc industry are made. Short of Federal intervention on a massive scale to shape the flow of money through the health care field via central controls, PM can put HSMHA potentially into a position to influence the attitudes and ultimately the behavior of large numbers of locally influential people, who are presently in the best position to effect systems transformation in health care. One can surrender to the immensity and difficulty of the task or one can use available instruments to tackle the job. Regardless of the future of national health insurance, the people in the health care professions will be the same people. What they believe to be appropriate and feasible will obviously strongly influence what actually can be done. We conclude that the qualities of RMP that may be usefxil to HSMHA are: a Sensitive Listening Posts-.,The majority of program coordinators, many core staff members, and increasing numbers of RAG members are sympathetically aware of the problems the Federal health establish- ment is trying to solve, and they are equally sensitive to the problems of the private health establishment. They are well located to hear and interpret wliat is going on. o Sounding Board - Program coordinators survive, contribute, and gain in reputation if they are active proponents of change, but slightly on the conservative side with respect to judging feasibility of pro- posed chpnges. If the cannier program coordinators tend to feel an action could succeed iii their own regions, it has a good chance. o Channel for Communications vith Physicians, because of the pre- ceding listed qualities as well as requirements for involvement of physicians in RNP as written into the law. 9 Test Bed to work out and improve system transformation processes and objectives. The widespread desire of core staff members"to " the skills and knowledge they have and continue make a difference, to accumulate, and the paradox that "demonstrations" are relatively rarely replicated suggests a large role for RNP. What is to be done to make sure that a health care delivery demonstration is as well adapted as possible to work with and constructively to modify the institutional arrangements around it? What is to be done to identify other sites where it can be tried, other champions for it, other auspices under which to sponsor it? RMP is well-equipped to conduct at least a share of the "action research" or field transforma- tion processes to accompany health services research, whether in its own right or in conjunction with other programs. 2. RMP, CHP, and Health Services R&D- We have sensed two somewhat inconsistent pressures on RMP, CHP, and Health Services R&D. On the one hand, all three are now subject to coTi.,iequences to follow from the creation of the new Health Services and Mental Health Administration. Like every comparable reorganization, this VI-24 Arthur 1) Little, Inc one brings with it the need for rational self-justification. Bureau of the Budget and Comptroller questions about roles and missions are one expression of this need for self-rationalization. But this is by no means merely an externalized pressure. If the people we have met are representa- tive at all, then the very personnel in the programs feel the programs should make sense, and be related to one another in terms of some over- all conceptual pattern. But the other pressure is equally strong. It acknowledges the differing constituencies and approaches to problem-solving that are em- bodied in the three programs It also acknowledges the pre-existing com- mitments and programs that w;re assembled under both the Partnership for Health and the Health Services R&D legislation. Through administrative processes, RNP has inherited some commitments, too, in its merger with the Chronic Diseases Control Pr-Qgram. The effect of these pressures is to retard rationalization and to delay the delineation of clear, conceptual distinctions among these programs. Political power to enforce any of the presently developed distinctions is insufficient, and this is the ultimate reality behind this pressure. Because this pressure is still dominant, we have seen little reason'to modify the tentative conclusion we reached nearly two years ago. CIIP and RMP can find their way with respect to one another more construc- tively on local levels than they can through administrative fiat at national levels. Integration between Health Services R&D and RMP also seems to be simple enough in specific projects, and should be similarly pursued. We can imagine a continuing gradual convergence of the three programs, if a more or less common strategy evolves for helping American health care professions and institutions to become still more responsive to changing needs. This convergence can be further supported by a gradual convergence of constituency, particularly between the CHP and RNP. Looked at in terms of the mechanism available under RNP legislation and under Sections 314(a) and (b) of Comprehensive Health Planning, both programs are saddled with the necessity of constituency development. In specific localities each can readily reach a point where it will be able to progress effectively only as it begins to come to terms with the constituency of the other. In our opinion, the conclusion is inescapable that all three programs are attempting to deal with essentially the same problem. Like another broad-based social programs with which we are familiar, they have to work on the basis of partly untested assumptions. This inevitably leaves a certain amount of room for conflict, for clash of interests (both bureaucratic and outside the Government), and for sheer intellectual con- fusion. Both Health Services R&D and RMP do partly share in the "research" strategy for social change; that is, both of them respond to the theory that adequately supported research-based findings will provide a strong and defensible basis for social change. RMP and CHP share the theory that participant planning and improved communication within communities will provide a desirable basis for acceptable and constructive social change. VI-25 Arthur D Little, Inc. Both theories have supporters; each can be either illustrated or decimated with examples; the ultimate truth of neither has yet been established; but they are two of the best bets going for bringing about change with a mini- mum of coercion. In our view, retaining all three programs entails two serious risks and one somewhat less serious one: 9 Conflict among the programs could lead to so much confusion that the assumptions on which each program is based would be inadequately explored. In that case neither proponents nor opponents will be any further along toward understanding what works or under what c:L-rcum- stances, even by the middle 1970's, after at least a decade of experi- ment with these approaches to assisting the health care delivery system to evolve. o The real overlaps conceptually and otherwise among the three programs will make them increasingly vulnerable to external criticism, which, though largely irrelevant, will require no less defensive energy to be expended in the future than has been the case in the past. Distinc- tions between "personal" and "public" health, between "planning" and "demonstrations," between the "categorical" and "comprehensive" represent some valiant efforts to clarify the boundaries between CHP and RMP, as are efforts to decide which of the three programs is really a department of which other one. But these efforts all seem to be subject to nothing but misunderstanding, and lead mostly to argument. o Less important, in our judgment, is the possibility that really inexcusable duplications will occur. This seems to be a much lesser risk because people are so sensitive to it and so many mechanisms exist for detecting it. At this stage the three programs may be approaching a point at which their joint investment and the HSMHA investment in minimizing duplication will soon have to be scrutinized to see if it is not costing more than would the duplication. Let us specify in a little more detail what we believe distinguishes the constituencies and some other aspects of the three programs.* Heal-th Services R&D has support among medical and public health academics (es- pecially epidemiologists, community health researchers, health economists, and some of those in public life who believe we need more facts before changing things around.) Health Services R&D is viewed as academic by activists who question whether its research will reach conclusions that can * It should be recognized by the reader that the study team made no con- certed study of CIIP or Health Services R&D. We have quite possibly ex- pressed emphases that some of the constituents of each of the programs would not recognize or agree with. We are willing to run the risk in order to try to advance the discussion of the differences and similarities among the programs by sharing our impressions, even though they could be made more complete. VI-26 Arthu r D l,i t t le, Inc be acted on in time to be of major use. While its constituency is certainly not limited to those who conduct research under its auspices, the health and medical care community in general is not aware of it or has little faith in its present activities or sponsored projects and grants. CHP appears to be supported by the health establishment, allied with the more moderate among those who favor community control of community services. A relatively strong infusion of hospital facilities planning council supporters brings CHP a not-quite-yet-realized step closer to the private providers of health services than is Health Services R&D. CHP ap- pears to be developing a constituency among its participants in local and state health planning councils. Although the economic and political frustrations of development so far have made this a spotty and variously composed group of people of 'apparently no great size, it has potential on both state and local levels for keeping CHP in a position to assist us in the integration of health planning with various other aspects of community planning, and to keep CHP connected to local social and political leader- ship. RMP has a small but significant part of its constituency in medi- cal school faculties, but more of it among the medical friends and supporters of the regional program coordinators. Its major constituency is composed of physicians who believe that voluntary processes are the best or only way to change the process of health care delivery, and who believe that physicians should take an active and constructive role. Hospital administra- tors, and voluntary association members and trustees sprinkle the RMP con- stituency as do members of allied health professions. RNP is generally quite weak in "community" representation among its constituency. Physicians and allied health professionals constitute its real potential, together with their allies in community leadership. Health Services R&D represents a dual strategy: find the facts that will.enable and justify change; avoid error by acting only when the facts are at hand. The CHP formula is community involvement and conceptual planning, with facilitative processes backed up by political sanctions where necessary. FM assumes that providers, given appropriate support, will overcome their own in-house disagreements to the point of becoming willing also to work with people outside their usual circles. RMP also assumes that a role in enforcement would be inconsistent with encouraging voluntary processes. In addition to the common ground among the three programs, since they must all relate to the health care org?mization crisis as such, each program has special concerns not necessarily shared with the others. Health Services R&D has vital research methodology concerns and needs to make very close examination of the processes in which medical or health care services are directly and actually delivered. CHP has to be involved strongly with general environmental impacts on health and the rational allocation of resources. RNP needs to create working communication and education networks among providers in the actual practice of delivering their day-to-day professional services. VI-27 Arthur D Little, Inc Health Services R&D reflected growing academic and political recognition that many pre-existing activities and some new ones could be profitably combined, because health services delivery and its organization was increasingly broadly regarded as a subject including many elements in addition to physicians' laying their hands on patients within or outside hospitals. Its organization also reflected steps toward the reorganization of the Public Health Service. RMP reflected the interaction of doubts about the adequacy of the biophysical research strategy, by itself, hopes for rationalizing the health care system, and recognition that voluntarism was the only method immediately acceptable to the medical profession and at least some other health care interests. CHP reflected both the long- held hopes of members of the Public Health Service to take a hand in rationalizing health care organization, and their ambivalence about the prospective impact(or lack of impact) of RMP as RNP was legislatively developed. Accordingly, in our view the three programs are significantly different in their origins and interms of the impulsions and assumptions on which they are based. We believe that all three sets of impulsions and potential constituencies could be valuable when applied to solving a set of problems that go beyond the abilities of any of the programs to solve alone. If we were in the Health Services and Mental Health Administration, our main concern would not be overlap but whether enough of the people in all three programs were trying hard enough and sensibly enough to carry out the charters they have, and to reflect theories that are supposed to impel them. In any case, as is clear by now, each program represents a sufficient constituency inside or outside the Government to make it hard to modify significantly through legislation. Without being privy to the discussions among the three program managers, we do not know how far they have gone toward creating a process for testing the relative effectiveness of the things they do, or toward working out a common understanding of where they are convergent and where divergent. But we suspect that initiating any such process would be ex tremely difficult from the program manager levels themselves. Such a process would have to have firm and consistent support from DHEW or great emphasis in the HSMHA front offices. Our inference is that such support would not be forthcoming or felt as "supportive," because the three programs together constitute a kind of embarrassment at higher administra- tive levels. Our impression is that each program manager has to conclude that one of his fundamental, continuing tasks is to show cause why one of the other two managers should not instantly swallow his 'program whole. Under these circumstances, to expect the program managers to work out a sensible accommodation or common strategy is fatuous at best. It may be, however, that the legislative process at work in 1970 and an annual budget cycle or two in which these three programs fared relatively well has already, or will soon, take care of this problem. If not, we hope a suc- cession of relatively consistent champions for the three programs will arise in HSMIIA, to provide an atmosphere in which all three program managers can consistently devote little or no energy to worrying about the other two. The job before them can be better done -- perhaps only VI-28 Arthur D 1-ittle, Inc done -- if the constituencies of all three begin to communicate and interact constructively. E. RECOMMENDATIONS 1. RMPS Reorganization: Effect Top-Level RMPS Accountability on a Region-by-Region Basis by "Assigning" 12 to 20 Regions to a Team of 2 to 3 Top Staff Members and Their Immediate Staff Assistants, Relieving These Senior Personnel of Most of Their Functional Respon- sibilities; Collect the Functional Branches Under One Official. If RMP is to be a successful facilitator, negotiator, systems integrator, and transformer, an acceleration of one of the tendencies in RNPS organization is in order. This tendency is toward more knowledge in depth on the part of more of the senior people in RMPS about the regions and their activities. This also implies realization that anything that happens through RNP has to happen in and through the regions. The present functional organization of RNPS, to an extent, tends to split the knowledge available about individual regions. The grants review specialists have a good deal of detailed understanding of what is in and immediately behind application documents; The better informed auong the liaison people are closer to some real-life dilemmas that are not quite confronted in the application documents or are left out altogether. The planning and evaluation staff very often is looking for broad trends and trying to make sense of all the regions taken collectively, but often without getting close to the specifics of what is really happening in many individual regions or taking part in discovering the political and social realities in the regions and with what problems the local RMP is coping. Our contact with the functional specialists in continuing educa- tion, computer technology, categorical disease areas, and the like, has been particularly intermittent and spotty. Some of these people have a basic understanding of regional activities, but the focus of the technical, expert almost has to be more on "telling" than on "listening." The pres- sure that technical experts feel and openly accept is a pressure to supply expertise, cognitive information, and solutions to problems. In consequence of all these factors, there are regions that nobody at @S "knows," a-,id more regions that are known only superficially, or from too limited a perspective, to efficiently carry out the central- regional dialogue advocated in Chapter V (Evaluation). Furthermore, until the last year or so, what did or could constitute a regional "program" was so uncertain that filing information in terms of regional "tmits" was easy to do but was, if anything, even less intelligible than classification by disease categories or project types. Certainly the basic regional ab- stracts and application files could give no one a picture -- let alone a clear picture -- of regional activities, in AThat context or to what end. These files were arranged to supply concrete information on the funding obligations for projects and regional offices, and minimum data on the activities funded. What was supplied was information crucially needed. VI-29 Arthur 1) Little, Inc It reflected the internal organization of the national staff. It went about as far as available energy and the existing consensus on RMP allowed it to go. Although all of these national staff members are, in fact, trying very hard to make the "mechanisms" of RMPS effective, each is also aware of the need to keep out of the way of other people who claim, or are as- signed, various pieces of the job. The result of this awareness is a further pressure against assembling knowledge about the regions. "Anniver- sary review" and the director's desire to develop his management informa- tion svstem must be stimulating some attempts toward integration. It will begin to be necessary more and more often, to look at a region and its activities as a more or less integrated whole. Somebody will have to be in a position to do so. We believe that the most competent information system and national-regional dialogue "mechanism" is embodied in the senior people in RMP-, the people called "directors" and some of those immediately reporting to them. Simultaneously, we suggest that the various functional branches should tend downward in the RMPS structure and be brought to- gether. They should report to a front office with responsibility for maintaining 4n adequate quality of functional expertise of the right kinds inthe several functional branches. These front office people should also milk the collected knowledge and impressions of the functional experts on regional activities. This implies that the senior person in charge of these functional branches should also be directly involved in regional liaison, as would be the other senior people in RNP, although he might be expected to deal with many fewer regions than the other senior RMPS officials, or in a less direct way. What,we envision is rather informally sorting out the 55 regions. We do not see any great importance in how the division is made., nor do we believe that the regions should be mysteriously associated with one another according to some over-reaching and permanent pattern. But we do advocate the explicit assignment of a senior staff member, or preferably two senior staff members, to-a series of regions, the regions to be associated with the national officials on some basis that takes account of mutual friend- ship, common perceptions, and mutual need. We think that up to 10 of the top RMP staff should be given this kind of assignment. Each pair or group of three of the senior staff members should be assigned between 12 and 20 regions. These staff teams would be almost completely composed of people who report quite directly to the RNP Director now. In this sense, organization structure would not change, for these "teams" of senior staff people would continue to report to the Director of RNP. We suggest that RMP staff members be teamed, and assigned to a rather large number of regions for three reasons. 9 There will never be enough available and qualified personnel to do it on a more intensive basis. Teaming at least increases the possibility that coverage can be meaningfully arranged when necessary. VI-30 Arthur D Little, Inc 9 Direct exposure to a large number and variety of regions, if carried to a reasonable depth, provides a base for comparison that should either reduce the problem of familiarity, breeding advocacy, or place the resulting advocacy on an easily justifiable basis, and in a context appropriate to evaluation as a mutual learning process. o Teams could be instructed to cross-check themselves on the issue of unjustified advocacy for "their" regions. However, we are proposing this system because it seems to us to be a viable method to bring the regional-national evaluative dialogue into being, not because we see it as a panacea. We have already said that the functional branches, for the most part, should be depressed in the organization, each left in the direct care of somewhat lesser senior officials. These branches might be re- arranged. Certainly some of them should be expected to be much smaller. The present liaison officers would disappear altogether, functionally speaking, and the grants review branch would be reduced to a small fraction of its present size. Many of the present members of those branches, through retraining, could relatively easily occupy other roles; some of them as technical consultants; others would become assistants to the senior staff. We want to make a few additional remarks about Grants Review, and to a lesser extent, Grants Management. These branches have been and will continue to be of considerable importance. But we are indeed sug- gesting that Grants Review be made still more responsive to what we see as the overriding purposes of facilitation in the interest of systems transformi--i.cion. An appropriate meshing of the functions involved would encourage Grants Review to devise criteria, procedures, and feedback tech- niques that would signal the importance of systems transformation to the regions, and would improve all application documents as sources against which to consider the appropriateness and concreteness of re- gional activities viewed in their relationship to systems transformation, as well as in other ways (e.g., technical diffusion, voluntary regional cooperation, adherence to sanctioned medical'standards.) Earmarking of funds is also a device that can be used in the interest of systems trans- formation, and one in which both grants branches have an interest. Assistants to the senior staff would be needed to do a good deal of writing and some data-gathering on behalf of the senior staff. We would imagine that most of the senior staff whose jobs were largely focused on the regions would require about two assistants for these pur- poses. Typically, these assistants would be experienced members of the RMP staff who had been members of the functional branches and were selected as possible alternates and eventual successors to senior staff members who had either retired or had gon@ on to other jobs. Each of the staff teams would carry on a major share of the regional-central dialogue. VI-31 Arthur 1) 1-ittle, Inc our knowledge of the bureaucratic affairs and Civil Service regulations is sufficient to make us aware of some possible vulnerabilities in trying to maintain a group of 6 to 8 very senior civil servants afloat without anchors in phalanxes of GS-7's. Special justification might be re- quired, either for the seniors or for the "thin" supervision over the func- tional branches. So a tradeoff might be required between the agony of justi- fication and the simplicity of working informally for awhile in the way we suggest without going through a reorganization hassle. At present, the fundamental difficulty in applying this recom- mentation is the small number of experienced senior staff members actually on board RMPS, and the difficulties faced in recruiting personnel of high quality and broad experience. Some of the problems involved are beyond the scope of RNP or its program manager. More can be solved overnight. What can be done? Attaining HSMHA agreement that RMP is centrally concerned with systems transformation through facilitation processes is the first step. The more real the commitment of the administration to this perception, the more completely can internal organization and recruiting be attuned to it. Some training and experimentation in facilitation with the top people remaining might also begin quite soon, though their number is so small that work pressures minimize the time and energy available. Finally, there is a range of possibilities with members of re- gional core staffs. While most of these are none too liberally staffed with facilitators, at this stage regional personnel resources are actually rich, compared to the senior national staff. One or two regional officials might appropriately be hired into the national staff. A few more could be conceivably seconded to Washington for temporary service, on leave from their regular employment. The executive committee of the coordina- tors could well be involved for its suggestions, too. Its members include some of-the most broadly experienced facilitators in RMP; their ideas of -what to do and who could do it should be of great interest and utility. Abstract qualifications for facilitators can be found in Chapter IV (Faci litation). Personnel sought for service in RMPS must have some good experience in fairly complex administrative or consultative work, outside the federal bureaucracy, as well as government experience. They need to be skilled interviewers, should be rather forceful personali- ties, and need to know how to listen. They should also have the ability to talk more or less as equals with most of the people they meet in the regions (a matter of experience more than status.) We recognize that the jobs we are describing are apparently un- like most of those ixow recognized in federal service. We believe the differences to be significant. The main differences, we expect, will lie, not in the explicitness of assigned responsibility, but in the paramount need to communicate and do work as influencers rather than as controllers. The demands placed on them to serve as live supervisors will be subordinate to this role. The men and women charged with regional liaison would con- stitute more or less collegially the top management of the national staff. VI-32 Arthur D Little, Inc Their power would stem from their direct involvement in national decision and policy-making rather than from their control over administrative pro- cedures. 2. Procedural Changes: Effect Self-Evaluation Through Regional- Central Dialogue. If RNP is to resolve on some such course, frankly viewing it- self as a facilitator-broker in the interest of sy'stems transformation, a number of activities and thrusts will have to be reduced proportionately. Some of these are implied or stated in the preceding recommendation on reorganization. These changes and further developments wili inevitably affect the procedures through which RMP administration is conducted. The most important RMPS procedure will be that through which regional strategies are discovered, evaluated, and reported. Obviously, this means being able to discuss the strategy in the context of the actual operating conditions in the region, and it means updating, verifying, and evaluating these strategies and the processes used to implement them. This set of knowledge and judgments is that on which decisions will be based. These decisions are not only those related to funding, but also those related to allocations of support and consultative services to the regions to work on specific problems. We see the process of discovering, updating, verifying, evaluating, and using the regional description in terms of an annual cycle. Not all regions will be in the same stage of the cycle simultaneously. A similar approach has been taken heretofore in the review cycle, resulting in three or more batch processings of applications annually. Obviously, the measures now required to adapt a series of overlapping cycles to the one basic an- nual fiscal cycle of the Federal Government would still be required. a. The Annual Cycle (1) Initial Step - The first move is up to the senior RMPS staff members w o@se job is to relate to the region in question. They should prepare a statement that expresses their views on what the region is doing and reasons for its activities. They-should say whether they believe that what the region has been up to recently is good, bad, about what could reasonably be expected, or indeterminate. This statement should be written, and may vary from 1 to 2 pages, to 100 pages. It should be based on what the authors already know, suspect,, hope, or fear about regional activities. It should attempt to outline starting conditions, problems, objectives, strategies, and how specific activities have either implemented or failed to implement these strategies. Obviously, once such a system has been implemented, the initial statement in the annual cycle will consist of revision of data and concepts in the file from the preceding cycle, hopefully informed further by impressions gained from other sources, from within the region, from elsewhere in the RNP community, or even from outside RMP. VI-33 Arthur 1) I.-ittle, Inc. A statement from a nondescript source will not suffice. It must be written by someone who is reasonably well-known and respected by people in the region, but distinct from it, because it should be slanted to reveal as nearly as possible the genuine problems and frustrations every region suffers and the in- adequacies experienced by insiders or observed by outsiders. In short, it should be a statement that incorporates at least some of those materials which, were the writer an outsider, in- siders would not have disclosed,, but will discuss with persons already partly aware of them. Thus, it would be an informal document which could be used as a first step in a dialogu,e. National staff members involved in the process also have the obligation to express their understanding of national priorities, and how these priorities -Ere reflected, or otherwise treated, in the regional program under scrutiny. The problem they face is accuracy. They have to understand (or to come to understand) what the program is trying to do, and to be able to express its aspirations and the difficulties it faces with considerable sympathy. They must also express national priorities clearly. For one of the reasons for the dialogue is to take account of the desirability of applying different priorities in different places and to varying degrees. The resulting variety owes (or should owe) to the success with which the dialogue constitutes evaluation as learning. The explicitness with which national priorities are treated results (or should result) in fairly straightforward justifications. (2) Informal site visit - When the program coordinator has read the document sent from RMPS and allowed others of his choice to do likewise, a site visit is in order. The site visit should be a very low-keyed affair at least two years out of three and the site visitors should number about two; one, a senior RNPS officer -- one of the top people we have focused the divisional organization around in the preceding recommendation; the other a junior staff member -- perhaps a promising member of one of the functional branches of the RMPS, or perhaps one of the assistants to the seniors -- or alternatively, a senior outsider of the kind who could be expected to have a significant knowledge and interest in RNP and real involvement in it. Members of the National Advisory Council and the Review Com- mittee represent one class; program coordinators from other regions represent another class; former RNP senior staff members are a third class. People from other branches of HSMHA, even without prior experience in RNP, might constitute still an additional class. The assistant directors in charge of public relations and the collected functional branches constitute an additional very important class; site visit experience is an important part of what they need to do. VI-34 Arthur D 1-ittle, Inc Such visits should last only a couple of days. The initial agenda would include discussion of the draft document sent earlier with the object of devising ways of testing any dif- ferences that exist between the interpretation offered in the document and the interpretation asserted by theprogram coordinator or other members of the regional establishment. Another agenda item of great importance would be the investiga- tion of activities about which the site visitors were dubious, or had not yet learned much about. This suggestion is not meant to contradict the recommendation of the FAST Task Force that site visits be held triennially. We fully endorse the idea that a full-dress site visit could be useful and appropriate at three-year intervals. Our espousal is for an interim uRdating and data-gathering exercise in the context of a dialogue. One objective of the annual visit would be to set the stage for making a formal triennial site visit much more meaningful and credible than the site visit at its worst; namely, a "dog and pony show" in which the "Feds" are moved from one performance to the next in rented cars. There are other differences between the formal site visit and the kind of session just described. The "formal" site visitor(s) is explicitly charged with the responsibility of making substantive judgments. The only judgment to which the "informal" site visitor aspires is whether agreement has been reached between the region and RMPS on the existing situation in the region and its meaning. Another difference is that a formal site visit delegation has to include persons of various backgrounds, some having prior acquaintance with the region and some not. Regional officials, accordingly, are in a position to know very little about what is expected and what preconceptions the visitors bring. In the informal setting, the document that preceded the visit reveals the preconceptions ofthe authors. Furthermore, the aim is informality, thus permitting a mutual test of what is com- municated well beyond the depth possible in formal site visits. The aspiration is to conduct genuine dialogue. Before the visitors depart, they owe their hosts a description of their observations. This description might include a list of points to be incorporated in the written document as re- visions; it might be a complete reformulation of strategy or the visitors' perception of local strategy. Whatever form this feedback takes, it should reveal what the visitors are taking away with them as views on what the region is doing, how it is performing its work, how well it is doing its work, why and how it has chosen the goals it is pursuing, and what difficulties it faces. (3) Written feedback to.the region - The site visitors should next revise the draft document appropriately. They should certainly add their most recent observations on problems or soft spots VI-35 ArtilLir 1) l.ittle, Inc and make suggestions for dealing with these issues. At this stage the document should deal with the questions listed in Section C, Chapter V (Evaluation); that is, there are criteria available to judge the adequacy of the document itself.* The obvious thing to do with the revision is to send it back to the region. Whatever process the regional program coordinator then wishes to use internally is up to him. If the document still proves unacceptable to the region in its revised form, telephone discussions between the program coordinator and the principal evaluator may well be in order. But not too much'effort should be devoted, nor too much time allowed to go by, to workout differences. It is almost sufficient that the differences come to light and are clearly specified. The issue is to decide whether further on-site Evaluation is necessary, whether changes in wording could permit substantial agreement, or whether the region should be invited to prepare a separate interpretation for all or part of the document before concluding this phase of the dialogue. A copy of the document and notice of any important disagreements should then be forwarded to the Director of the Regional @ledical Programs for his decision and response. The Director may de- cide on further promulgation. Possibly the document itself or a summary should go to the National Advisory Council, to various officials in HSMHA, or to particular local regions in which, for one reason or another, the report looks as though it would serve a useful purpose. Alternatively, of course, the Director may want a revision made or further development undertaken before the document is distributed, or ask for additional tests of specific issues that go beyond mere verbal development. (4) "Negotiations" - Informal conversations between people in the region and people on the Washington staff should then take place. These discussions should be undertaken and moderated by the program coordinator and the senior RNPS staff member assigned to the region, with the object of shaping firm courses of action in selected areas that have emerged both from the on-site dis- cussions and subsequent revisions of the initial statement. This discussion should be more than a planning consultation; it should also lead to agreement on what assistance RNPS itself can or should offer-the region in trying to cope with any dif- ficulties under discussion. (5) Regional budgeting estimates - The annual document descriptive ofthe regional program has by this time reached a final stage. Enough agreement or clear enough disagreement should have been We continue to feel the need to note that these propositions are not direct criteria, but function as "meta-criteria." VI-36 Arthur D 1-ittle, Inc staff members from RMPS who carry out the evaluative site visits and conduct the scheme outlined here. The purpose of the work- shop is to evaluate regional evaluation procedures, and it should be so structured that the representatives of the coordinators can confront the evaluators. Properly done, such a confrontation enables one or both groups to come to a more mature understanding of crises or disagreements that may have occurred in the year preceding. The workshop should also take an overview of the gaps, deficiencies, and problems that the evaluations of the year pre- ceding may have revealed. From these discussions a number of results could be sought. They might include: 11 Judgments on methods of strengthening national-regional or interregional capabilities so that better consultant expertise becomes available to cope with endemic or epidemic problems; 'o Decisions on sharpening, softening, broadening, or nar- rowing the evaluative criteria or processes that emerge from behind-the central-regional dialogue; *- Consensual decisions of the workshop participants to recommend concerted consultation for or with specific regions in chronic difficulty; and o Consensual decisions to recommend conceptual consultation with or replacement of specific central "evaluators" (them- selves very likely participants in the workshop itself.) All these specific possible outputs from the workshop would have to be appropriately directed. Some of them might be channeled to the Director of RNP, some to specific program coordinators, some to specific Regional Advisory Groups or Boards, some to the Executive Committees of the coordinators themselves, or even to any interregional bodies that happen to exist in the RMP structure. b. Other Uses of Annual Evaluation Documents Both the process of preparing the annual evaluation and descrip- tive document and the document itself are keys to a number of other WES procedures. They would, as already mentioned, furnish materials for National Advisory Council agenda. In some cases, the descriptions themselves should be seen and discussed by the National Advisory Council; in other cases, the staff would base policy suggestions and priority concepts in part on the content of these documents. These annual descriptive evaluations would also provide "a way into" all the regions for the public information specialists on the national staff. These documents could enable the national public relations staff to be even more active than it now is in seeking out and identi- fying activities either to publicize- or clarify. VI-38 Arthur D Little, Inc. reached that the region itself can then prepare budget estimates for the next fiscal year, after specifying as well as possible what their monetary requirements will be. Such a budget document can easily take the form of a program budget because it is pre- pared against a program evaluation and implicit program plan. Supporting budget breakdowns by disease category, project, or appropriation account will also be administratively required. Obviously, RMPS'would have to establish some ground rules that would take account of the continuing preparation of these docu- ments on an annual cycle that would overlap the annual federal budget cycle in an awkward way. Two obvious options present themselves: (a) the final preparation of regional budget submis- sions could be deferred in those regions where the site visit takes place at the wrong time of the year, so that all the re- gions could turn in their budget estimates on the same date suitably documented with revisions updating the evaluation document in terms of changes that happen in the period between the preparation of the site evaluation document and the budget deadline; or (b) budget revision made just before the deadline could be permitted to regions whose annual cycle was out of phase with the federal budget cycle, and whose budgets were submitted earlier. Although at the time this report was drafted we had seen none of the "anniversary review" applications so far submitted or in preparation, we hoped that responses prepared by regional core staffs to the general questions about "program" and "pro- gram" strategy would elicit much of the kind of material described in this discussion. Subsequent events confirm our belief that the hope is worth retaining. The regions seem to be making attempts to cope with the issues. The practice of preparing applications, descriptions, and justifications at the regional level may persist, consistent with other grant programs. The region could incorporate the central-regional dialogue document in its application, suitably supported, or could provide an alternative. Preparing the documentation locally provides a basis for accountability ex- pressed in the words of those who are accountable. Including the words and judgments of outside evaluators means that the outsiders' perceptions of the region have to be made explicit. People within the region then have a much harder time avoiding the external perception if it is presented in the words of outsiders. (6) Annual assessment - A workshop should be convened annually by the Director of RNP to examine the process just described. It should include representative program coordinators chosen by their Executive Committee. (It might be sensible for the Executive Committee itself to be the group representing the co- ordinators.) Other participants should include the principal VI-37 Arthur 1) l,ittle, Inc. The experience of the senior people in conducting the site evalua- tions and drafting and developing the documents would also become very important background for choosing, phrasing, testing, and inter- preting items on special information-gathering instruments that would still be required. The various functional branches of RNPS would still have to beget their own specialized views of certain issues and gather information responsive to questions from on high. In other words, the presence of 10 to 12 senior RNPS people who among them know all 55 regions in considerable detail would enable'these questions to be put with maximum sophistication, so that the chance of getting real information would be maximized. 3. Management Information System Since the spring-of 1970 we have been aware that a new effort was underway to design a national management information system for RNP. Our contract is monitored by the Assistant Director for Planning and Evalua- tion. Most naturally, our contacts with this process have been through Planning and Evaluation personnel. We have attempted to transmit ideas about several broad guidelines through the conversations we have had. o Keep the system as simple and as inexpensive as possible. This means clearly recognizing that a lot of information does not need updating very often, because it does not change very fast and because it does not really affect things much when it does change. Having the names and addresses of all RAG members in the country recorded in computer memory would, for example, seem to be "gilding the lily," and expecting such information to be updated more than annually would be quite unnecessary. For almost all of this kind of information the "anniversary review" application should suffice, and the appli- cation documents themselves should constitute all the management information system required for these classes of data. o It is a real issue whether computerization has a significant place with respect to an RMP national management information system. We could imagine that certain aspects of current financial analysis ought to be handled by a machine. The several cycles of expenditure, obligation, and budgeting are hard to keep straight. Knowing how much money is available from what point of view at what points in the RNP system is not easy. We hope that RMPS has gone through the arduous exercise of revising its various accounting systems, so that they can produce information that is up-to-date and accurate for management purposes. Having completed that process, managers can then rationally decide whether actual computerization would be of any benefit. o What is most important about RMP for management purposes is either nonquantitative or reduced to trivialities if presented in purely quantitative form. The strategies and processes actually in being or contemplated in the several regions is what we are referring to. While brief summaries on these matters can be recorded VI-39 Arthur 1) Little, Inc in a central system and probably should be in such a system to pro- vide a management check on whether anybody in RMPS is able to ex- plicitly answer the relevant questions, the most important item in the management information system in terms of "strategy" is "who in RMPS knows what about which regions." In this sense, the heart of the real management information system is constituted by the top-level people who have direct connection with the regions in terms of the organizational form suggested in an earlier recommendation. o The information which managers want is that which relates to emergent problems. Emergent problems are those which have not yet been subjected to a routine. Ordinary management information systems are quite incapable of handling data that are not quite completely subjected to a routine. It is in the nature of RMP that a good part of the information wanted for next year's planning, for next year's justification, and for dealing with today's questions from on high will simply never be found in any mechanized information system now devised. A large number of items will have to be produced by very specific investigations. While the functioning of a com- prehensive national management information system could somewhat reduce the.number of special questionnaires and information-gathering activities of other kinds initiated by the Division of Regional Medical Programs, it would take much specific and concrete detail to convince us that the reduction in these special investigations would result in cost-savings sufficient to defray any significant fraction of the costs of building and operating the management in- formation system. vi-40 Arthur D little, Inc t APPENDIX A Arthur D Littk Inc APPENDIX A PROGRAM PLANNING AND BUDGETIN%'A FOR RMP Our contract with the RMP requires that we deal with the applica- bility of Program Planning and Program Budgeting 'to the Regional Medical Programs. Program Planning and Budgeting was developed in the Department of Defense under Robert McNamara, and in 1965 President Jckhnson directed that other administrative departments undertake to apply the concept. In the interval, many attempts have been made and much has been learned about the possibilities and limitations of "PPBS" (as it has become known.) In this brief paper, we set forth our views of the applicability of PPBS to RNP. 1. 14HAT PPBS IS AND ATTEMPTS TO DO PPBS groups activities that have similar goals or are expected to yield closely related results. Essentially, when applied to government it sorts'activities of a department by program elements to facilitate comparisons of similarly directed programs across jurisdictions. PPBS identifies program goals and matches all activities aimed at achieving them. Thus, within a department, e.g. PYPS, programmed attacks on heart disease, cancer, or stroke may be grouped in such a way that the costs and results of different programs can be directly compared. At the HSMHA level, programs in several RMPs would be grouped with programs in other HSMHA departments when the objectives of such programs were essentially similar. In form, PPBS consists of a long-tem strategic manpower and finan- cial plan and a short-term budget or control instrument for authorizing the current year's expenditures of manpower and money against the strategic plan. Typically, the programming to implement strategies covers several years beyond the current year, recognizing that the strategies may require long investments before payoffs are realized in concrete terms. PPBS does not deal with the ordinary budgetary controls with which we are all familiar. For example, it has no direct bearing on controlling efficiency, costs, selection of personnel, or performance measurement. As originally conceived, PPBS assumes that all elements of a program can be converted directly into monetary terms. In this conception, once programs have been identified and their elements determined, all operating costs can be assigned to them in such a manner that all costs are accounted for. For anyone familiar with cost accounting, it is quickly evident that this assumption requires the allocation of joint costs where individuals work on more than one program or have some supervisory or administrative role with respect to more than one program. Arthur D Little, Inc 2. LIMITATIONS TO PPBS FOR MANAGING SOCIAL PROGRAMS PPBS was designed for use in a setting where central control was strong, goals readily definable, and progress measurable in quantitative terms. Its applicability is severely limited in the arena of social programs, subject to multiple control, political pressures that often seem irrational, evolving goals, and progress that takes intangible forms. In an ideal system, PPBS starts with a framework of all activities in a particular department or bureau. The framework consists of multi-year plans, multiyear programs, and a budget for at least the current year. Fitted into this framework are manpower, facilities, and fund requirements to carry out the plans and programs. Since the objective f-s to compare programs across jurisdictions, PPBS as presently conceived contains an extreme centralizing bias. In a complete federal system, theoretically the President would be in a position to make decisions of manpower and funding support across all departments for programs having a common social or economic objective. Within departments, the department head is given a similar role. For him choices are seen as being available from all sub-departments, bureaus, regions, and so on whose function is to meet common objectives. In miniature, at the lowest level (let us say, for example, within an individual RMP) the Executive Director or Coordinator would have complete power over balancing all of the programs under his supervision and would be expected to group them in such a way that he could make direct choices among them. In the social arena such commonness of objectives is not always clear, except in the broadest terms. For example, within an individual there may be several cancer programs with the objective of reducing mortality or morbidity from cancer; but the approaches to this accomplishment may be so different as to make the programs quite uncomparable. Where the effects of social programs may be far in the distance, it may be impossible to judge in the early years how effective one program is compared to another in reducing mortality and morbidity. Often the only directly comparable elements of the program are the amounts of energy -- man- power and money -- that go into two or more programs aimed at the same long- term objectives. PPBS is essentially an evaluative mechanism; it may be prospective or retrospective. When it is prospective it takes a starting point and lays out steps for achieving some different ending point. Since under PPBS programs can be measured only in quantitative terms, there must be a quantitatively re- cordable starting point as well as a quantitatively recordable ending point. A peculiarity of social programs is that often quantitative measures do not exist; and even more frequently, data-gathering systems for pulling together such quantitative measures as do exist are not themselves available. This is especially true in medicine, where even the most basic measures of the health status of the nation are primitive and weak. 2 ArthurDLittle,inc In social programs of variable product, and subject to few fixed routines, prospective budget estimates are difficult to make because so many of the detailed expenditures are truly unpredictable. So far as retrospective analysis is concerned, variable assumptions allow wide scope for reaching a broad range of conclusions. If contesting groups can agree on these assumptions, the retrospective program analysis may yield important information; but when this happens it is most likely superfluous. Any system of management which distributes largesse to some while withholding it from others will face manipulation by those who wish to shar@ in it. One of the features of social programs is that they carry with them political "benefits" and "costs" which may well be of overriding importance in determining what is done. Thus, any decisions on public issues will distort the material going into PPBS to match political reality at the expense of what outside observers might czCfl rationality. It is difficult for rationality, even when ostensibly supported by data, to have much of an impact in the social and political arena. The PPBS analyst will in the first place always be suspected (often correctly, in our experience) of reflecting the very special rationality of those he regards as his bosses. Secondly, the prospective program budget is just as dependent on "cooked" estimates as any other form of budgeting. In fact, the difficulty of eliminating bias may be the greatest obstacle to the application of PPBS to social programs. Even the explicit statement of goals can be a problem with social programs. Since such programs are essentially political, the publication of goals can at times set in motion political forces that foreclose their realization. Thus, the identification of the real purpose of some programs can be impossible, and comparisons with other programs aimed in the same general direction are blocked off. In addition, goals change. One of the peculiarities of public programs is that, though they may start out with simple concepts, the simpli- city begins to vanish as political reality becomes evident. Given that al- most no program that affects people can live without appropriate recognition of all the kinds of people who are affected by it, firm adherence to stated goals is often just not possible. In the long run, the objectives of a public program grow, develop, and shift as the awareness of political forces reaches those in responsible positions. As Aaron Wildavsky*said, "It is difficult enough to adjust clear-cut objectives'to resources in the present or near future. The task is im- mensely complicated by considering future states of affairs about which Wildavsky, Aaron, "The Political Economy of Efficiency," The Public Interest, No. 8, Summer 1967 3 Arthur D Little, Inc there is a huge amount of uncertainty. What used to be constants become variables; little can be taken for granted." Further along, he quotes Quade as saying, "Systems analysis is associated with that class of problems where the difficulties lie in deciding what ought to be done -- not simply how to do it -- and honors go to people who . . . find out what the problem is. PPBS is in a very early stage of development. It has not yet reached (and may never reach) a point where it can give full weight to in- tangibles which do not lend themselves readily to monetary expression. We are a long way from the time when errors prevented by PPBS wilL outweigh the disasters of excessively rigid reliance on PPBS perspectives and evalua- tion techniques. Furthermore, although PPBS practitioners have been aware from the beginning that many defJsions are outside their scope and must be made on the basis of intangibles rather than "cost benefit analysis," case- hardened program managers (veteran survivors of many campaigns conducted in the name of economy) have tended not to believe that PPBS analysis would confine itself within the limits of its applicability. In his testimony before the Subcommittee on National Security and International Operations, Thomas Schelling** said, "PPBS if too much focused on costs and other 'tangibles' may even divert attention from those elements of a decision, sometimes dominant elements,that cannot be translated straight- forwardly into budgetary terms. There is consequently genuine concern that PPBS and other techniques of management that are essentially budgetary or quantitative may be of not only less positive value when applied to foreign affairs but even, through their tendency to divert criteria and to elevate particular kinds of analytical competence, of positive harm." Schelling was talking about the prospects of applying PPBS to United States foreign affairs programs. He concluded that the spirit of PPBS for comparing the total results of different programs across all countries with which the United States has active relationships was desirable, even though most of it would have to be non-quantitative. He felt that the ap- plication of PPB S philosophy to objectives in Jordan or India might call attention to the sacrifice of certain objectives in Egypt, Algeria, Israel, 6r Pakistan, and that the process of making judgments about international policy should bring such relationships to the fore. We heartily agree that to the extent the spirit of PPBS can be applied in such a way as to put de- cisions in the most complete possible perspective, it should be applied to all social programs. It is not the spirit of PPBS that we find poorly adaptable to social pro'rams; it is the violation of that spirit which becomes inevitable 9 in some contexts if a methodology is employed which expresses everything in monetary terms. Strictly construed, PPBS works best the closest it comes to straightforward economic choices. If the costs and benefits are directly Schelling, Thomas C., Testimony before the Subcommittee on National and International Operations., U.S. Senate, 90th Congress, First Session, 1968 4 Arthur D Little, Inc comparable, and if the political implications of two programs are directly comparable, then PPBS can be an effective and even critically important aid to reaching decisions. However, as the costs become less comparable and include increasingly complex political elements, even though the objectives may be comparable, PPBS becomes indeed a weak method for reaching decisions. This is even more compelling where, as in RNP, costs include those of per- suading other organizations to change their practices and procedures. At that point in the analysis of a.social system one 'must include the changed, costs of the organizations not directly controlled as well as those of the persuader, which becomes quickly impractical. Quoting Schelling*once more, "The budget does not yet exist to which PPBS might be applied in the field of foreign affairs." We can add that in our experience withvthe RMP we did not see a budget to which PPBS might readily be applied except in the most super- ficial sense. In our chapter on Evaluation we make it quite clear that the most important accomplishments of @ can best be evaluated through dialogue between those in a position to carry out programs and those who are responsible for seeing that all of the programs are worthwhile. Very little of what we describe as appropriate dialogue can be converted directly into quantitative measures. Dialogue, however, fits the spirit of PPBS rather well because it insis'ts on identifying the real purposes of every activity and requires that progress against identified goals be carefully monitored. In the sense, then, that programs must be compared to each other in terms of the ultimate goals of RMP, the kind of dialogue described in our Evaluation chapter is philo- sophically compatible with PPBS. In the dialogues between responsible RMP executives and their subordinates, progress must be compared in such a way that the intelligent executive is well equipped to make choices from among several programs. What is not possible, we believe, is the translation of all programs into comparable monetary terms without the loss of much that is at the heart of the RMP. The danger to be avoided at all costs is that when an organization cannot measure what it wants, it may begin to want what it can measure. 3. THE APPLICATION OF PPBS TO RMP Public Law 89-239 clearly places improvement in the general health of the nation as a primary objective. For some years the Department of Health, Education, and Welfare attempted to apply PPBS to the health scene. As Drew** reports about early PPBS efforts in HEW, "The study of disease control was similarly, but somewhat more surprisingly, beset by quandaries about what as- sumptions could be made. There is not even material, as William Gorham points out, to make a documentable argument that people who receive regular medical attention are healthier than those who do not! It is known that the wealthier are healthier than the poor, but how much of this has to do with doctors, and how much with characteristics of the entire environment, no one can honestly say." It has become widely believed by those in public health that the nutri- op. cit. Drew, Elizabeth B., '-HEW Grapples with EPBS", The Public Interest, No. 8. Siimmer, 1967 5 Arthur D Little, Inc tion of people has more bearing on their health than the amount of direct medical care. Thus, if the medical profession cannot pinpoint the causes of good health, then we are in a very weak position to make quantitative com- parisons between programs whose results in terms of actual improvements in health will not be apparent for many years. We call attention to the levels of evaluation described in Chapter V. In an evaluation system that must rely strongly on peer judgments of the effectiveness of different programs, it makes no sense to assign to those judgments the monetary values which are so much a part of PPBS. Public Law 89-239 sets as an independent objective "cooperative relationships." Implicit in this is the assumption that if the various ele- ments of the medical profession and related service organizations can be brought together to work out improved ways of extending health care, there will be very real benefits. How does one measure cooperative relationships in monetary terms? Clearly, when people do cooperate and agree to accomplish something it may result in direct changes in costs with presumed changes in benefits to follow; yet there is no realistic way of estimating more than very crudely in advance -- or even determining after the event -- the extent of either kind of change. As for the cost shifts in medical practices and in the relationships among the professional groups in medicine, these can fan out in ever-spreading effects in such a way that no accountant could possibly hope to keep track of them. It is a peculiarity of RMP that large numbers of people are brought together for discussions (and even for taking action) whose time cannot realistically be evaluated. We call attention once more to the fact that the principal mechanism available to RMP is facilitation -- bringing about improvements through the action of other people. It is at least theoretically possible to attribute the payroll and other costs of a particular RMP to several different facilita- tion programs. An accountant would quickly recognize that almost all of such costs would have to be attributed by allocation rather than by direct attri- bution. So much for trying to compare facilitation costs. At the benefits end, the problem is not nearly as susceptible to rational explanation. As has already been said, there is no certainty that an "improvement" will, in fact, raise the status of health care in the nation. Even if objectives thought worthy are realized, there can be no absolute conviction that one result is better than another. At best we must lean on that crutch called "medical sanctions" -- a non-quantitative support at best. This does not mean that RMP should dismiss entirely the type of thinking reflected in PPBS; what it does mean is that PPBS cannot be applied literally on any major scale. Though strategy and facilitation are at the heart of a successful RMP, there is plenty of room for applying thoughtful comparisons at both the regional and central levels. RMP can identify program elements. They may be single projects; they may be groups of projects aimed at similar targets; they may be groups of projects which together constitute a strategy for building new relation- 6 Arthur D Little, Inc ships, with individual project results of only secondary significance. Activities may be arranged by geography so that the effects on different areas, municipalities, or the rural countryside can be compared. Thus, it may be possible to highlight the comparative effects of programs on different kinds of populations across regions. RMP sometimes arranges its information by the institutions or professions involved, i.e., medical schools, teaching hospitals, community hospitals, voluntary agencies, punlic agencies, etc. This makes it possible to compare results involving these different kinds of institutions with results involving other kinds of institutions. Activities can be arranged by type, through a spectrum running from research to demonstration, from prevention to rehabilitation, from plan- ning to action. Thus, prdirams falling at either end of these ranges may be compared with each other. Finally, they may be arranged by how they are approached and how structured. For example, is the principal activity handled by core staff, by the RAG, by a committee, or by a task force? Is the mechanism a project, a contract, or just facilitation? By looking at all of these, a coordinator can begin to learn which groupings of what institutions under what structural form produces the most effective results. All of these arrangements of projects and programs make it pos- sible for a coordinator to think more clearly about the purpose and effect of what is going on in a region. But it is important to remember that strategy is the name of the game. Phased activities cannot necessarily be time-predicted, nor their ultimate scope defined. Dependencies, and hence sequences, may be unclear until the strategy is applied. Priorities will shift in midstream. Accomplishments in the short term may have very little traceable relationship to accomplishments in the long term. In short, PPBS is a useful attitude for an executive in RNP to apply to his responsi- bilities, but in any 'strict methodological sense it has little or no bearing. The value of PPBS as a management tool depen ds on the illingness :w of its practitioners to use it introspectively. When imposed from the out- side, PPBS is soon interpreted (as it was from the earliest days in the military departments in the early 1960s) as being a way to direct huge amounts of staff time from more important work, thus keeping subordinate organizations off-balance. Thus viewed, PPBS becomes just another set of devices by which people attempt to conceal from one another what is really going on. Good program analysts know that they must somehow make the idea of program analysis one that is useful from the point of view of those who "own" the data required, if they are not to fall into these difficulties. In the case of RNP, this means that there has to be agreement on what the basic program elements are. Since there is almost no intelligibility in the concept of program elements applied to RMP nationally, "program bud- geting" is not very meaningful at the national level. Even such gross dis- tinctions as those between'core staff activities and operational projects do 7 Arthur D Little, Inc not constitute national program elements in any sensible way, because nationally there can be no real attribution of elements to the levels of change on which benefits may be anticipated (i.e., levels in the planning process, institutional relationships, access or quality of care, and health of people.) These changes happen locally, not nationally. Furthermore, .most of the interactions take place intra-regionally. Aggregations of the data at national levels reveal the total quantities involved, but cover up the significance of sub-regionally important effects. The only approach we can see to program budgeting nationally would be one that responded to the question, "How much is it worth to lubricate the processes of voluntary systems transformation in the health and medical care field, anddoes the RNP way seem to give you as much as some other way?" Given the local inter- actions and interdependencies of RMP, CHP, community mental health planning, hospital facilities planning, etc,., this question can be answered with only the utmost uncertainty. It may @e worth a few days' work on the part of an imaginative researcher; but what can be learned beyond the first rough formulations is very problematical, for all the reasons laid out earlier. On the other hand, in our view an equally imaginative self-motivated program analysis on the part of regional program coordinators could be very useful. Actually, program analysis cannot make much sense until the coordina- tor can tell you what his problems are and what he is trying to do about them. Answers to these two questions constitute a statement of regional strategy (see Chapter V) and will provide a perfectly adequate framework for such rudimentary program analysis as has value for the coordinator in his role as manager. The activities and processes under his direction have costs. The PPBS issue is to test consistency between these costs and the priori- ties of the region. Thinking in program budget terms for a few hours a year more than justifies itself, for example, if it reveals or makes ex- plicit to a program coordinator that almost none of his or anybody else's resources are being devoted to dealing with a high-I)riority problem. .Approached from this point of view, program budgeting is an attractive device for regional program coordinators to use in RMP; RMP is such a fluid program that it would benefit from a periodic review and readjustment of its "accounting" categories. Accounting of a conventional sort, systematically carried out in detail, of course, must continue. 4. COST/BENEFITS ANALYSIS IN RMP The ADL-OSTI contract requirements go beyond PPBS in inquiring in- to the economics of RMP and its costs and benefits relative to those of other ways of accomplishing similar objectives. Early in our study, DRMP executives agreed with us that these considerations are not susceptible to meaningful mathematical analysis. However, we feel constrained to elaborate on this conclusion. There are three on-going programs in HSMHA all having the over- all objective of improving the effectiveness and efficiency with which health care is provided to people. They are The Regional Medical Program, Comprehensive Health Planning, and Health Services Research and Development. 8 Arthur 1) l,ittle, Inc Each of these programs makes its own unique contribution to the common ob- jective, but all of them shade into areas where any two or all three might be considered as appropriate tools for reaching toward the objective. RMP is unique in calling on the medical professions (with help from lay people) to design their own programs for improving health care. It is generally regarded as responsible for emphasizing and upgrading the quali of care. CHP is unique in giving prominence to the consumer and his intere@t in achieving both availability and accessibility of appropriate care for all people. It tries to align consumers, state health and planning officials, and concerned medical interests behind rationalization of the health system. Health Services-i&D is.unique in sponsoring experiments with promising new ways of providing health care. Unlike the other two programs it is not subject to policy guidance by councils of private citizens. All three programs have budgets, and all three could undoubtedly Find ways of doing things that would use more funds than they are likely to get. -We know that in the case of RMP, the National Advisory Council has approved projects during the last two fiscal years which the program could not support. On occasion, more than one of the three programs might engage in projects addressed to the same problem. Take the North Carolina stroke program as an example. (See Addendum I to Chapter V.) Under the guodelines of the three programs this might quite reasonably have been sponsored by RMP, CHP, or HS R&D. The emphasis would have been somewhat different in each case: under RMP it was on the dissemination of advanced medical knowledge through cooperative arrangements; under CHP the attention might have been focused on rationalizing the flow of patients to the nearest point at which suitable expertise might be found; under HS R&D the emphssis might have been on experimentation with differently constituted professional stroke teams. Who is to say in advance which of these approaches was likely to yield the most favorable cost/benefit ratio? The hospital network in Western North Carolina (described in Ad- dendum ]to Chapter IV.) might have been advanced through either RIT or CHP. There is little evidence to indicate which program might have accomplished the building of a more effective hospital network on the most favorable cost/ benefit terms. Another example is the rural health care project described in Ad- dendum 2 to Chapter IV. RMP happened to be on the spot when it could help to develop promising new relationships. CIIP might well have designed an ap- propriate program for achieving a similar result. Even HS R&D might have seen value in a pilot model for getting better health care to sparsely populated rural areas. One would be hard put to it in advance to decide in cost/benefit terms which was best suited for the job. Decisive cost/benefit choices -- or even satisfying approxima- tions -- cannot be made among these programs with assurance. Results are 9 ArtilLir I) I.it tie, fric. difficult to compare on any absolute scale, and true costs are usually beyond the range of accurate calculation. What comparisons can be made are possible only in retrospect, when all the facts are in. These programs are all in the developmental stage; they have had only limited experience from which to learn how much of what kinds of energy will produce what degree of result, how fast, at which levels of the health system. At best, judg- ments based on very limited data will have to be relied upon when choices are, to be made in advance. Unless the choices, expressed in judgment terms, are orders of magnitude apart in prospective costs or benefits, there can be no meaningful comparisons between them. All of our comments in this Appendix about the dominance of poli- tical considerations in trying to apply PPBS to RNP applies equally to all other formalized systems of cost/benefit analysis. Thus we conclude that analyses of this sort can serve-no useful purpose beyond that of making dis- ciplined comparisons of whatever programs seem to be closely related in purpose, and then only in terms of judgments based on controversial data. 5. BIBLIOGRAPHY Black, Guy, The Application of Systems Analysis to Government Operations, New York,, Frederick A. Praeger, 1968. Churchman, C. West, The Systems ApproachNew York, Dell Publishing Co., 1968. Cleland, David I. and King, William R., Systems Analysis and Project Manage- ment, New York, McGraw Hill, 1968. Dorfman, Robert (ed.), Measuring Benefits of Government Investment, Washington, D.C., Brookings Institution, 1965. Drew, Elizabeth B., "HEW Grapples with PPBS," The Public Interest, No. 8, Summer, 1967. Frankel, Marvin, Federal Health Expenditures in a Program Budget," Chapter 8 in Program Budgeting, David Novick (ed.), Cambridge,, Mass., Harvard University Press, 1965. Froomkin, Joseph, "Cost/Effectiveness and Cost Benefit Analyses of Educa- tional Programs," Socio-Economic Planning Sciences, Volume 2, 1969,, pp. 381-387. Grosse, Robert N., "Problems of Resource Allocation in Health," in The Analysis and Evaluation of Public Expenditure: The PPB System. A Compendum of Papers submitted to the Subcommittee on Economy in Government of the Joint Economic Committee, U.S. Congress, Volume 3. Hartley, Harry J., Educational Planning, Programming and Budgeting, A Systems Approach, Englewood Cliffs, N.J., Prentice-Hall, 1968. 10 Arthur D Little, Inc Hatry, Harry P., Criteria for Evaluation in Plannin& State and Local Programs. State-Local-Finance Projects of the George Washington University, Washington, D.C., May 1967. Hitch, Charles J., "What Are the Programs in Planning, Programming, Budgeting?" Socio Economic Planning Sciences, Volume 2, 1969, pp. 465-472. Joint Economic Committee, 90th U.S. Congress, First Session, Hearings of the Subcommittee on Economy in Government, "The Planning-Programming- Budgeting System: Progress and Potentials." I- Klarman, Herbert E., The Economics of Health, N.Y. Columbia University Press, 1965. Lyden, Fremont J., and Miller, Ernest G. (eds.) Planning Programming, Budgeting: A Systems Approach to Management, Chicago, Markham Pub- lishing Co., 1967. McKean, Roland H., Efficiency in Government ThrouSh Systems Analys'SL With Emphasis on Water Resource Development, New York, John Wiley and Sons, fnc., 1958. Monograph Series in Planning Programming Budgeting for City State County Objectives,.State-,Local.Finance Project of the George Washington University, Washington, D.C., 1967. Morse, Philip M. (ed.). -Operations Research for Public Systems, Cambridge, Mass., M.I.T. Press,, 1967. National Center for Health Statistics, Conceptual Problems in Developing an Index of Health, U.S. Public Health Service Publication No. 1000- Series 2. No. 17, U.S. Govt. Printing Office, Washington, D.C., 1966. Novick, David, (ed.) Program Budgeting: Program Analysis and Federal Government, Cambridge, Mass., Harvard University Press, 1965. Olson, Mancur, Jr., "Social Indicators and Social Accotmts," Socio- Economic Planning Sciences, Volume 2, 1969, pp. 335-346. Packer, A.H., "Applying Cost-Effectiveness Concepts To The Community Health System," JORSA, Volume 16, No. 2, March-April 1968. "Planning Programming Budgeting System: A Symposium," Public Administration Review, Volume XXVI, No. 4, December 1966, pp. 243-310. Rice, Dorothy P., Economic Cost of Cardiovascular Diseases and Cancer, 1962. (Health Economics Series No. 5) U.S. Public Health Service, May 1965. Saunders, B.S., "Measuring'Community Health Levels," American Journal Public Health 54, 1063-1070 (1964) 11 Arthur D Little, Inc Schelling, Thomas C., Testimony before the Subcommittee on Nation@il and International Operations, U.S. Senate, 90th Congress, First Session, 1968. Schick, Allen, "The Status of PPB inthe States: An Interim Report," Brookings Institution, Washington, D.C., presented at the 35th National Meeting of the Operations Research Society of America, Denver, Colorado, June 1969. U.S. Bureau of the Budget, "Planning-Programming-Budgeting," Bulletin No. 66-3, October 12, 1965, and Supplement to Bulletin No. 66-3, February 21, 1966. t' University of Pennsylvania, Fels Institute of Local and State Government. Government Studies Center. "General Design for an Educational Plan- ning-Programming-Budgeting System." Philadelphia, Pa., June 28, 1968. Wildavsky, Aaron, "The Political Economy of Efficiency," The Public Interest, No. 8, Summer 1967. 12 Arthur D Little, Inc. I lr APPENDIX B Arthur 1) Little, Inc- APPENDIX B OTHER REGIONALIZATION This paper was commissioned as an attempt to answer the question,. "What other regionalization programs have something to offer to RMP?" RMP led us gradually to the conclusion that there is little that is posi- tive to be learned from other programs, either because the circumstances are different in clearly significant ways, or because the programs failed for reasons that RMP cannot control but is already taking account of. So the summary paper presented here is essentially a corroboration of certain themes already well understood in most parts of RMP. There are Several Reasons-for Regionalizing In a large-scale and highly involved society, regionalization of necessity becomes a way of life. There are several reasons for re- gionalizing activities, and we find each of.them accounting for particular regional breakdowns of large programs. The most common reason for regionalizing is to break up admini- stration into manageable units. For example, most national industrial marketing organizations are broken up into regions. Most large industries separate their manufacturing into a number of factories. In fact, it be- comes apparent upon investigation that most businesses operate with the smallest factories that can be regarded as above the critical mass level required for efficiency. Clearly, for them the advantages of close direct management-employee relationships outweigh the economies of scale once this critical size level has been reached. The political management of our states is usually regionalized in one way or another. A particular example of breaking up the administra- tion of a state program is the university system in California. Obviously, it was felt that beyond a certain size several universities were-better than one. Another common reason for regionalizing is to diffuse knowledge under controlled conditions. The DeBakey model for heart, cancer and stroke was conceived in this mode. So was the Agriculture Extension Ser- vice, which is so often given credit for spreading the high technology that made our agriculture famous worldwide. While closer scrutiny of what happened in agriculture suggests that seed and equipment and chemical companies have had significant and often crucial roles in spreading the latest in agronomy, it is notable that they too regionalized their ap- proach in spreading the technology. The State Technical Services Program was set up regionally to try to bridge the gap between the high technology in certain industries and the less advanced industries which presumably lacked access to it. B-1 Arthur 1) l.ittle, Inc The Volkswagon Company brings mechanical supervisors to its headquarters from regional service organizations throughout the world for training, which they can then pass on to their fellow mechanics. Regionalization is sometimes applied for the purpose of accom- modating operations to local conditions, interests, and prejudices. The several Corps of Engineers' water basins are a case in point. Many national magazines today publish several regional editions to match the divergent interests of audiences in different parts of the country. Most religious organizations are broken up into regional groupings to recognize the special social environments in which they operate. The Federal Reserve Bank system is divided into regions which are based on national economic sub-groupings of the country. Sometimes a regionalized approach is applied to facilitate quick response of the parent organization to feedback from the field. The MM Corporation in Minnesota has perfected a system which "regionalizes" new products it is hoped will grow into entire businesses. It sets up a small group of executives in a distinct organization given the product idea and the problem of how to get going. (company support is afforded it through a start-up phase. By giving its managers autonomy and personal recognition if they succeed, MMM hoped to get the product off to a better start than would be the case if it got buried in the larger corporation. The Alcort Company, manufacturers of the famous Sailfish, for several years after starting up in business kept close contact with individual users scattered all over the country, and even out of the country, to learn just how the product was performing in different environments. Regionalization that accommodates to local conditions and re- gionalization that seeks to facilitate feedback from the field have other advantages running in their favor. They may provide a vehicle for co- opting local power groups so strongly entrenched that they could thwart any objective if their enthusiastic support were not won. They usually also take advantage of local resources, leaning on pride, initiative, imagin- ation, and self-interest to widenthe horizon of what is possible beyond what any central body would be likely to construct. Tyrical Models for Regionalizing The most common regionalization model is what we refer to as the center-periphery model. This is a model in which there is a 'policy- setting center fanning out intelligence, instructions, and usually training to a dependent periphery made up of a number of branches. Returning to the center from the periphery is sufficient information to permit the center to keep control, and to make decisions that are in touch with reality. Almost all industrial regionalization that has come to our attention takes this form. Success depends upon the power and authority of the center over all the elements needed to accomplish whatever tasks are set for the periphery. B-2 Arthur D Little, Inc Another common shape for regionalization separates the country intoits natural geographic units. Natural watersheds of economic activity are familiarly used in grouping activities that have very little economic or political connection. Often rivers and mountains have a considerable bearing on the shape of geographical regions. The third most common breakdown is along political or social lines. These are often based on historical divisions sanctioned by law, such as the division of the United States into its 50 states. While social boundaries are often not entirely synonymous with political ones, there is the tendency for them to be sufficiently alike so that these boundaries can serve the purpose as a matter of convenience. Some Examples of Regionalization of Social Programs Long before the Regional Medical Program was enacted into law there were activities of a regional nature within the health system. The first in this country of any consequence was the Bingham Associates program, embracing the state of Maine, and later parts of Massachusetts and Connecti- cut. The Bingham Associates program began with a continuing education alliance between the Rumford Community Hospital in up-state Maine and Dr. Pratt's Boston Dispensary. By the time the RNP came into being, the Bing- ham Associates incorporated some 61 hospitals in three states, and in ad- dition to the Pratt Clinic, the Tufts-New England Medical Center and the Tufts University School of Medicine. Residents from the center rotated through these hospitals, and patients needing special diagnostic workups were moved from the periphery into the center for special diagnostic workups and treatment. Some time after the program had been initiated, Maine itself was divided into two regions, one around the Lewiston Hospital and one around the Eastern Maine General Hospital in Bangor. These hospitals filled the role of referral centers within the State to screen and pass on cases that were of suf- ficient complexity to justify their being transported to Boston. The Bingham Associates program has had its ups and downs over the years, and sometimes some parts of the program have operated better than others. l'uition-free post-graduate courses in the Tufts-New England Medical Center for physicians, nurses, technicians, and administrators have progressed continuously but unevenly in their acceptance by those they were designed to help. Experts in all of these areas have been used for consultation in the hospitals. Residents from Boston have moved through the larger hospitals. It is notable that though the program has not always been embraced with unqualified enthusiasm, it is still in existence nearly forty years after its founding. Surgeon General Thomas Parran in 1944'recommended to the Senate Subcommittee on Wartime Health and Education, legislation for Federal grants in aid to the states for construction of hospitals. His proposal incorporated a condition that there be prepared in advance statewide plans based on need and priority. The hope was that there would be a country- wide system of regional medical networks modeled after the Bingham Associates B-3 Arthur 1) I.ittle, Inc. program. Somewhere between Dr. Parran's recommendation and the effectua- tion of what became the Hill-Burton Act, regionalization of any formal kind diminished to relative insignificance. The country was provided over the succeeding 25 years or so with a very large number of small hospitals filling what had been voids, largely in rural areas. The re- quirement prevailed that there had to be a state plan, but regionalization as such was never a significant consideration. The Kellogg Foundation in 1950 tried to introduce a regional scheme in a rural section of Michigan. Three isolated communities were given aid to join in a common effort represented by the construction of small health centers in each of the three towns, combined with'an affilia- tion arranged with two larger hospitals in some distance from them. Re- gionalization was abandoned after a few years, though the small health centers survived. Apparently,-neither the public nor members of the health professions had been adequately prepared in advance to think through the full import of regionalization. Local pride could not be adequately dealt with, andthe arrangements among the hospitals were abandoned. Another regionalization scheme in Rochester, New York, has sur- vived right up to the present with considerable change over the years. Sup- ported in 1945 by the Commonwealth Fund, the Rochester Hospital Council served as a center through a new unit called The Council of Rochester Regional Hospitals. The two councils were merged into the Rochester Re- gional Hospital Council in 1951. The program became self-supporting in 1954. Thirty hospitals in 11 counties comprised the Regional Council, and they are joined through the Council with coordinated services and educational and administration functions. Major planning and construction was placed in an integrated framework. The University of Rochester School of Medicine is tied into the continuing education and training of physicians, nurses, and other hospital personnel. There is limited rotation of residents. Moving from the health field across to another Government agency, the State Technical Services Act of 1965 established a program in the Depart- .ment of Commerce which was designed to help universities expand and reorient their technical assistant programs to provide improved technical service to industry. Each state designated an agency to administer the program in the state, and in general the activities of these agencies fell into three categories: (1) Preparing and disseminating technical reports, abstracts, computer tapes, microfilm reviews, and similar scientific or engineering information disseminated from technical in- formation centers established for that purpose; (2) Providing a reference service to identify sources of engineering and other scientific expertise; (3) Sponsoring industrial workshops, seminars, training programs, extension courses, demonstrations and field visits designed to encourage the more effective application of scientific and engineering itif ormation. B-4 Arthur D Little, Inc As might be expected, there was very uneven performance among the several states. The program never really earned a strong political constituency. Based on the notion that some types of industries in certain places had better access to available technical knowledge thar others, the theory was that if access were made easier the underprivileged industries would be able to engage in self-improvement. This theory assumed the existence of a kind of vacuum -- places and companies and whole states where little or nothing was going on. It was assumed that universities could fill this vacuum. A dozen or more had long since begun to try to do so. In Georgia, North Carolina, and Iowa (among others) there were more or less flourishing experimental stations. Georgia, in particular, was combed by roving field agents who were perceived as disseminating to the periphery technologies andtechniques which had originated or flowed through the university at Athens. Proponents discounted the fact that previous attempts at federally sponsored technology diffusion programs had failed. There were, of course, those who had considerable interest in pushing the program, but the right formula was never found. It turned out to be too difficult to find people in the less advanced industries who saw STS as an opportunity for them. Those in the delegated agencies had special interests of their own and tended to be more or less passive about-Pursuing the purposes of STS except where their interests happened to coincide. The gap never was conspicuously bridged, and the program was allowed to lapse in 1970. The Be,-iring of these Experiences on RNP When we began looking for experiences in regionalization that might have some bearing on RMP, we found ourselves going through a process of elimination. Starting with scores of examples of regionalization, we quickly found reasons why most of them were so radically different from RMP as to make detailed comparisons meaningless. What remained after this process of elimination is discussed in the following section. RMP was established by Public Law 89-239 as a program built essentially around regional cooperative arrangements. Local (regional) autonomy became a key philosophy behind the program. RNP was given no power of authority over those whose cooperation it chose to seek to achieve. Of the four reasons for regionalizing described at the start of this Appendix,* the first two and the last assume that the agency which has chosen to regionalize has control over the people and resources with- in their regional offshoots. P.L.89-239 does not give RNPS such control; it went further and madethe regional boundaries a matter of regional, rather than national, determination. Thus, the only reason that has applicability to RNP is that of accommodating to local conditions. Power is placed in the regions, and in this case it is essentially the power of persuasion. (1) Manageability; (2) Diffusioon of knowledge under controlled con- ditions; (3) Accommodating to local conditions; and (4) Quick response of the parent organization to feedback from the field. B-5 Arthur D Little, Inc The State Technical Services never fulfilled the hopes of its original backers. Two basic problems of design appear to have been some- what slighted: (1) It may be in the general interest for technology to be shared, but the specific injection of technical information into indivi- dual business firms is potentially adverse to the interests of competitive firms already in possession of that information; (2) Technical diffusion occurs across organizational boundaries when there is a "pull" on the technology much more easily than when the technology is being merely lipushed" from outside. STS rarely regarded itself as a "pull" program, that is, one intended to create an atmosphere in which people capable of "pulling" technology into their own organizations could be easily identi- fied and turned into effective bridges for diffusion. % In summary, those few regionalization programs that are sig- nificantly analogous in concept to RNP -- those that are not center- dominated and rely heavily on cooperative behavior -- point to one common conclusion: their success depends most importantly on a full appreciation of the political nature of inducing cooperation and on the ability of the participants to handle the political element with skill and imagination. B-7 Arthur D Little, Inc I f APPENDIX C Arthur 1) I.ittle, Inc.