Health Promotion and Aging
  "Injury Prevention"

       Richard W. Sattin, M.D.
  Chief, Unintentional Injuries Section
Division of Injury Epidemiology and Control
  Center3 for Disease Control, Atlanta

Michael C. Nevitt, Ph.D.

Adjunct Assistant Professor of Medicine and Epidemiology
    University of California, San Francisco

      Patricia F. Wallet, Ph.D.
Director, Injury Prevention Research Center
University of North Carolina, Chapel Hill

Richard H. Seiden, Ph.D., M.P.H.
   The Glendon Association
  Los Angeles, California

THE FREQWENCY AND IMPACT OF' FALLING IN THE ELDERLY

Falls (IO-9 codes E880-E888) are the leading cause of death from injury in per-
sons over the age of 65.  Approximately two-thirds of reported injury-related
deaths of persons 85 years of age and older are due to falls (1). Of the 8200
fatal falls that occurred in the United States in 1985 for persons aged 65 years
or older, 59% were those that occurred in the home. This large number of fatal
falls listed on death certificates, however, may understate by one-half the num-
ber of deaths in which falls are contributing causes (2).

Approximately 250,000 hip fractures per year among persons ages 45 and over
result from falls, with an annual medical cost exceeding $7 billion (3). There
may also be six times as many fractures of other bones as there are hip frac-
tures in persons over the age of 65 (4), most of which are attributable to falls
(5) .

The majority of falls in the elderly result in minor physical injury (6), with
only a small percentage of falls causing severe injury, such as a fracture. Es-
timates of the proportion of falls causing a fracture range from four to six
percent in ambulatory populations, with one percent or less resulting in hip
fractures  (6,7). Slightly higher rates of hip fractures per fall have been
reported among institutionalized populations (8,9,10).

Most falls, however, go unreported and are not medically attended. Respondents
to the 1984 U.S. Health Interview Survey (HIS) Supplement on Aging were asked
about <heir history of falling in the previous 12 months. Among women,  19% of
those ages 55-69, 23% of those ages 70-79 arid 34% of those age 80 or more said
they had fallen in the past year. For men, the percentages were 15, 17 and 28,
respectively. Overall, about half of those reporting a fall said they fell more
than once.  The proportion of multiple falls was highest for those age 80 and
over for both sexes (11). The incidence of falls is considerably higher in long-
term care facilities than among elderly living in the community (7,8, 9,12,13).

Psychological trauma resulting from falls may be severe and can result in loss
of confidence in the ability to perform daily routine,  restriction of activity,
decreased mobility and increased dependence. Deconditioning, muscle atrophy and
joint stiffness or contractures that result from immobility may lead to more
falls and further mobility restrictions.

CRITICAL CONCERNS REGARDING HOST,  BEHAVIORAL A ENVIRONMENTAL FACTORS

1s PROBLEMS WITH DATA ON THE OCCURRENCE OF FALLS,  RISK FACTORS AND INTERVENTIONS

Reporting Fall-related injuries are not consistently reported in public health

                                      D-l


surveillance systems or medical care data bases. Although records of falls are
required in most long-term care facilities,  there is no standardized reporting
system.  In addition, efforts are needed to insure standard application of coding
systems for falls. Inaccurate self-report of falls by the elderly may alSO af-
fect research and prevention efforts.  One recent study found that 13% of per-
sons with a confirmed fall during a 12-month period failed to recall having a
fall when questioned at the end of the 12 months (14). Other problems include
absence of a definition of falls that is easily understood by study participants
and the "response burden" sometimes placed on participants who report falls.

Risk factors and interventions Until recently,  there have been few prospective
studies of risk factors for falls or injury from falls. Many presumed risk fac-
tors for falls exist among those who do not fall as well as those who do. There
are few existing studies that have included statistical analyses of fallers com-
pared to a control group, or that have evaluated the strength of suspected risk
factors using odds ratios, relative risks,  or attributable risk measures- No
studies have used adjustment or multivariate analysis to control for multiple
risk factors. Case-control studies of falls are relatively numerous.  However,
the risk of falling,  the risk of injury and factors leading the person to,seek
medical care,  such as perceived severity of the injury or preexisting medical
conditions, all contribute to differences between cases and controls, making in-
terpretation difficult.

Finally, there are very few controlled trials of the effectiveness of interven-
tions to prevent falls,  and none of the yield and effectiveness of medical work-
up3 of previous fallers. Therefore, existing preventive and treatment efforts
are empiric in nature,  relying mostly on common sense and descriptive studies
of the characteristic3 of falls and fallers.  Host and environmental risk fac-
tors and behavioral interventions remain an uncertain and an ad hoc undertak-
ing.

2. PROBLEMS ARISING FROM THE COMPLEX NATURE OF FALLS

Falls in  the  elderly   are  the   common   result of a  very large   number of
pathophysiological processe3,  primary aging processes,  behavioral and environ-
mental factors (15).  Some fall3 are an acute symptom of a chronic or episodic
cerebrovascular,  cardiovascular or neurologic disorder (16-19). Syncopal falls
are prototypical of this kind of fall. However, a variety of acute illnesses may
manifest nonspecifically in falls in persons with multiple risk factors (12,20).
Falls can also be a nonspecific result of disease-relatedand age-related declines
in gait, balance, sensory perception, strength, coordination,  reflexes and other
aspect3 of neuromuscular function  (12,21).  When the healthy,  active older per-
son falls,  it often involves an unexpected hazard in the environment, one which
could>also cause a much younger person to take a spill (slippery surfaces or un-
seen obstacles). However, the older person may be influenced by more subtle en-
vironmental factors,  such as lighting and visual and spatial design.
The multifaceted, multifactorial nature of falls has prompted attempt3 to develop
a typology of falls (22-25). These typologies focus on the circumstances of falls
and provide information about the probable etiologic factors that guide inter-
vention efforts (24). In addition,  greater specificity about outcome3 would en-
hance  understanding of  risk  relationships by allowing researchers to link
specific risk factor3 or biologic measurements to specific types of fall. Some
examples of these typologies follow:

.  unexplained falls versus falls with a self-evident etiology (i.e. syncope,
   seizure,  stroke)

.  falls due to host (intrinsic) factors versus falls due to environmental (ex-
   trinsic) factor3

o pattern or recurrent falls versus occasional or isolated falls

                                   D-2


  falls occurring in the sick or older elderly (age 75 and over) versus $hose
  occurring in the healthy or younger elderly (ages 60-74).

Unfortunately,  work on classifying falls is still developmental and may be of
limited value in understanding and preventing falls for the following reasons:


a. It is sometimes difficult or impossible to obtain valid information about the
  circumstances of a fall.

b. Syncopal falls may have an etiology similar to many "unexplained falls" which
do not progress to full loss of consciousness but do involve the effects of
decreased cerebral perfusion on muscle tone and balance (26,27).

c. Most falls probably have a mixed etiology,  involving both host and environ-
  ment as contributing factors.

d. Trips and slips involving a definite hazard may also implicate 1) age-related
  changes in gait (28), and 2) decline in the speed and organization of dynamic
  postural responses to external displacement  (29) I  blurring the distinction
  between environmental and balance falls.

e.  What constitutes an environmental hazard depends on the individual's func-
  tional capacity.  With functional decline,  features of the environment which
were once negotiated without difficulty can become major barriers.

f. The same individual can fall for different reasons on different occasions.
  This makes it difficult to classify individuals as one type of faller or
  another.

g. Persons at risk for falls because of abnormalities of gait or balance may SC
  restrict their activities that they fall infrequently over the near-term.
   Such  "adaptations"  to diminished capacities may be dysfunctional over the
  long-term,  accelerating loss of function and leading to multiple falls.

Prevention of falls must address a large number of risk factors. At present, we
know very little about the interaction between risk factors which will be neces-
sary in the development of effective prevention efforts.

3. PROBLEMS ARISING FROM THE UNCERTAIN SIGNIFICANCE OF ANY FALL

Because the etiology of falling is complex,  the significance of any individual
fall is difficult to determine,  both for health care professionals and for the
person who falls.  This may lead to inappropriate actions at several extremes,
including: extensive medical work-ups which have little yield; dismissal of the
fall as of no consequence; inappropriate reductions in mobility and activities,
including use of physical and chemical restraints (30); or extreme fear of fall-
ing again.
In someJ persons,  a fall or series of falls signals serious acute illness,
precipitous functional decline and, possibly,.imminent death (7,31,32). For these
reasons, any fall must be taken seriously by clinicians. However, most falls in
the elderly do not carry this meaning. Falling is an ubiquitous experience
throughout life,  usually resulting in no or only very minor injury.

Though the circumstances of falls appear, on average, to change with age (12,33),
the most frequent fall in the elderly is a consequence of persons with diminished
functional capacity attempting to meet the intrinsic and external demands of mo-
bility within specific environments.  For the relatively fit and functionally
able, mobility entails constant exposure to and successful negotiation of a wide
range of physical environments. Risk of fall is spread over many diverse situa-
tions and environments. As function declines,  success in mobility focuses in-
creasingly on basic movements, such as transfer,  short walks and quiet standing
within a familiar environment.  These basic movements then become the focus of
exposure to fall risk.

The behavioral response t0 falling and postural instability affects the trans-

D-3


lation of physical decline into reduced mobility (34). A fall or a near fall
provides information about activities and circumstances which place a particular
person with a given set of capacities  at risk as well as information about a
mismatch between the external and intrinsic demands of mobility and individual
competence (35). This information may motivate a reduction in mobility, in turn
resulting in reduced exposure to the risk of falling by decreasing the range of
environmental exposures and by decreasing the time at risk while walking, trans-
ferring and standing. Indeed, persons  who do not adjust their activities to
declining capacities may be at especially high risk (12,36).

Clearly, however, adjustments in activity and mobility in response to falls are
neither universally appropriate nor  sufficient to eliminate the risk of fall-
ing.  1) Fear and excessive restrictions in activity may reduce exposure to the
risk of falling in the short term,  but only increase the long-term risk by un-
dermining self-confidence and physical conditioning.  2) For the elderly whose
functional capacity is severely compromised,  maintenance of even a minimum of
independent mobility may entail substantial risk- 3) Some risk of falling is
probably unavoidable if mobility and independence are to be maintained in the
presence of functional decline.

The goals of prevention should be realistic and based on our best understanding
of the problem.  It is realistic to aim for modest reductions in the frequency
of falls and perhaps to prevent a recurrence of falling in some individuals. It
is not realistic, given our current understanding, to eliminate falls as a fea-
ture of aging.  Even if every fall does have a set of causes, there will remain
a random element in many falls beyond our ability tomodel, predict or anticipate.

Equally important,  prevention efforts must strike a balance between protection
from risk and the maintenance of mobility, function,  personal autonomy and an
acceptable quality of life.  To optimize the latter,  it may be necessary to ac-
cept a certain level of risk. Prevention should focus on modifying risk factors
that reduce that level of risk as much as possible while impinging on inde-
pendence and autonomy as little as possible.

Prevention efforts would benefit from an increased understanding of behavioral
and psychological responses to the onset of instability and falls. The nature
of this response may have important implications for the individual's short-term
and long-term risk. Fear and excessive activity restrictions may only increase
risk in the long run. On the other hand,  failure to make some behavioral accom-
modation to aging and disease may also increase risk in the near term. Adapta-
tions to diminished function,  while perhaps inevitable,  should be appropriate
to the threat and emphasize and strengthen residual capacities. In addition, re-
search is needed on what constitutes "risk-taking behavior" in the context of
speci$ic functional disabilities.

Finally, preventing the adverse consequences of falls, including injury, fear
and the "long lie," may be as important a goal as preventing falls. Severe in-
jury may precipitate maladaptive behavioral responses as well as lead directly
to physical deconditioning and further falls.

REVIEW OF HOST AND BEHAVIORAL FACTORS

The following risk factors are limited to controlled studies in which comparisons
were made between "failers" and "nonfallers." Specific study designs vary con-
siderably. Nearly all of the associations between risk factors and falls reported
here are univariate and do not control for confounding.

1.  GENERAL RISK FACTORS

Age and Sex. These variables may contribute to identifying persons at risk, but
tell us little about actual causes of falls or where to intervene to reduce risk.

D-4


There is substantial variation in risk within age groups. Biologic and function-
al variability within age groups may be more important determinants of fall risk
than age-dependent variations.

History Of Previous Falls and Dizziness. It is not known how the risk of injury
is related to the frequency of falling. The ratio of injuries prevented per fall
prevented may vary considerably between frequent and infrequent failers. For ex-
ample, those who fall frequently may do so in a way that has a low risk of in-
jury or learn to protect themselves from injury. Research is needed on how the
mechanics of falling affect the risk of injury (5).  Certain interventions may
be less effective after a person has fallen. For example, extreme fear Of fall-
ing may reduce acceptance of exercise programs to improve neuromuscular func-
tion.

Health Status,  Mobility Limitation and Functional Disability. General health
variables appear valuable in identifying elderly at increased risk of falling
because their association may reflect a common origin in underlying diseases and
conditions.  However,  mobility limitations and functional disability may also
have a direct bearing on prevention to the extent that they indicate a mismatch
between the external and intrinsic demands of mobility and personal competence
increasing the risk associated with routine activities (35). Residual capacities
may be enhanced by environmental modifications that reduce the demands placed
on the individual. However,  it is not known whether environmental and behavioral
interventions  that improve function also reduce fall risk. Moreover, the
relationship between mobility and fall risk is complex and not well understood.

Mental Status,  Psychological Status,  and Psychosocial Factors.  Cognitive,
psychosocial and psychological risk factors  for falls in the elderly are not
well  understood,  but are being evaluated in  ongoing prospective  studies.
Neurological disorders affecting cognitive  function are often clinically as-
sociated with neuromuscular deficits and falls in the elderly, but it is not
known if the association is causal (18,37-39).  The causal relationships are
potentially complex (40). Confusion, impaired judgement, distraction, agitation,
depression and lack of awareness may increase exposure to hazardous situations.
Associated gait and balance deficits and psychomotor depression may increase the
chance that a fall will result.  Depression, in turn, may result from falls, in-
jury or physical illness.  Antidepressant and sedative medication may increase
the risk of falls (41,42). The behavioral aspects of depression that affect fall
risk are not well understood.  There are no studies of the effect of cognitive
or psychological factors on the coping strategies and adaptations of elderly in
response to falls and instability.
Physic&Activity. Longitudinal studies of physical activity and falls are needed
since reduced activity levels may result from previous falls, fear of falling
or gait and balance problems. Moreover,  increased physical activity could in-
crease exposure to environmental hazards.

lbrironwmtal Hazards.  Environmental hazards include such factors as stairway
design and disrepair, inadequate lighting, slippery floors, unsecured mats and
rugs,  and lack of non-slip surfaces in bathtubs, among others. These factors
have been implicated in about one-third to one-half of all falls or falls in-
juries in the home (43-45). Most studies that deal with home environmental hazards
are difficult to interpret, however,  because of differences in case selection
criteria,  information collected,  and presentation of data. Definitions,  espe-
cially those of environmental hazards,  were  not provided,  making valid com-
parisons difficult.

Only two investigators who studied the environment  used a referrent group of
nonfallers for comparison: one of case-control and one of cohort design (31,45).
However,  these investigators described the environment only for cases and not

                                     D-5


for the referrant group,  and the environment was not assessed visually. In-
stead,persons were interviewed to determine what-the respondents felt were the
causes of their falls.  Few studies actually defined the environment or an en-
vironmental hazard,  and none provided a uniform approach to assessing the en-
vironment.  Despite  indication3 that several potential risk factor3 might be
interrelated, only one study explored the possibility of such interaction3 in a
limited way (45).  More analytic studies need to focus on where falls occur in
the home and on the prevalence of various home hazards. The risk attributable
to each of these home hazards, especially in relation to a person's time at risk
to these exposures, is criticalto the design of prevention strategies. Moreover,
the definitions of a room, dwelling unit,  and home hazard need to be clearly
stated,  reproducible and valid.  The use of analytic techniques that determine
risk factor3 for diseases,  such as the determination of the interaction of host
factor3 and the environment,will be key to the etiology of falls and fall in-
juries. Intervention strategies would then be based on sound epidemiologic prin-
ciples hnd would take into account acute and chronic health problem3 as well as
contributing environmental factors.

2. COMMON INTERMEDIATE PATHWAYS: NEUROMUSCULAR FUNCTION

Gait and Balance.  Clinical and laboratory assessment of gait and balance is in-
creasingly sophisticated and show3 significant promise as a method of assessing
fall risk.  Important research issue3 remain, however,  including 1) identifica-
tion of the modifiable causes of gait and balance abnormalities, 2) the relation-
ship of clinical assessments of balance  and gait to laboratory measurement of
the biological mechanism3 of balance and gait,  3) the utility of computerized
gait analysis for fall risk assessment and research,  4) the utility of gait and
balance measures a3 intermediate outcome measures for risk factor modification
studies,  and 5) the relationship of falling to the determinant3 of total motor
reaction time (46) -

In addition,  the relationship between balance and gait is not well understood.
For example,  slowed walking speed may be caused by balance problems,  fear of
falling, or both,  or it may be due to pathology not directly related to balance
(47).  The effect of musculoskeletal condition3 on gait and balance performance
is  not well understood.  The ability to influence  corrective and protective
response through training and learning should be investigated (46). Finally, an
understanding of how specific gait and balance problems transform environmental
features into "fall hazards"  would help focus environmental intervention ef-
forts.

FACTORS AFFECTING NEUROMUSCULAR FUNCTION AND OTHER SPECIFIC RISK FACTORS

Muscle Strength. The effect of improvements inmuscle strength on gait or balance,
or on <he risk of falling, is not known.  Upper and lower limb strength may be
important in protective responses during a fall which prevent injury, but this
has not been studied (5).

Afferent and Efferent Systems.  Published studies linking  either clinical or
laboratory measurement of the sensory and motor nerve systems to data on falls
are lacking,  though some studies are in progress. There is indirect evidence,
however,  that simultaneous abnormality in somatosensory, visual and vestibular
systems critically impairs dynamic balance (29,48). Clinical assessment of pro-
prioception,  vibration sense and reaction time need to be validated against
laboratory measures.

Vision.  Visual function is *degraded by conditions  affecting the crystalline
lens and other intervening structures, reducing and scattering the light reach-
ing the retina (49).  Early detection and treatment of common condition3 such as
glaucoma and cataract should improve visual function (50) andmight reduce falls.
Improved correction of acuity is feasible (updated prescriptions, improved com-
pliance) and could also reduce falls. EnGironmental and behavioral adaptation3

                                  D-6


to reduced visual performance may improve visual function and also reduce falls,
but this has not been studied systematically.
Decreased cerebral perfusion. True syncopal falls in the elderly may be uncom-
mon (24,421, and dizziness may accompany these falls more often than cause them
(37) -  However I there are no data on the incidence of these falls. Falls dlle to
postural hypotension may a130 be relatively uncommon since many elderly adapt
readily to the symptoms of homeostatic dysfunction (51,SZ). Decreased cerebral
perfusionmay cause falls by compromisingmuscle tone and postural control without
other symptoms being present  (261,  but this is not well studied.

Medications and alcohol.  The use of individual drugs should not be blamed for
falls without appropriate evidence. The most difficult problem is distinguish-
ing the effect on falling of specific drugs (e.g.,  antihypertensives or hyp-
notic-anxiolytics) from the effect of underlying diseases (e.g., hypertension
or dementia)  or from the effect of drug-disease  and drug-drug interactions
(27,41). The effect of polypharmacy on falling is a potentially important problem
(53) that has not received sufficient study. Detailed studies of falls and drug
and dose-specific physiologic effects in the elderly are needed to focus inter-
vention efforts. Better methods are needed for collecting data on alcohol use
in the elderly and linking it to data on falls. The effect of medication review
programs for the elderly on the risk of falling is not known.
Specific diseases. Intensive clinical work-ups of elderly who fall will uncover
large numbers of diseases and pathologic conditions thought to contribute to
falling (54,55). But many of these conditions will also be frequent in elderly
who do not fall. It is not known if the diagnostic search for treatable medical
causes of falls has sufficient yield to justify the effort and cost. The presence
of multiple pathology and abnormality may be as important a determinant of the
risk of falling as individual conditions  (20) .  The effect of specific condi-
tions,  such as arthritis and peripheral neuropathy, on falling and intermediate
risk factors is relatively unexplored and warrants increased attention.
REVIEW OF PROGRAMS TO PREVENT FALLS BY MODIFYING HOST AND BEHAVIORAL FACTORS
Although fall prevention efforts in the community and health care settings are

not uncommon, few of these programs have been systematically evaluated in terms
of outcomes or cost. As a result, these efforts have yielded remarkably little
useful information.  Some of the reasons for this situation are as follows:
1. Many programs are part of comprehensive community-oriented health promotion
efforts that treat falls and injury as only one of a sometimes daunting array
of objectives. The result is that inadequate resources are devoted to evaluat-
ing the pfogram's effect on the incidence of falls or fall-injury.

2.  The outcome3 of fall prevention programs  can be difficult and costly to
measure, a problem faced by both clinical and community-based interventions.
-Anticipated effect sizes are small (e.g., lo-15% reductions in the number
   of falls), necessitating relatively large sample .sizes.
-Frequent and intensive contact with individuals is required to assess falls
  per se in community settings.
- Injurious falls treated in emergency rooms are only a small portion of all
   falls, increasing sample size requirements, and may be insensitive as in-
  dicators of the effect of community programs.
-Falls treated in doctors' offices are more numerous, but may be subject to
  reporting inaccuracies and are expensive to monitor in an entire community.

  - Measurements of intermediate outcomes which could reduce sample size re-
  quirements,  such as gait and balance performance,  are in a developmental
  stage and require individual contact by trained personnel.

                                       D-7


4. The appropriate outcome measure is not always clear (e.g., number of f:llz,
number of injurious falls, fallerz, failers with injury, severity of injurious
falls).  Individuals with very large numbers of falls may skew the results of
analyzes of number of falls,  even for relatively large sample sizes.

5. Fall prevention programs usually combine several intervention components ad-
dressing multiple risk factors.  While this is appropriate given the multifac-
torial causes of falls, most studies have neither an appropriate design nor the
statistical power to determine the effectiveness or cost of the separate inter-
vention components.

6. There is an over-reliance on one group, pretest-pozttezt designs, the results
of which are often incorrectly analyzed. The selection criteria for subjects in
these studies are often inadequately specified.

7.  Many intervention programs in health care settings are undertaken without a
rigorous approach to methods resulting in a number of limiting deficiencies.

Review of Prevention Programs

The programs reviewed here were evaluated with a controlled study design and
were published with sufficient detail in methods to evaluate the validity of the
study findings.

1. Interventions in community populations. There are only a handful of published
intervention studies in community populations meeting the above criteria.

a) A multiple risk factor intervention program using the Portland, Oregon, Kaiser
Permanente Medicare population focuses  on the ambulatory elderly (56). Two
thousand five hundred households of the elderly were randomized to intervention
and control groups.  Both groups were provided with the results of an initial
home safety audit,  were given a home safety publication,  and underwent an as-
sessment of risk factors for falls,  including performance measures of strength
and balance. Both groups are being followed for 24 months to ascertain falls and
fall-related medical care utilization.

The intervention received by the experimental group included: 1) encouragement
and assistance in completing safety repairs and modifications identified in the
home audit; 2)  a series of four falls-prevention workshops covering a) exercises
to improve strength, balance, flexibility, posture, and conditioning, b) drug
safety and calcium intake, c) falls risk awareness and risk control, d) develop-
ment of social support skills and an environment for group reinforcement; 3)
guidelines for preventive health care,  and screening and follow-up care for
vision and hearing problems.
This i's the most rigorous and best designed.fall prevention intervention study
to date.  Preliminary results indicate that both the intervention group and the
control group are experiencing reductions in the rate of self-reported falls per
100 person-years during year 2 compared with year 1 of the study (56). However,
with follow-up nearing completion, no consistent differences have emerged in the
incidence of fall-related injuries between the control and intervention groups.
This study will be one of the few to provide data on the costs of the interven-
tion.  However,  since the entire intervention group received all of the program
components,  only limited information will be available on the relative effec-
tiveness of the various tactics.

b) One hundred elderly persons seen in community'medical practices in Binning-
ham,  England,  and who had fallen in the previous 4 weeks were randomly assigned
to two treatment groups (57). One group underwent an intensive, "long" (up to
12 visits) course of home-based exercise physical therapy and the other group
received a  "short" course  (no more than three visits).  Participants were fol-
lowed for four months for falls and changes in balance and mobility.   After

                                     D-8


four months,  there were no differences in outcome between the two groups, but
both groups showed improvement in mobility and balance measures. The fact that
only half of the study population fell during follow-up was cited as an indicator
of effectiveness.  However,  the historical controls used for comparison provide
an insufficient basis for this conclusion.

2.  Interventions in Institutions. A large number of institutions have undertaken
fall prevention programs, but few have been systematically evaluated.
a) Fife et a1.(58) evaluated an individualized nursing intervention to prevent
falls in a community hospital in Cleveland, Ohio.  The program was implemented
in two hospital units, with two units comparable in census and in age of patients
used as controls. Patients deemed at high risk of falling were identified (based
on an informal risk analysis) at admission to the experimental units (83% of all
unit admissions)  and an individualized prevention program developed for each
patient. At the end of the 12-week study period, there were no differences be--
tween the number of falls reported in the experimental and control units. The
authors reported contamination of control groups. A reduction in falls on the
experimental units compared to historical controls was cited as evidence of ef-
fectiveness.   However,  the appropriateness of the control period and method of
comparison were not demonstrated.
b) Rainville (59) reports the result of a similar hospital-based study. Patients
on an experimental unit of the hospital were assessed for fall risk upon admis-
sion and a  "standard care plan"  implemented for these patients. The interven-
tion consisted of patient and family education in the hospital on fall hazards,
increased staff awareness of the high-risk patient,  and a special care protocol
focusing on transfer to and from bed.  Evaluated with a pretest-posttest design,
this study also failed to demonstrate a clinically or statistically significant
decrease in falls following implementation of the program.
c) A study reported by Gray-Vickery (60) demonstrates an increased short-term
awareness of potential fall hazards in the home among elderly hospital patients
receiving safety education classes. However,  the study is limited by small num-
bers and a one group,  pretest-posttest design.

3. Clinical Interventions. In spite of numerous articles on the clinical evalua-
tion of the faller,  very few studies have systematically assessed the yield of
treatable medical disorders in failers or the effectiveness of treatment in
preventing falls.

A "falls clinic" was established at a geriatric care center on Long Island, New
York, which emphasized an interdisciplinary approach to diagnosis, treatment and
coundeling of patients who had previously fallen (54). Intensive medicalmanage-
ment was accompanied by a home visit from'occupational and physical therap,ists
and by falls-prevention education for patients and their families. A total of
120 potential fall-related diagnoses were identified in 36 elderly fallers by
cardiologists,  neurologists,  physiatrists and geriatricians. A substantial
decline in the number of patients who fell during 12 months of post-interven-
tion follow-up compared to before the intervention (22% vs 100%) was cited as
evidence of effectiveness. However, patient selection criteria that may have af-
fected the likelihood of falling again are not specified and no information is
provided on the cost of the intervention.  It is not possible to determine the
contribution of the medical workup and treatment to the observed reduction in
falls.

Are common sense and the existing state of knowledge of risk factors sufficient,
in the absence of adequate experimental evidence to warrant the endorsement and
propagation of specific prevention tactics? In the case of low-cost, low-risk
interventions such as risk awareness and exercise programs, the answer is prob-
ably yes, especially when there are other benefits besides fall prevention. For

                                     D-9


higher cost strategies such as home modifications and falls clinics the answer
is probably no.  In addition to falls, intervention studies should monitor
mobility,  functional status,  dependence and psychosocial outcomes to determine
if there are benefits of intervention in addition to reduction in falls and if
a reduction in falls comes at the expense of autonomy, level of activity, and
quality of life.

MOTOR VEHICLE AND PEDESTRIAN INJrJRY TO THE ELDERLY

OVERVIEW OF THE TOPIC

Motor vehicle and pedestrian injuries (ICD-9 codes E805.2, E822.7, E810-E819)
to older persons (age 65 and above) occur primarily as a result of the transfer
of mechanical energy to people in amounts or  rates that exceed their injury
thresholds (61). Motor vehicle injury is by far the leading cause of fatal in-
jury for all ages,  and for the elderly it is the leading cause to age 75 when
it is exceeded only by falls.

More than any other injury type, motor vehicle injuries are systematically in-
vestigated,  and a variety of information is routinely provided on circumstances
surrounding the event. Centralized,  computerized databases exist at both state
and national levels on these injuries, although there is variation in the quality
from state to state. Because of the amount of information available on the cir-
cumstances surrounding motor vehicle injury,  it has been possible to develop in-
terventions aimed at alleviating specific causes of injury.

Although major shortcomings continue to exist,  these state crash files can be
used alone or in conjunction with other databases to address a wide variety of
problems. They can be linked to driver history files. They can provide addition-
al information on highway design characteristics associated with certain kinds
of injury-producing crashes and on specific vehicle design characteristics as-
sociated with injury to elderly occupants.  In some states they can be linked to
ambulance run data to evaluate the time required to provide medical attention
and how the time delay is related to subsequent outcome.  In a few states all
these databases can be linked to hospital data,  including trauma registries and
hospital discharge data.  The potential for such data linkage provides the op-
portunity for addressing such questions as whether certain vehicle design charac-
teristics are associated with a greater probability of post-crash fire resulting
in burn injuries to occupants. Likewise,  data from trauma registries, if linked
to crash files and then to highway files,  can be used to determine whether there
are certain highway design characteristics,  including the presence or absence
of pedestrian controls, that are associated with injury to elderly pedestrians.
These data can be used in conjunction with other databases such as census data
and economic indices to address such questions as how fluctuations in the economy
relate to motor vehicle injury and whether some groups are at different risks.
    3
One special problem with using available hatabases concerns the older person's
greater vulnerability to injury and later death from a given crash. The elder-
ly motor vehicle crash victim may die long after the injury producing event as
a result of infection or another problemresulting fromthe crash. If the proximate
cause of death is pneumonia,  the death may not be readily identified as motor-
vehicle-related and thus be lost to the analysis.  Investigators in this area
need to be sensitive to this problem in attributing causes of death and time
frames for defining motor vehicle deaths.

Background National Rates by Age, Race, and Sex

For the population age 65 and older, mortality rates per 100,000 for the years
1978-1982 were much higher for males (34.2) than for females (14.9). Black and
other males have 1.6 times the rate of white males until age 75, at which age
the rates become similar.  Black and other females have between 0.7 and 0.8 times
the rate of white females at all ages 65 and over.   However,  the rates for all
groups are gradually decreasing.

                                    D-10


When death rates are divided into ages 65-74 and those age 75 and over, rates
are much higher for the older age category. These figures do not take into con-
sideration amount of exposure to risk,  but there is evidence that exposure
decreases with age, so that the increased rates associated with older age would
be even greater if exposure were considered.

According to the National Safety Council (62), in 1985 there were 45,600 motor
vehicle deaths and 1,70O,OOf1 dtsabling injuries.  Approximately 6400 of the
fatalities were age 65 and older. Based on the number of licensed drivers, those
age 65-74 had lower rates of both fatal crash involvement and total crash in-
volvement than was true for the driving population as a whole. In contrast,
drivers age 75 and older had 59 fatal crashes per 100,000 drivers, compared to
38 for drivers of all ages.  For total crashes (fatal and nonfatal), this age
group had 26 per 100 licensed drivers, compared to 21 for drivers of all ages.
While,these rates were higher than those  for the general driving population,
they were lower than those for drivers below age 25.

Crash involvement rates based on mileage driven, rather than number of licensed
drivers show drivers over age 65 to be overrepresented in crashes (63-68). Some
studies report that on a per mile basis the older driver is second only to the
young driver in crash risk,  while other analyses  rank the elderly driver as
having the highest crash risk per mile driven.

Although pedestrian fatalities account for more than 18 percent of all motor
vehicle fatalities (62), they account for 23 percent of motor vehicle deaths oc-
curring to persons ages 65-74 and 35 percent of those occurring to persons age
75 and older. Although this overrepresentation may be attributable to greater
exposure,  a comprehensive study by Tobey et al.  (69) reported that persons age
60 and over represented only 7.7 percent of pedestrians observed but accounted
for 12.8 percent of the pedestrian injuries  (both fatal and nonfatal). Thus it
appears that the elderly are overrepresented in pedestrian casualties whether
the analysis is based on population or exposure.

Effect of Aging Population on Importance of Motor Vehicle Injury as a Public
Health Issue

The elderly are the most rapidly growing segment of our population and it may
be anticipated that without efforts to facilitate mobility for the elderly, the
magnitude of the motor vehicle and pedestrian injury problem will increase.

One factor that may ameliorate the injury problem is related to driver cohort
effect.  Many elderly drivers today,  particularly women drivers, have limited
driving histories, but because of a spouse's death or failing health have in-
creased the,ir driving time or even obtained a license for the first time. As a
result,  their driving inexperience is reflected in their performance. In con-
trast, larger proportions of the younger population are obtaining licenses than
was previously true. When these young drivers become elderly, they will have a
lifetime of driving experience behind them,  which in turn should result in more
proficient performance and a reduction of the number of collisions and injuries
occurring to the elderly.

ENVIRONMENTAL ASPECTS OF THE INJURY PROBLEM

The highway transportation system was designed without taking into account the
capabilities and limitations of the elderly (70).  Relatively little attention
has been given to the special characteristics of the elderly driver. Vehicle
design does not consider what features might facilitate or impair the function-
ing of older operators and highway design has been developed primarily on the
basis of performance measures obtained from young males. Sivak, Olson, and Pas-
talan (71) found that older drivers who had similar results as younger drivers
on standard tests of visual acuity still'had to get much closer to highway signs
at night in order to read them. Consequently they had less distance remaining

                                    D-11


in which to react to the information provided. Older people as a group need more
light in order to see as well as younger drivers and they have more difficulty
with glare recovery.

In both violations and crashes,  older drivers are overrepresented in problems
associated with intersections. They are especially likely to have problems with
yielding right of way,  observing stop signs and traffic signals, and maneuver-
ing left turns (66,72,73).
One critical need is to modify vehicle and highway design to facilitate the driv-
ing task for the elderly.  The National Academy of Sciences (NAS) is Currently
conducting a major study on improving mobility and safety for older persons and
is commissioning a number of papers addressing specific aspects of the problem.
Some issues being investigated are improvements in intersection design (for both
drivers and pedestrians), vehicle modification to increase crash protection for
the elderly,  and sign legibility and conspicuousness. However, no attention is
given to modifying vehicles so as to facilitate the driving task for the elder-
ly,  nor is the highway design in general given serious attention.
An environmental modification that has been given considerable attention in
Europe but relatively little in this country is that of exterior vehicle design
that would reduce pedestrian injury in the event of a collision. Most pedestrian
crashes occur at relatively low speeds ( 20 mph), and modification in vehicle
design,  such as softer structured bumpers, can change the probability of victim
survival and impairment  (74) -

Review of Programs or Efforts Attempting to Deal with the Problem
Modifications to the transportation system that simplify the task of driving or
walking should aid the elderly.  Three specific modifications are far-side bus
stops,  increased delineation of highways,  and redesigned residential areas to
reduce the amount and speed of traffic.

The location of bus stops is associated with the probability of injury occur-
ring to passengers disembarking from the bus.  Traffic coming from behind a bus
cannot see passengers disembarking and crossing an intersection until they are
very close. If the bus crosses the intersection before stopping, passengers dis-
embark and cross the intersection behind the bus where they have a better view
of traffic. Thus relatively simple changes can reduce the probability of injury.

Increasing the delineation on highways (centerline and edge and lane lines, as
well as guide signs and delineator posts)  increases the probability that they
will be seen. Uniformity of delineation is also important to reduce confusion.

The Woonerf, a design for a living environment implemented in Europe, the Nether-
lands and Japan that includes joint and safe utilization of space by people for
recrea)tional purposes and vehicles reduces ,the danger of vehicle crashes  (75) -
Residential areas are designed to severely limit through traffic and to reduce
drastically the speeds of local traffic.

BEHAVIORAL ASPECTS OF THE INJURY PROBLEM

There is evidence that the elderly as a group restrict their driving as they
recognize their increasing limitations (63,76,77). However, elderly drivers are
not necessarily aware  of some of their problems.  Planek et al.  (78) examined
older drivers' errors that resulted in crashes with how older drivers perceived
their driving problems. Failure to yield right of way was the major cause of the
crashes,  but the older drivers ranked it ninth out of ten maneuvers posing
problems. Even with self restriction, on a per mile basis, risk of crash among
the elderly is among the highest of any driving group.
A critical concern is that so little has actually been attempted to help older
drivers and pedestrians. As a group they appear as responsible and motivated to
do well,  but it is likely that those most in need of behavioral modification
would be least likely to seek and respond to intervention programs.

                                    D-12


Review of Programs or Efforts Attempting to Deal with the Problem
Licensing Programs: very few states have special procedures for driver licen-
sure after a certain age, and none have been carefully evaluated.

A study by Zaidel and Hocherman (79), examining Israel's requirement for vision
and medical evaluations for renewal applicants age 65 and older, found that 25
percent of the applicants had to wear corrective lenSe3. Because many of the ap-
plicants had already obtained glasses, only seven percent actually started wear-
ingthem as a result of the required vision test. The authors therefore concluded
that the requirement was not cost effective. However, vision screening in the
licensing station can be conducted at minimal cost, and those applicants already
using corrective lenses would not be required to seek professional evaluation.
Furthermore, the seven percent of applicants detected who were in need of len-
ses would translate into many thousands of drivers in this country.
The Florida Senate Committee on Transportation reported that the 70 year and
over age group experiences a license examination  failure rate that is nearly
double that of all drivers similarly tested and that their mileage crash rate
is higher than that for all other age groups over 25. Once involved in a crash,
their chances of being seriously injured or killed are the highest of any age
group.
Related to increased evaluation of older drivers for license renewal is the need
for a system to ease older drivers out of the driving population gradually. To
be effective,  these programs must pay attention to the capabilities of older
drivers and the demands placed on them by specified driving circumstances.
In addition to improved evaluation and appropriate restriction of older drivers,
there is a need for increased attention to meeting the transportation needs of
the elderly through public transportation or other systems such as volunteers
or carpooling. The transportation needs of the elderly will increase, and they
will be met either by individuals struggling with them one by one or with the
help of a more organized, coordinated approach where resources and solutions are
shared.

Driver Education:  The National Retired Teachers Association/American Associa-
tion of Retired Persons (AARP) has developed a program for special needs of the
elderly called "55 Alive/Mature Driving"  (80) _ The American Automobile Associa-
tion (AAA) has also developed a program called "Safe Driving for Mature
Operators." The AARP program is muchmore widely used, but neither program reaches
more than a tiny fraction of the older driver population. Furthermore,  it is
highly likely that those taking the course are in least need of improvement.

Pedestrians:  Interventions focused on the elderly pedestrian appear to make a
difference.  These include far-side bus stops,  separation of motor vehicle and
pedestrian traffic, cessation of all traffic in any direction during the
pedestrian signal,  enforcement  of existing laws and ordinances concerning
pedestrians,  improved vehicle design,  improved lighting of streets used by
pedestrians, increased conspicuity of pedestrians, and improved emergency medi-
cal services.

HOST FACTORS

In considering the increased impairment accompanying age, J. Wailer (81) dif-
ferentiates between changes that are associated with what he refers to as nor-
mative aging and issues involving pathological aspects of aging. While many of
these conditions can be identified through careful evaluation, some of them are
less readily detected so that increased risk of injury may occur despite efforts
to identify potential problems before they arise.
In addition to the older person's higher risk of collision involvement as a
driver or pedestrian,  it is clear that the elderly are more vulnerable to in-
jury from any given crash- Data from the Fatal Accident Reporting System (FARS)
show that a much lower proportion of older drivers survived with no injury and,

                                   D-13


conversely,  a higher proportion suffered fatai injury (72). Data from FARS as
reported by Partyka  (73)  also show that this increased probability of death
steadily increased with increasing age.  Partyka concludes that the increased
probability of fatality for the older driver is more related to decreased resis-
tance rather than to crash configuration per se. This finding and conclusion are
consistent with an earlier report by Baker.and Spitz (82).

One more consideration is the use of drugs including alcohol since the elderly
are more likely to be on medication,  and all too often they have been given no
medical counsel as to the potential effects on their functioning (77). An older
person of a given weight is likely to reach a higher blood alcohol concentra-
tion (BAC) from a specified quantity of alcohol than a younger person of the
same weight (83,84).  Older persons may recognize this change, for it appears
that they modify their alcohol intake (84,85). While frequency of drinking chan-
ges little with age,  the quantity of alcohol consumed tends to decrease.

The major concern in dealing with host factors is that chronological age is not
a good measure of the aging process itself. While there are clear relationships
between chronological age and the overall average performance of the elderly,
the predictions for individuals are not that reliable. As a result, interven-
tions must be carefully designed to accomplish what is needed while avoiding any
discrimination or appearance of discrimination against the elderly.

INTENTIONAL INJURIES: SUICIDE AND HOMICIDE


SUICIDE:  Reports in the scientific literature have consistently concluded that
suicide (ICD-9 codes E950-E959) increases with advancing years (86-89). The
elderly (age 65 and older)  constitute about 12% of the U.S. population but com-
prise 18.7% of all suicides. Moreover, this phenomenon is not limited to the
United States,  but is a characteristic of all industrialized nations.

The United States ranks at about the median for elderly male suicides and at the
lowest for suicides among the female elderly (90). Actually, the suicide rate
among the elderly has been decreasing over the past twenty years. However, older
persons still are at the greatest risk of suicide despite this relative improve-
ment (91).

The suicide attempts of older persons tend to be more serious and dangerous from
a psychological and a medical point of view (92).  The first attempt is likely
to be successful, and gestures or attempts as cries for help are relatively rare.
The elderly tend to use methods that are deadly and certain such as firearms
(employed by 64% of the elderly,  57% of the younger cases) and to avoid those
methods such as drug overdoses or gas which allow greater opportunity for in-
tervention and rescue (14% of elderly suicides vs.  21% for youthful suicides).

However, suicide rates do not increase with increasing age for all elderly. Rates
increase only for white males while they actually decline for black males and
for females of both races. Accordingly, prevention measures might be focused on
reduction of suicides among white males,  the high-risk population, although
prevention efforts might benefit all groups.

Age-related Factors

Phvsical health Sainsbury's  study (88) of coroner's cases  (N = 409) indicate
that physical illness was a factor in 35% of the cases over sixty years of age
as contrasted to the youthful cases  (ages 15-29)  in which illness figured in
only 10%. Dorpat, Anderson and Ripley (93) found that in their sample of 80 cases
of completed suicide, physical illness was the most frequent precipitating fat-
tor by those who were over sixty years of age. The reality of increased physi-
cal illness in the elderly  leads one to consider the question of voluntary
euthanasia and "the right to die" via so-called "rational suicide." The dilem-
ma over whether a person in a free society enjoys the right to take his own life
is fraught with emotional conflict almost as pointed as that of the counterpart
"right to life"  controversy (94,95). As a result of our increasing medical,

                                      D-14


pharmacological and technological sophistication, life can be extended quantita-
tively at the same time it may deteriorate qualitatively.
Despite the reality of physical decline in later life and the heightened risk
of suicide therein, certain preventive steps are suggested. First, a study by
Motto and Greene (96) found that 42% of a cohort of 175 suicides had some type
of medical contact within the six months preceding their  suicide.  Studies by
Robins and Vail have confirmed this finding (97). Educational efforts aimed
toward this population by physicians and the medical community could be extreme-
ly effective in diagnosis, referral and even treatment of the susceptible elder-
ly patient. Secondly, changes in the budget priorities of this nation that direct
greater funding for human and social services  could relieve the anxiety that
medical expenses can quickly wipe out a lifetime of savings.
Mental Hea1t.h The incidence of clinical depressive illness preceding serious
suicide attempts is quite high among older persons (92,98,99). Research also in-
dicates that the frequency of completed suicides is linked to depressive disor-
ders more than any  other form of mental disorder (97). Typical depressive
conditions in the person contemplating suicide are rarely of full-blown psychotic
dimensions,  but rather an individual with no previous history of serious emo-
tional problems whose defense mechanisms of a lifetime are now insufficient to
withstand the combination of biological,  psychological and social stresses of
old age. Butler and Lewis (100) enumerate some of the major diagnostic considera-
tions for suicide in old age in addition to physical illness and being a white
male as discussed earlier.  These risk factors include depression, withdrawal
from social and personal relationships (101), bereavement, especially within the
first year of a loss (102), isolation due to widowhood,  illness or disengage-
ment (103), expectation of death from some cause, not necessarily suicide (93),
induced helplessness often stemming from removal from home and placement in an
institutional setting (104) alcoholism (105), and hopelessness (99), which is
one of the most serious diagnostic signs. The feeling of exhaustion and failure,
the fear of being a burden to others, financially and otherwise, the loss of in-
terest in pleasurable activities are all symptoms of high suicide potential and
their overt expression must be taken very seriously.
Treatments for depression include anti-depressive medications, milieu therapy
such as the establishment of caring social support networks, and psychotherapy
of a supportive nature.
Social forces play a critical role in suicide prevention.  In our society, a
premium is placed upon appearance, occupational status,  and power and financial
standing, attributes which almost always decline with age. And yet, as previously
noted,  there is a significant exception to the finding that suicide increases
with age, sven in our society. An increase is found only among white males. Con-
versely, the rates for blacks, Native Americans,. and Mexican-American males and
for both white and black females actually decline in later years. Are there posi-
tive factors that inhibit suicide in certain communities? Several explanatory
hypotheses have been proposed (106):
&I-& of the elderly differs noticeably in the white community and communities
of other races.  Families in cultures that tend to be extended rather than nuclear
provide more interaction and purposeful activity- for the elderly, such as child
care,  housekeeping,  and food preparation.  Although dictated by necessity, the
net result is to establish and strengthen social bonds which Durkheim (86) con-
sidered the greatest bulwark against the commission of suicide. Mar-y white elder-
ly, because of economic affluence, live apart a.nd alone without the social nexus
and sense of participation found in more trarditionally orientec: societies.

Status of the elderly is also greater  in the communities of races other than
whites where wisdom and experience are honored,  and the elderly actually enjoy
an increase in status as they gKOW older (107-109). However, the trend is toward
changing these traditional values when societies industrialize and modernize.

                                    D-15


               . .
&&tive degrlvaw is a concept that helps expiain the great plurality of white
male suicides at the upper ages. White males,  as a group,  have enjoyed the
greatest advantages and the greatest prestige during their active years. Thus,
they have the farthest to fall when confronted with the.vicissitudes of aging.
Presumably the white male who begins to feel his power and influence eroding ex-
periences a greater sense of loss than women or minorities who have long been
accustomed to a lowered status due to racism,  sexism and economic hardship.
Therefore,  women and minorities do not have to suffer these losses in old age.

What are the policy and theoretical implications of these findings? First in-
dividuals must learn to be realistic in their goals and aspirations since in-
ability to achieve these high-reaching aspirations may lead to depression and
other emotional problems as earlier discussed.  Second is an appreciation of
respect for age underscored by greater acceptance of the aging process and the
inclusion of the elderly in the family structure. But foremost is understanding
the importance of human relationships and response. The most distressingly dif-
ficult part of aging is not the material adversity nor the physical pains and
illnesses,  but rather the feeling of being useless, unwanted, superfluous. Com-
munities that deal successfully with these issues,  through necessity as much as
design, provide a role,  status and purpose for their elderly population.

HOMICIDE:  In 1983, the latest year for which fully reliable and complete vital
statistics are available, 20,141 people died by homicide (ICD-9 codes E960-E969)
in the United States. Of these,  some 1,244 cases or 6.2% were elderly (65 years
and older). Since the elderly constitute about twelve percent of the general
population it is clear that they are underrepresented by half in the population
of homicide victims.

Homicide is very much a condition of the impetuous,  youthful years of life. The
rate characteristically declines after the first year of life and then begins
accelerating through the teen years,  reaching a peak at ages 25 to 35 and declin-
ing steadily from that point with very minor fluctuations.  Males run a risk about
3 to 4 times that of females with the exception of infanticides during the first
year of life when the victim is more likely to be female. Blacks have a homicide
risk five times greater than that for whites and the risk for young black males
can only be described as astronomical.  These relative race and sex differences
are perpetuated throughout the lifespan; however,  the absolute numbers are con-
siderably reduced by the time one reaches later life.  In addition to these quan-
titative differences there are qualitative distinctions between homicide in the
elderly as contrasted to the general homicide picture. For instance, the dis-
tribution of homicide by method indicates that the elderly are much more like-
ly to be killed by beating or strangling than are their younger counterparts.
The predominant motive is robbery and, therefore,  they are much less likely to
know their assailant (110).  Preventive programs or even research studies on the
topic of elderly homicides are extremely scant (110). If homicides are considered
within%he context of elder abuse,  there is slightly more to review including a
thorough overview by Pedrick-Cornell and Gelles (111) and a systematic study
recently published by Powell and Berg (112). However,  neither of these sources
deal with the problem of homicide per se but rather with the topic of physical
abuse of a nonlethal nature.  The neglect of this topic in the literature may
reflect its underrepresentation  in the elderly population.  (The incidence of
homicide is less than one-fourth of the suicide incidence for this age group.)
Accordingly the following recommendations are proposed:
A-It is a truism that the odds of being a victim of a violent crime are tied to
neighborhood income. Poor people bear a disproportionate share of the crime bur-
den. For the most part this means the minority elderly who are trapped by pover-
ty and discrimination in areas described as subcultures of violence. This entails
a subculture in which members are recruit?ed into a vicious cycle of   street
crimes followed by prison terms where antisocial behavior patterns are rein-
forced, encouraging more violence (113). This vicious cycle will only be remedied
when we address the conditions that underlie and perpetuate it (114).

                                   D-16


8. Since the elderly are perceived by the criminally motivated as representing
vulnerable prey, any moves toward strengthening their position through increased
social networks (strength in numbers),  assertiveness training and political
demands for increased police protection will be helpful in changing the percep-
tion of vulnerability. However,  this cannot be completely overcome since the
elderly are realistically more vulnerable due to their relative frailty. There
are strong implications here for community education programs directed toward
the elderly. Most elderly homicides occur in the victim's home. TO decrease the
possibility of having to open their doors to strangers,  the installation of an
inexpensive optical viewer in the front door would be a sufficient preventative
step which could be underwritten by community agencies. Other strategies might
involve the establishment of mutual aid networks through telephone linkages and
the provision of an "eyes on the street" crime prevention program which would
have the added benefit of fostering closer interpersonal relationships and emo-
tional support. Such approaches have the advantage of reducing robberies which
are the primary ways  in which the elderly are murdered and also of reducing
through neighborhood contact, the isolation and loneliness which frequently ac-
company aging-

REFERENCES

l.Baker S, O'Neill B, Karpf RS.  The Injury Fact Book.  Cambridge, MA: Lexington Books, 1984.
2.Hongladarom GC, Miller WF, Jones JM, et al.  Analysis of the Causes and Prevention of Injuries
  Attributed to Falls.  Office of Environmental Health Programs, Department of Social and Health
  SerViCtS3.  Olympia.  Washington, 1977.
3.Kelsey JL, Hoffman S.  Risk Factors for Hip Fracture.  NEJM 1987; 316(7):404-406.
4.Grazier KL, Holbrook TL, Kelsey JL, et al.  The Frequency of Occurrence, Impact, and Cost of Mus-
  culoskeletal Conditions in the United States.  Chicago: American Academy of Orthopedic Surgeons,
  1984.

5.Melton LJ, Riggs BL.  Risk Factors for Injury after a Fall.   Clinics in Geriatric Med

6.Nevitt MC,  Cummings SR,  Kidd S, et al.  Risk Factors far Recurrent Falls in the Ambulatory Elder-
  ly-  Unpublished paper, 1987.

7.Gryfe C, Amies A, Ashley M.  A Longitudinal Study of Falls in an Elderly Population: I. Incidence
and Morbidity. Age and Ageing 1977; 6:201-210.

8.Baker SP, Harvey AH. Fall Injuries in the Elderly.  Clinics in Geriatric Medicine. Philadelphia:
W.B. Saunders, 1985; 1(3):501-512.

7.Berry G, Fisher R, Lang S.  Detrimental Incidents,  Including Falls in an Elderly Institutional
Population.  J Am Ger Sot 1981;29(7):322-324.

lO.Morris E, Isaace B.  The Prevention of Falls in a Geriatric Hospital. Age and Ageing 1980;9:181-
  185.

11,National Oenter for Health Statistics.  Unpublished data from the 1984 Health Interview Survey:
Supplement on Aging.

12,Nickens H.  Intrinsic Factors in Falling among the Elderly.  Arch Int Med 1985;145:1089-1093.

13.Venqlarik JM, Adams M.  Which Client Is a High Risk? J Gerontol Nursing 1983; 11(5):28-30

14.Cummings SR, Nevitt MC, Kidd S.  Forgetting Falls.  Unpublished paper, 1987.
15.Radebauqh TS, Hadley E, Suzman R (eds.). Falls in the Elderly: Biologic and Behavioral Aspects.
  Clinics in Geriatric Medicine.  Philadelphia: W.B. Saunders, Vol. 1, August 1985.
16.Lipsitz L.  The Drop Attack: A Common Geriatric Symptom.  J Am Ger Sot 1983;31(10):617-620.

17.Lipsitz L.  Syncope in the Elderly.  Ann Int Med 1983;99:92-105
18.Sabin T. Biologic Aspects  of Falls and Mobility Limitations in the Elderly.  J Am Ger SOC
1982;30(1):51-58.

19.Kapoor WN.  Evaluation of Syncope in the Elderly.  J Am Ger Sot 1987;35:826-828.

20.Tinetti ME, Williams FT, Mayewski R.  A Fall Risk Index for Elderly Patients Based on Number of
  Chronic Disabilities.  Am J Med 1986;80:429-434.

21.Wolfson LI, Nhipple R, Amerman P.  Gait and Balance in the Elderly:  Two Functional Capacities
that Link Sensory and Motor Ability to Falls.  Clinics in Geriatric Medicine.  Philadelphia: W.B.
Saunders 1985;1(3):649-660.

D-17


22.Isaacs B. Clinical and Laboratory Studies of Falls in Old People: prospects for Prbvention.
  Clinics in Geriatric Medicine.  Philadelphia: W.B. Saunders 1985;1(3):513-524.
23.Brocklehurst JC, Exton-Smith AN, Lempert-Barber SM,  et al. Fracture of the Femur in Old Aye: A
Two-centre Study of Associated Clinical Factors and the Cause of the Fall.   Age and Ageing
  1918;1:2-15.

24.Wild D, Nayak U, Iaaacs B.  Description,  Classification and Prevention of Falls in Old People at
  Home.   Rheum Rehab 1981;20:153-159.

25.Campbell A, Reinken S, Allan B, et al.  Falls in Old Age: A Study of Frequency and Related Clini-
  cal Factors.  Age and Ageiny 1981;10:264-270.
26.Lipsitz L, Nyquist RF, Wei 3, et al.  Postprandial Reduction in Blood pressure in the Elderly.
N En91 J Med 1983;209(2):81-83.

27.Lipsitz LA.  Abnormalities in Blood Pressure Homeostasis that Contribute to Falls in the Elder-
  ly.  In Falls in the Elderly: Biologic and Behaivoral Aspects. Clinics in Geriatric Medicine.
Philadelphia: W.B. Saunders 1985;1(3): 637-648.

28.Imms F,  Edholm 0. Studies of Gait and Mobility in the Elderly.  Age and Ageing 1981;10:147-156.

29.Woollacott MH, shumway-Cook A,  Nashner L. Postural Reflexes and Aging.   In Mortimar JA, Piroz-
  solo FJ and Maletta GJ (eds): The Aging Motor System.  New York: Praeger Publishers, 1982.

30.Robbins LJ.  Restraining the Elderly Patient.  Clin Ger Med 1986;2(3):591-599.

31.Wild D, Nayak U, Isaacs B.  HOW Dangerous Are Falls in Old People at Home? Brit Med J1981;282:266-
  268.

32.Tideiksaar R,  Kay AD.   'what causes  Falls? A Logical  Diagnostic Procedure.    Geriatrics
  1986;41(12):32-50.

33.Prudham D, Evans J.  Factors Associated with Falls in the Elderly: A Community Study.  Age and
Ageing 1981;10:141-146.

34.Tinetti ME. Performance-Oriented Assessment of Mobility Problems in the Elderly. J Am Ger Sot
  1986;34(2):119-126.

35.Hogue C.  Mobility.  In Schneider C, et al.(eds).  The Teaching Nursing Home: A New Approach to
  Geriatric Research, Education,  and Clinical Care.  New York: Raven Press, 1984.

36.Brody EM, Kleban MH, Moss MS, et al.  Predictors of Falls Among Institutionalized Women with
  Alzheimer's Disease.  J Am Ger Sot 1984;32(12):877-882.

37.Isaacs 8.   Are Falls a Manifestation of Brain Failure?  Aye and Ageing 1978;7 (Supplement):97-
  111.

38.Visser H.  Gait and Balance in Senile Dementia of Alzheimer's Type.  Age and Ageing 1983;12:296-
  301.

39.Buchner DM,  Larson EB. Falls and Fractures in Patiennts with Alzheimer's Type Dementia. JAMA
1987;257(11):1492-1495.

40.Mossey JM. Social and Psychologic Factors Related to Falls among the Elderly. Clinics in Geriatric
Medicine. Philadelphia: W.B. Saunders 1985;1(3):541-554.

41.Ray WA,  Griffin MR,  Scchaffner W, et al.  Psychotropic Drug use and the Risk of Hip Fracture. N
Eng J Med 1987;316(7):363-369.
42.Granek E, Baker SP, Abbey H, et al.  Medications and Diagnoses in Relation to Falls in a Long-
Term Care Facility.   J Am Ger Sot 1987;35:503-511.
43.Sheldon i".  On the Natural History of Falls in Old Age.   Brit Med J 1960; 5214-1690.
44.Lucht D. A Prospective Study of Accidental Falls and Re'sulting Injuries in the Home among Elder-
ly People.  Acta Sociomed Stand 1971;2:105-120.

45.Waller J. Falls among the Elderly--Human and Environmental Factors.   Act Anal Prev 1978;10:21-
  33.

46.Stelmach CD, Worrinyham CJ.  Sensorimotor Deficits Related to Postural Stability: Implications
for Falling  in the Elderly.    Clinics  in Geriatric Medicine.  Philadelphia:  W.B.  Saunders,
1985;1(3):679-694.

47.Mathias S, Nayak USL.  Issaacs B. Balance in Elderly Patients: The "Get-up and Go" Test.   Arch
Phys Med Rehab 1986;67:387-389.

48.Nasher LM. Black FO, Wall C. Adaptation to Altered Support and Visual Conditions During Stance:
  Patients with Vestibullar Deficits.  J Neuroscience 1982;2:536-544.

49.Sekuler R, Kline D,  Dismukes K (eds). Aging and Human Visual Function.  New York: Alan R. Liss,
  Inc.,  1982.

SO.Applegate WB, Miller ST, Elam JT.  Impact of Cataract Surgery with Lens Implementation on Vision
and Physical Function in Elderly Patients.  JAMA 1987; 257(8):1064-1066.

Sl.Overstall P.  Prevention of Falls in the Elderly.  J Am Ger Sot 1980;28(11):481-483.

52.Rubenstein L, Robbins A. Falls in the Elderly: A Clinical Perspective.  Geriatrics 1984;39(4):67-
  78.

D-18


53.Wells BG, Middleton B, ~arrence G, et al.  Factors Associated with the Elderly Falling in Inter-
   mediate Care Facilities.  ~r,.,~ Intelligence and clin Pharm 1985.
54.Foley CJ,  Wolf-Klein GP. Prevention of Falls in the Geriatric Patient with OsteoporosiS. Clin
  Rheumat Pratt 1986;136-143.
55,~apoor W, Snustad D, Peterson J. et al.  qrncc,pe in the Elderly. Am J Med 1986;80:419-429.

56. Hornbrook MC, Wingfield DJ. Falls Hazards in the HOIIBS of the Elderly.   Presentation at Annual
  Meeting of Gerontological Society of America.  Chicago: November 22, 1986.

51 .Obonyo T,  Drummond M, Isaacs B.  Domiciliary Physiotherapy for Old People Who Have Fallen.   1nt
  Rehab ued 1984;5:157-160.

58.Fife DD, SolomOn P, Stanton M.  A Risk/Falls Program: Code Orange for Success. Nursing Manage
  1984;15(11):50-53.

59.Rainville NG.  Effect of an Implemented Fall Prevention Program on the Frequency of Patient Falls.
  dual Rev Bull 1984;10(9):287-291.

60.Gray-Vickrey M.   Education to Prevent Falls.  Geriat Nursing 1984;5(3):179-183.

61.Baker SP, O'Neill B, Karpf, RS.  The Injury Fact Book. Lexington,  Mass.: D.C. Heath and Company,
   1984.

62.National Safety Council. Accident Facts.  Chicago: National Safety Council, 1986.

6i'.Planek, TN.  The Aging Driver in Today's Traffic: A Critical Review. In Wailer PF (ed).   North
  Carolina Symposium on Highway Safety: The Elderly in a Mobile Society.  Fall 1972;7:3-38.

64.Wiener EL.  Elderly Pedestrians and Drivers: The Problem That Refuses To Go Away.   In Waller PF
  (ed). North Carolina Symposium on Highway Safety: The Elderly in a Mobile Society 1972;7:53-95.
65.Waller PF, Reinfurt DW.  The Who and When of Accident Risk: Can Driver License Programs Provide
  Countermeasures?  Chapel Hill: UNC Riqhway Safety Research Center, 1973.

66.Waller PF, House EG, Stewart JR. An Analysis of Accidents By Age. Chapel Hill: UNC Highway Safety
  Research Center, 1977.

67.Joksch HC, Thoren S.  Car Size and Occupant Fatality Risk,  Adjusted for Differences in Drivers
  and Driving Conditions.  Hartford, CT:   The Center for the Environment and Man, Inc., January
  1984.

68.Maleck TL, Hummer JE.  Driver Age and Highway Safety.  In: Transportation Research Record 1059.
Washington: Government Printing Office, 1986.
69.Tobey HN, Shunamen EM. KnoblauchRL. Pedestrian TripMakinqCharacteristics andExposure Measures.
  Washington:  U.S. Department of Transportation, 1983.

7O.Waller PF.  Preventing Injury to the Elderly.  In: Phillips HT, Gaylord SA (eds) Aging and Public
  Health.

71.Sivak M, Olson PL, Pastalan LA. Effect of Driver's Age on Nighttime Legibility of Highway Signs.
  Human Factors 1981;23(1):59-64.

7Z.Brainin PA.  Safety and Mobility Issues in Licensing and Education of Older Drivers. Washington:
National Highway Traffic Safety Administration, 1980.
73.Partyka S. Comparison of Age of Drivers in Two Car Fatal Crashes. Washington: National Highway
Traffic Safety Administration, 1983.
74.Ashton SJ, Peddler JB, Mackay GM.  Pedestrian Head Injuries. In:  Huelke DF (ed). Proceedings
American Association for Automotive Medicine,  22nd Conference and the International Association
for AccideQt and Traffic Medicine VII Conference.
75.Ichikawa K, Tanaka K, Kamiya H. Living Environment and Design of Woonerf. International Associa-
tion of Traffic and Safety Sciences Research 1984;8:40-51.

76.Ysander L, Herner B. Elderly Male Automobile Drivers in Gothenburg and Their Traffic Behavior in
  the Year 1971.  Proceedings of First International Conference on Driver Behavior.

77.Yee D.  A Survey of the Traffic Safety Needs and Problems of Drivers Ages 55 and Over.   In: Mal-
fetti JL (ed). Drivers 55+ Needs and Problems of Older Drivers:  Survey Results and Recommenda-
  tiona.  Ed. by J.  L. Malfetti, 96-128.  Virginia: AAA Foundation for Traffic safety, 1985.
78.Planek TN, Condon ME, Fowler RC.  An Investigation of the Problems and Opinions of Aged Drivers.
Chicago: National Safety Council, December 1968.

79. Zaidel DM, Hocharman,  I. License Renewal for Older Drivers:   The Effects of Medical and Vision
  Tests.  Jnl Safety Research 1986;17:111-116.

BO.Seaton M.  55 Alive/Mature Driving. Proceedings of Twenty Third Conference of the American As-
  sociation for Automotive Medicine, 1979.

81.Waller JA.  Injury Control.  A Guide to the Causes and Prevention of Trauma. Lexington, Mass.:
D.C. Heath and Company,1985.

82.Baker SP, Spitz WU.  Age Effects and Autopsy Evidence of Disease in Fatally Injured Drivers. JAMA
  1970;214:1079-1088.

                                    D-19


83.Veetal RB, McGuire EA, Tobi JD, et al. Aging and Ethanol Metabolism.  Clin Pharmacol Therapeutics
  1977;21:343-354.

84.Voqel-Sprott M, Barrett P. Age, Drinking Habits, and the Effects of Alcohol.   J Studies Alcohol
  1984;45(6):517-521.

85.Stall R. Change and Stability in Quantity and Frequency of Alcohol Use Among Aging Males:  A lP-
  Year Follow-up Study.  Brit J Addiction 1986;81:537-544.

86.Durkheim E.   Suicide.   Glencoe, IL:  Free press, 1951.   (Original French edition of Le Suicide
  published in 1897.)

87.Dublin LI.  Suicide: A Sociological and Statistical Study. New York: Ronald Press, 1962

88.Sainsbury P.  Social and Epidemiological Aspects of Suicide with Special Reference to the Aged.
  In: Williams RH, Tibbits C, Donahue W (eds). Processes of Aging:  Social and Psychological Perspec-
  tives (Vol. II). New York: Atherton, 1963.

89.Weiss JMA.  Suicide in the Aged.  In: Resnik HLP (ed). Suicidal Behaviors.  Boston: Little Brown,
  1968.

90.Statistical Abstract of the United States:  1987 (107th ed.). Washington: US Bureau of the Cen-
  sus,  1986, p. 823.

9l.Seiden RR, Freitas Rp.  Shifting Patterns of Deadly Violence.  Suicide and Life-threatening Be-
  havior 1980;10(4): 195-209.

92.O'Neal P, Robins E, Schmidt E.  A Psychiatric Study of Attempted Suicide in Persons over Sixty
  Years of Age.  Arch Neural Psych 1956;75:275-284.
93.Dorpat TL, Anderson WF, Ripley HS.  The Relationship of Physical Illness to Suicide.   In: Res-
  nik HLP (ed) Suicidal Behaviors.   Boston:  Little Brown,  1968, pp. 209-219.

94.Battin Hp.  On the Relationship Between Suicide-prevention and Suicide-advocacy Groups.  Suicide
  and Life-threatening Behavior 1982;12(4):254-260.

95.Seiden RH.  Self-deliverance or Self-destruction?  Euthanasia Review 1986;1(1):48-56.

96.Motto JA, Greene C.  Suicide and the Medical Community.  Arch Neural Psych 1958;80:776-781.

97.Rabins E, Murphy GE, Wilkinson RH Jr, Gassner S, F&yes J.   Some Clinical Considerations in the
  Prevention of Suicide Based on a Study of 134 Successful Suicides.  AmJPublicHealth 1959;49:888-
  899.

PB.Batchelor IRC,  Napier M. Attempted Suicide in Old Age.   Brit Med J 1953;2:1186-1190.

99.Farbedrou NL, Shneidman ES.  Suicide and Age.   In: Schneidman ES, Farberow NL, Litman RJZ. The
  Psychology of Suicide.  New York: Science House, 1970, pp. 165-174.

100. Butler RN, Lewis MI.  Aging and Mental Health. St. Louis: C.V. Mosby, 1977

101. Cumminqs E,  Henry WE. Growing Old: The Process of Disengagement.  New York: Basic Books, 1961.

102. Holmes TH, Rahe RH.  The Social Readjustment Rating Scale.  JPsychosomaticResearch1967;11:213-
  218.

103.Galle OR, Gove WR. Overcrowding, Isolation and Human Behavior. In: Taueber KE, Bumpass LL, Sweet
  JA (eds). Social Demography.   New York:  Academic Press,  1978, pp. 95-132.

104.Braceland F. Psychological Aspects of Aging. In: Vedder CB,  Lefkowitz AS (eds). Problem of the
Aged. Springfield,  Il.: Charles C Thomas . Publisher, 1965, pp. 203-214.
105.Seiden RH. Suicide: Behavioral Aspects.  In: The Encyclopedia of Crime and Justice. New York:
Free Press, 1983, pp, 1521-1526.
     >
106.Seiden RB. Mellowing with Age: Factors Influencing the Nonwhite Suicide Rate. Int J Aging Human
  Develop 1981;13(4):   265-284.

107.Messer U. Race Differences in Selected Attitudinal Dimensions of the Elderly. Gerontologist
  1968;8(4):245-449.

108.Robina L, West P, Murphy G.  The High Rate of Suicide in Older White Men.  Sot Psych 1977;12:1-
  20.

lOP.Wylie FM.  Attitudes Toward Aging and the Aged Among Black Americans:  Some Historical Perspec-
  tives. Aging and Human Development 1971;2:66-70.

llO.Allen N. Homicides in the Elderly.  Unpublished research report,  U.C.L.A. Neuropsychiatric In-
  stitute, 1985.

lll.Pedrick-Cornell c, Gellee RJ.  Elder Abuse: The State of Current Knowledge.  Family Relations
  1982;31:457-465.

112.Powell S, Berg RC. When the Elderly Are Abused:  Characteristics and Intervention. Ed Gerontol
  1987;13:71-83.

113.Wolfgang ME, Ferracuti F. Subculture of Violence--A Psychological Theory.  In: Wolfgang ME (ed).
  Studies in Homicide.   New York:  Harper and Row,  1967, pp. 271-280.
114.Kastenbaum R, Aisenberq R.  The Psychology of Death.  New York: Springer, 1972.

                                   D-20