• Improving Dietary Self-Monitoring and Adherence with Hand-Held Computers: A Pilot Study
  Karen Glanz

• Pocket PC-Delivered Interventions to Improve Physical Activity and Dietary Behaviors in Adults at Least 50 Years Old
  Brian Oliveira

• Context-Sensitive Measurement of Physical Activity
  Julie Wright

• A Validation of Mood Adjectives in Ecological Momentary Assessment (EMA) Using Two Cancer Populations
  Hoda Badr

• Cortisol and Mood in Breast Cancer Survivors: Baseline Levels and Responses to Mammography Screening
  Laura S. Porter

• Electronic EMA with Tobacco-Dependent Cancer Patients Scheduled for Surgery
  Jack E. Burkhalter

• Negative Affect and Smoking Lapses: A Prospective Analysis
  Andrew J. Waters

• The Effects of PDA Vs. Paper-Based Monitoring on Recording Compliance and Nicotine Gum Use Among Smokers Attempting to Quit.
  William T. Riley

• Using Electronic Diaries for Real-Time Data Capture in Clinical Research
  Alan L. Shields

• Proximal Determinants of Temptations to Drink Among Alcoholics Discharged From Inpatient Treatment: Application of Real-Time Monitoring Methodology
  Alan L. Shields

• Craving and Affect as Antecedents of Cigarette Smoking in Alcoholic Smokers
  Ned L. Cooney

• The Theory of Self-Control Strength Applied to Smoking Cessation
  Kathleen A. O'Connell

• Signaling Does Not Adequately Improve Diary Compliance
  Joan E. Broderick

• End-to-End Communications Infrastructure for Real-Time Clinical Trial Operations, Administration and Management
  Augustine Samba

• Real-Time Data Collection and Quality Assurance / Quality Control in the National Health and Nutrition Examination Survey
  Lewis Berman

• Daily Experiences of Sense of Control and Smoking Behavior in Adolescents
  Thomas A. Wills

• Development of a New Watch-Type Computer for Ecological Momentary Assessment and Application to CFS Patients
  Kazuhiro Yoshiuchi

• Wearable Relational Devices for Real-Time Annotation of Physiological Data
  Karen K. Liu

• Development of an Electronic Diary Method for Use with Older Adult Medical Patients and Caregiving Spouses
  Mary Ann Parris Stephens

• Daily Process Research on Cognitive Therapy for Depression
  Lawrence H. Cohen

• The Reciprocal Relationship Between Drinking and Guilt: Studying Processes Over Time Using Ecological Momentary Assessment
  Mark Muraven

• Feasibility of Daily IVR Symptom Monitoring Among Patients with an Alcohol Use Disorder and PTSD
  Tracy L. Simpson

• Uses of Electronic Monitoring to Improve Adherence to Prescribed Medications
  Marc I. Rosen

• Theoretical and Practical Issues in the Measurement of Momentary Self-Reports
  Tamlin C. Christensen

Authors:
Suzanne Murphy Joanne Moylan, Pitt County Health Department, Pittsboro, NC Diana Evensen J. David Curb

Abstract:
BACKGROUND. Interventions that use personalized, or tailored, feedback, hold great promise for promoting health behavior change. Innovations in computerized information technology offer new opportunities for creative applications of tailored strategies for improving adherence in clinical trials. METHODS. This pilot study involved developing and testing a diet monitoring and feedback system using hand-held computer technology, or Personal Digital Assistants (PDA's), for real-time dietary assessment. Participants were women in the Dietary Modification arm of the Women's Health Initiative (WHI). After conducting focus groups with 29 women to determine the preferred monitoring and feedback features, 36 women used the PDA-based system for one month, and received both immediate and weekly tailored feedback. ANALYSIS. Qualitative analysis of focus groups. Descriptive data and comparison of pre-post responses of participants. RESULTS. During that month participants significantly increased self-monitoring, improved their attitudes toward self-monitoring, and met their dietary goals more often. Reported total fat intake and percent energy from fat decreased significantly, and there was a slight decrease in total caloric intake. DISCUSSION. The pilot study findings support the largely untapped potential of hand-held computers for improving diet monitoring and diet adherence.

Funding Source:
Developmental Funds, University of Hawaii Cancer Research Center

Disclosures:
None. No financial conflict of interest

Authors:
Brian Oliveira, Abby King, Christopher Gardner, & Jessica Morton, Stanford Center for Research in Disease Prevention, Palo Alto, CA, USA (all)

Abstract:
B. Research in Progress or Research in Development: Background: Physical activity and diet are important factors that can impact cancer and other chronic diseases. Making healthy changes to improve physical activity and dietary behaviors can be very challenging, especially among older adults who may be most susceptible to these chronic diseases. Handheld computer technology may offer a useful new approach for delivering physical activity and dietary interventions that is both convenient and cost effective. Methods: The efficacy of using handheld computers in delivering an intervention strategy designed to increase targeted physical activity and specific dietary intake among 80 adults at least 50 years old is being evaluated. In two separate 8-week, randomized, controlled studies, physical activity (n=40) and nutrition (n=40) interventions are being delivered via handheld computers to adults aged at least 50 years old. It is expected that those participants who are randomly assigned to the computerized intervention approach will show greater increases in targeted physical activity and specific dietary intake when compared to the participants assigned to an educational, "usual-care" approach. Analysis Plan: Post-intervention differences on standardized measures of physical activity and dietary intake for participants in the computerized intervention and the "usual-care" approaches will be evaluated. Also, process data from the handheld computer will be analyzed. Discussion: It is expected that these efficacy studies will demonstrate that handheld computers can be used to improve physical activity and dietary behaviors. Data from these studies will be used to inform larger-scale studies in this area.

Funding Source:
Stanford University Office of Technology Licensing Intramural Grant

Disclosures:
None

Authors:
Robert H. Friedman, MD, Boston University, Boston, MA, USA. Stephen Intille, PhD, Massachusetts Institute of Technology, Cambridge, MA. Abby King, PhD, Stanford University, Palo Alto, CA. William Haskell, PhD, Stanford University, Palo Alto, CA. Susan Blake, PhD, George Washington University, Washington DC. Jeffrey Migneault, PhD, Boston University, Boston, MA.

Abstract:
Background: Existing measurement methods for physical activity (PA) in free-living populations which include self-report, direct observation and special purpose devices (pedometers, accelerometers, and heart rate (HR) monitors have significant limitations (reliability, validity, completeness, fidelity, cost). Methods: The goal is to develop and test a Context-Sensitive Ecological Momentary Assessment (CS-EMA) system. CS-EMA consists of a PDA connected wirelessly to a HR monitor which continuously transmits data to the PDA where it is stored and analyzed to estimate PA intensity. MOD+ PA triggers a CS-EMA prompt (PDA beep) for the subject to enter PA diary data about the period (e.g., PA type, duration, perceived intensity). To evaluate CS-EMA, 25 housekeepers and 25 materials handlers will use CS-EMA PDA during working hours for one week. They will use CS-EMA and a paper-and-pencil PA diary on different consecutive 2 days, in random order. During all four days they will be monitored by a trained PA observer. On the fifth day, the subjects will only be monitored by HR and accelerometers to evaluate reactivity. Analysis Plan: We will compare CS-EMA and paper diary recording to the gold standard of the observer. We will assess their satisfaction with the two measurement methods, their adherence to the reporting protocol and their reactivity to the measurement methods and the observer. Discussion: In future work we will use data gathered during the study to create algorithms for improving MOD+ PA recognition and for recognizing specific PA types. These algorithms will be incorporated into a second generation CS-EMA system.

Funding Source:
NA

Disclosures:
None

Authors:
Hoda Badr, Ph.D., Karen Basen-Engquist, Ph.D., Cindy Carmack Taylor, Ph.D., University of Texas M.D. Anderson Cancer Center Carl de Moor, Ph.D. University of Texas School of Public Health

Abstract:
Background: Identifying valid and reliable brief instruments for EMA is important - particularly when assessing individuals coping with the ongoing sequelae of cancer. In two samples, the validity of a 16-item EMA measure (Affleck et al., 2000) based on the Mood Circumplex was examined. Comparable validity of a reduced-item mood measure was also explored. Methods: Four times daily 18 breast cancer survivors (post-treatment) and 42 ovarian cancer patients (receiving chemotherapy) entered information in palm-top computers regarding mood and physical symptoms. Analysis Plan: For each sample, to confirm the mood measure factor structure, principal components analyses with varimax rotation were performed using centered adjectives. A two-factor solution was sought for each set of 8 adjectives. Correlations between centered pain and fatigue scores and mood were also examined. Results: In both samples, adjectives loaded on their predicted components (>.60 or <-.60), accounting for 51-58% of the variance. A condensed 8-item measure comprised of opposite adjective pairs with the greatest correlation on each factor performed comparably. Bipolar mood vectors comprising four adjectives apiece were computed with higher scores indicating positive mood. In the ovarian sample, correlations of the mood vectors with pain and fatigue were significant (r = -.10 to -.57); for the breast sample, 3 of the 4 mood vectors (r = -.17 to -.58) were significant. The condensed measure yielded correlations of similar magnitude. Discussion: This study supported the validity of mood adjectives in EMA in two cancer samples; however, a reduced number of adjectives may perform comparably well, reducing patient burden.

Funding Source:
NCI 1R21CA89519 (PI: Karen Basen-Engquist) "Breast Cancer Survivors, Physical Activity and Quality of Life" NCI R25 Training Grant CA57730 (PI: Robert M. Chamberlain, Ph.D)"M.D. Anderson Education Program in Cancer Prevention Education"

Disclosures:
None

Authors:
Laura S. Porter, Ph.D., Duke University Medical Center, Durham, NC; Merle Mishel, R.N., Ph.D., Virginia Neelon, R.N., Ph.D., Micheal Belyea, Ph.D., Etta Pisano, M.D., University of North Carolina; Mary Scott Soo, M.D., Duke University Medical Center

Abstract:
Background: For breast cancer survivors (BCS) repeated exposure to stressors such as mammograms, which trigger fears of cancer recurrence, may lead to important biologic effects as well as reduced quality of life. The purpose of this study was to examine mood and cortisol levels at baseline and in response to a mammogram in BCS compared to women without a history of cancer. Method: Subjects were 33 women 3-5 years post-treatment for breast cancer and 21 control women, all of who were scheduled for a routine mammogram. They provided saliva samples 6 times/day and mood ratings 3 times/day for 6 days. The first 3 days (baseline) occurred one month prior to the mammogram; the second 3 days were the day before, the day of, and the day following the mammogram. Analyses: Mixed linear models were used to analyze group differences in baseline levels of mood and cortisol, diurnal slopes of cortisol, and mood and cortisol responses to the mammogram. Covariates were age, smoking status, race, and education. Results: At baseline, BCS had significantly higher levels of cortisol than controls. There were no group differences in diurnal slopes of cortisol. There were group differences in cortisol responses to the mammogram: In BCS cortisol levels decreased between baseline and mammogram periods, while in control subjects cortisol levels increased. There were no significant findings with regard to mood. Discussion: BCS showed elevated levels of basal cortisol and suppressed cortisol response to a cancer-related stressor compared to controls, despite reporting similar levels of positive and negative mood. These findings suggest the importance of utilizing physiologic as well as subjective measures to assess stress and well being in cancer survivors.

Funding Source:
Lineberger Comprehensive Cancer Center, UNC UNC School of Nursing Grant RR00046 from the General Clinical Research Centers program of the Division of Research Resources, NIH

Disclosures:
None

Authors:
Jack E. Burkhalter, Ph.D. & Jamie S. Ostroff, Ph.D., Memorial Sloan-Kettering Cancer Center, New York, NY, USA

Abstract:
B. Research in Progress or Research in Development. For tobacco dependent cancer patients, smoking cessation decreases risks of cancer recurrence and second primary cancers. Despite these health benefits, 1/3 or more continue to smoke. Continued smoking is largely due to barriers (e.g., psychological distress) rather than lack of intention to quit. NCI has funded us to conduct a randomized, two-arm intervention trial to help recently diagnosed cancer patients quit smoking prior to surgery by testing whether the addition of an enhanced behavioral cessation strategy, scheduled reduced smoking (SRS), to hospital counseling and nicotine replacement therapy (Usual Care) improves smoking cessation outcomes. The cancer context provides an opportunity to understand the interplay of psychological distress, quitting self-efficacy, physical symptoms, social support, and smoking urges on adherence to SRS and cessation outcomes. EMA methodology via hand-held computers will guide SRS delivery and the planned process assessment. Methods: The SRS cues patients to smoke based on a personalized algorithm to taper smoking before surgery. EMA will be informed by item analysis of standardized scales used in prior work and subject burden, given the stresses of imminent surgery for cancer. Analysis: We will use HLM for analyzing repeated measures of within-subject variables impacting adherence to SRS and smoking outcomes and multivariate logistic regressions for examining binary smoking outcomes. Discussion: This study will advance understanding of the processes of behavior change in a cancer context by applying EMA methodology for data collection and intervention delivery. Rationale for item selection for assessing psychosocial constructs and sampling strategies will be discussed.

Funding Source:
National Cancer Institute, R01 CA090514

Disclosures:
None

Authors:
Andrew J. Waters, MD Anderson Cancer Center, Houston; Saul Shiffman, University of Pittsburgh, Pittsburgh

Abstract:
Background: Retrospective recall data on smoking relapse episodes indicate that stress and negative affect (NA) promote relapse, but these data may be biased. We report an EMA-based prospective analysis of stress and NA prior to initial lapse episodes. Methods: 215 smokers quitting smoking carried an electronic diary, monitoring quit experiences for up to four weeks after cessation. Momentary EMA assessments of NA were administered 5 times daily at random times, and when participants reported temptations to smoke. Participants also reported daily on stressful events and perceived stress. Analysis Plan: We analyzed day-to-day influences on lapse by relating daily stress and NA to the subsequent days' lapse risk, and by examining trends in stress and NA in the 4 days preceding the first lapse (logistic regression and GEE). Analyses of proximal within-day influences examined trends in NA in the hours preceding the first lapse, compared to a control day (GEE). We examined a subset of smokers who endorsed stress or a bad mood as the top trigger for the first lapse. Results: Day-to-day changes in stress and NA did not influence lapse risk. Within the lapse day itself, however, NA increased significantly over the 6 hours before lapses that were triggered by stress or a bad mood. Discussion: Slow-changing day-to-day shifts in stress and NA do not affect lapse risk, but more proximal and rapid increases in NA may precipitate lapses. Intensive EMA-based assessments are essential to understanding such rapid transient changes in affect.

Funding Source:
Grant R01 DA06084 from the National Institute on Drug Abuse to Saul Shiffman

Disclosures:
Saul Shiffman has undertaken research and consultancy for manufacturers of smoking cessation products, is currently consulting exclusively for GlaxoSmithKline, and holds a patent and interest in a new nicotine medication. He is also a founder of invivodata, inc., which provides electronic diaries for clinical trials.

Authors:
William Riley, PICS, Reston, VA Steven C. Carson, University of Phoenix, Phoenix, AZ Valerie Forman, University of Iowa, Iowa City, Iowa Jami Obermayer, PICS, Reston, VA

Abstract:
Background: This study sought to evaluate the difference between PDA-based vs paper-based monitoring on both compliance to recording and on nicotine gum use. Methods: Forty-nine smokers who desired to quit using nicotine gum were entered into the study and randomly assigned to either the pocket PC (PDA) or paper-based self-monitoring (LOG) condition. All subjects were to quit smoking using nicotine gum and then to record their gum and/or cigarette use for a three week period. The self-monitoring protocol consisted of three scheduled recording sessions each day in which the subject recorded for the prior 5 hr. period the number of cigarettes smoked, number of pieces of gum used, and ratings of cigarette cravings or nicotine gum side effects. Results: Significant differences were found between groups in recording compliance for all three weeks, particularly in the later two weeks of the trial (e.g. 69% vs. 39% in week 3). Participants in the PDA group reported chewing significantly more pieces of gum per assessment period than participants in the LOG group (3.5 vs. 2.4) but did not differ significantly on reports of concomitant gum and cigarette use during the same assessment period (14% PDA, 19% LOG). Discussion: These results indicate that handheld, computer-based recording improves self-monitoring recording compliance, especially over time and even in difficult recording contexts (i.e. relapsed smokers continuing to record gum and cigarette use). Computerized recording also appears to have an effect on nicotine gum use although a variety of factors could contribute to these observed differences.

Funding Source:
National Heart, Lung, and Blood Institute (NHLBI; R43HL65107).

Disclosures:
All authors are current or former employees of PICS, a private research and development firm with commercial interests in the self-monitoring of smoking and other health behaviors.

Authors:
Alan L. Shields McLean Hospital, Harvard Medical School Belmont, MA, USA Michael R. Hufford invivodata, inc. Pittsburgh, PA, USA

Abstract:
Background: Electronic diaries are increasingly used to collect health related outcome data in real-time from patients in the field. This comprehensive literature review examines how handheld computing and momentary data sampling strategies have impacted the collection of reliable and valid diary data in clinical areas. Methods: We performed a systematic review of original, empirical studies in peer reviewed journals published in the English language. Data sources include bibliographic searches of medical databases, manual bibliography searches, and institutional contacts. Analysis plan: Descriptive statistics are used to provide summary illustrations of the clinical phenomenon assessed, sampling strategies used, current reporting practices, and compliance and other measurements that assess data quality among studies employing electronic diaries for real-time data collection. Results: Approximately 100 identified studies used electronic diaries to assess clinical phenomenon ranging from urinary incontinence to smoking cessation and employ a variety of technologies and time-, event-, and signal-dependent sampling strategies to collect real-time data. Despite tremendous variability in reporting practices, trends emerged. The rates of patient compliance with electronic diary protocols differed widely across studies (50-99%) but studies that include subject training, compliance features, and/or livability features are associated with higher rates of success in the field compared to diaries without these features. Discussion: Improved standards are needed for how researchers describe their real-time data collection methods and subsequent results. While not all electronic diary studies are successes in the field, variation in technical success and patient compliance appears to be a function of the maturity of the electronic diary system.

Funding Source:
None

Disclosures:
Michael R. Hufford, Ph.D., is the Vice-President for Scientific Affairs at invivodata, inc. Invivodata provides electronic diary support for clinical trials.

Authors:
Alan L. Shields Mclean Hospital, Harvard Medical School Belmont, MA, USA Michael R. Hufford invivodata, inc. Pittsburgh, PA, USA

Abstract:
Background: One reason for the discrepancy between our understanding of alcoholism and the efficacy of our interventions is that our perspective of the struggle for abstinence relies almost exclusively on patients' retrospective reports, which can be inaccurate and biased. An alternative to retrospective data is Ecological Momentary Assessment (EMA): the collection of self-reports of momentary states on multiple occasions in participants' natural environment. We sought to better understand the proximal determinants of temptations to drink among alcoholics discharged from an inpatient treatment center by using EMA to collect data regarding participants' momentary experiences in near real-time using palmtop computers. Methods: Nineteen participants took part in real-time monitoring for one month subsequent to discharge. Participants completed research interviews, self-report measures, and were trained to use a palmtop computers. They responded to signal- and time-contingent prompts and reported their mood, activities, social setting, and coping several times each day. In addition, participants completed event-contingent reports when they were tempted to drink alcohol and when they consumed alcohol. Results: Our real-time monitoring protocol resulted in the observation of 2660 daily assessments of mood and activities, 107 temptation reports, 2 lapse reports, and a total of 532 person-days of monitoring. Compliance rates were assessed to be good to excellent. Analyses examined the following questions: Do self-report measures of high-risk situations correspond to real-time reports of temptations to drink? What is the magnitude of inaccuracy and bias when comparing retrospective versus real-time reports of temptations to drink? Discussion: The unique perspective gained from the use of real-time monitoring methodology may lead to significant advances in our understanding of the proximal determinants of temptations to drink among patients with alcohol dependence.

Funding Source:
This research was supported by grant #292145 from the MONTS project (National Science Foundation, EPSCoR) awarded to the second author.

Disclosures:
Michael R. Hufford, Ph.D., is the Vice-President for Scientific Affairs at invivodata, inc. Invivodata provides electronic diary support for clinical trials.

Authors:
Ned L. Cooney(1), Mark D. Litt(2), Judith L. Cooney(2), Cheryl A. Oncken(2), David T. Pilkey(1), and Howard R. Steinberg(1)(1) Yale University School of Medicine, New Haven, CT(2) University of Connecticut Health Center, Farmington, CT

Abstract:
Background: Tobacco use is the leading cause of death among alcoholics. Clinical trials have examined smoking cessation interventions for alcoholic smokers in alcohol treatment, but smoking quit rates have been unimpressive, underscoring the need for investigation of antecedents of cigarette smoking in alcoholic smokers. Methods: Participants (N= 102) in a randomized clinical trial of concurrent smoking cessation and alcohol treatment were instructed in monitoring moods, self-efficacy, urges to drink or smoke, and drinking and smoking behavior using an electronic diary (ED) that signaled them for assessment at quasi-random intervals, four times/day. ED data were collected for 14 days after completion of a three-week intensive alcohol treatment. Analysis Plan: Antecedents to reported cigarette smoking were examined using Generalized Estimating Equations procedures in which ED ratings of mood, craving, self-efficacy, and smoking were entered as predictors of smoking in subsequent ED assessments. Results: The strongest predictor of reported smoking in an ED assessment was reported smoking in the previous ED assessment. In addition, higher reported urge to smoke and lower ratings of positive, high arousal mood (e.g., active, peppy) predicted subsequent smoking. Thirteen participants reported the onset of drinking during the 14-day monitoring period. The first day of drinking was significantly predicted by lower confidence to resist drinking and higher urge to smoke on the day prior to the lapse. Discussion: Data suggest that urges to smoke are important antecedents for both smoking and drinking after concurrent alcohol and tobacco treatment. Future research should examine interventions targeting cigarette craving in concurrent alcohol-tobacco treatment programs.

Funding Source:
Supported by NIAAA grant # R01 AA11197

Disclosures:
None

Authors:
Kathleen A. O'Connell, Teachers College Columbia University, New York, NY, USA Joseph E. Schwartz, State University of New York--Stony Brook, Stony Brook, NY Saul S. Shiffman, University of Pittsburgh, Pittsburgh PA

Abstract:
Background: Ecological Momentary Assessment (EMA) methods were used to apply the theory of self-control strength (Muraven & Baumeister, 2000) to smoking cessation. Self-control is posited to be a limited and consumable resource; once it is depleted, time and rest are required to replenish it. Because high urge episodes and coping strategies are likely to utilize self-control resources, we hypothesized that temptations to smoke that follow high urge reports will be (1) more likely to end in a lapse, and (2) will be associated with fewer coping strategies than temptations that follow low urge reports. Methods: Smokers in two different samples (N= 61; N=240) carried palm-top computers during the first two to four weeks of their cessation attempt. One sample also used tape recorders to report coping. Participants made reports at randomly prompted occasions and immediately after successfully resisted temptations and lapses. Analyses: Multi-level, random effects regression analyses were performed. Results: The theory was partially supported. Temptation episodes occurring within 4 hours after high urge reports were more likely to end in lapses in both data sets. However, using many strategies in one episode did not lead to using fewer strategies in a subsequent episode. Discussion: This EMA study was unusual in that it used prospective analyses, and included data derived from tape recordings as well as handheld computers. The methodology enabled us to test a theory heretofore tested only in the laboratory in the context of real-time experiences during smoking cessation.

Funding Source:
This study was supported by NIH grant #NR03145-05, Kathleen A. O'Connell, Principal Investigator

Disclosures:
Saul Shiffman's Smoking Research Group provided the palm top computer technology and software used in these studies.

Authors:
Joan E. Broderick, Ph.D.1 , Joseph E. Schwartz, Ph.D. 1, Saul Shiffman, Ph.D.2,3 , Michael R. Hufford, Ph.D.3 ,Arthur A. Stone, Ph.D.1 1Department of Psychiatry & Behavioral Science Stony Brook University, 2University of Pittsburgh, 3invivodata, inc., Pittsburgh, PA

Abstract:
Background: Prior research has demonstrated that compliance with the reporting schedule in paper diary protocols is poor when respondents are not provided with reminder signals. We hypothesized that compliance with a paper diary protocol would be improved by using auditory signaling. Methods: Adults with chronic pain (N= 27) were recruited from the community to participate in a 24-day experience sampling protocol of 3 pain assessments per day (10 AM, 4 PM, 8 PM). Paper diaries were instrumented to record openings and closings, thereby permitting determination of date and time when the participant could have made diary entries. Participants were signaled with a programmed wrist watch at the onset of each 30-minute assessment window. Two compliance windows were defined: +15 minutes and +45 minutes of the targeted assessment time. Results: Self-reported compliance based on participants' paper diaries was 85% and 91% for the 30 and 90-minute windows. Verified compliance was 29% and 39% for the two windows. Signaling produced a significant increment in verified compliance when compared with an identical trial without signaling [Stone, A.A., Shiffman, S., Schwartz, J.E., Broderick, J.E., Hufford, M.R. Patient compliance with paper diaries. British Medical Journal, 2002, 324, 1193-1194]. A significant eroding of verified compliance was observed across the three weeks of the study. Conclusions: This study provides additional evidence that self-report dating of diary entries may be misleading investigators about compliance with diary protocols. While auditory signaling enhances compliance, the result is still unsatisfactory.

Funding Source:
This research was supported by a grant from the National Cancer Institute (CA-85819; Arthur A. Stone, Principal Investigator) and in-kind support from invivodata, inc., Pittsburgh, PA.

Disclosures:
invivodata, inc. supplies software for the use of electronic diaries in clinical trials. Saul Shiffman is a co-founder and the Chief Science Officer of invivodata, inc. Michael Hufford is Vice President for Scientific Affairs at invivodata, inc. Arthur A. Stone is Associate Chair of the Scientific Advisory Board of invivodata, inc.

Authors:
Augustine Samba, Ph.D. PrimeCo Enterprises Aurora, USA

Abstract:
B. Research in Progress. A Clinical Trial is a research study to answer specific questions about new therapies or new ways of using established treatments. This paper describes the end-to-end communications infrastructure and architectural elements that are needed to facilitate efficient real-time clinical trial Operations, Administration and Management. The architectural components consist of a Remote Gateway Server, which runs the "Service Package Application" that defines the type of Clinical Trial "Service" being provided. The Gateway Server communicates with Clinical Trial participant End-User Devices and Researchers in real time. It also communicates with external network elements, such as Centralized Databases. The Service Type specifies a programmed behavior for processing one or more Clinical Trial studies. Multiple Services can be supported on a single Gateway platform. The Service definitions may include characteristics of momentary thoughts, feelings, timing elements, study schedules, tracking, study report types and statistical analysis. These Services are deployed on the Remote Gateway Servers, via a Service Management System (SMS). The SMS is used to provision and administer participant End-User Devices and Service Package Applications for each Service type deployed on the Remote Gateway Server. SMS receives the definition of the service's provisionable portion from a Service Creation Environment, manual entry or from external systems. This data is validated and routed to the appropriate Gateway Servers across multiple interfaces, depending on the type of data being provisioned. SMS also provides a reporting environment to view results of clinical trials. SMS supports migration of data from an older to a newer.

Funding Source:
I am seeking funding to support the proposed Research and Development project

Disclosures:
I am Independent Contractor of QuikCAT Technologies.

Authors:
Lew Berman, Centers for Disease Control and Prevention

Abstract:
The National Health and Nutrition Examination Survey (NHANES), developed by the Centers for Disease Control and Prevention, is a large and comprehensive health survey utilizing leading edge technologies to produce national estimates of health measures and the nutritional status of the United States population. The fourth National Health and Nutrition Examination Survey (NHANES) took a new direction beginning in 1999. The major differences from previous surveys are that NHANES will be continuous and that it will be linked to related Federal government data collections of the general United States (U.S.) population. NHANES is designed to collect data that can be obtained by direct physical examination, clinical and laboratory tests, and related measurement procedures. This information is used to estimate either the prevalence of some disease or to estimate the normative distribution of the characteristic in the total population.

To support NHANES a real-time state-of-the-art integrated survey information system (ISIS) was developed for the collection, review, editing, and analysis of the questionnaire and biomedical data. The impact of ISIS has been significant in that it has resulted in more accurate and timely data release. In addition, the ISIS allows for field oversight and intervention to resolve problems much more quickly then done in previous NHANES surveys.

Funding Source:
CDC

Disclosures:
None

Authors:
Thomas A. Wills, Albert Einstein College of Medicine; Larry D. Jamner and Carol K. Whalen, University of California, Irvine

Abstract:
Background: Adolescent cigarette smoking is of great health relevance and epidemiologic studies have shown measures of self-control related to early onset of tobacco and alcohol use (Wills, Cleary et al., 2001), but exactly how these factors operate on a daily basis is not well understood. Recent research has used experience-sampling methodology to study daily behaviors of teenagers (Whalen, Jamner et al., 2001). Methods: In a cohort of 400 adolescents (12 17 years of age), participants logged their experiences and behaviors every 25 (± 10) minutes across 4 day intervals including 2 week and 2 weekend days, and the diary included an item on being "in control of things." Analysis: SAS GEE procedures determined the Odds Ratio of a diary report of smoking as a function of diary ratings of "In Control" 30 minutes earlier. Results: There was an overall odds ratio of 0.88 (CI 0.80 - 0.97; p < .01); likelihood of smoking was decreased when persons felt more in control. A median-split blocking on the Barratt Impulsivity Scale indicated diary ratings of control were predictive of smoking among high-trait-impulsive youth (OR = 0.91 (CI 0.83 1.00; p = .03), but not among low-trait-impulsive youth (OR = 1.01 (CI 0.85 1.21; ns). Data on compliance indicated that overall, participants responded to 81% of all diary prompts; good adherence rates were observed among males and females, and among smokers and nonsmokers (range 72% 89%). Discussion: We discuss methodological refinements for studies of smoking behavior and directions for using self-control constructs in smoking prevention research.

Funding Source:
This work was supported by a Research Scientist Development Award K02 DA00252 from the National Institute on Drug Abuse (TAW) and by a TTURC grant P50 DA133332 and an R01 grant CA08031 from NIDA and the National Cancer Institute (LDJ and CKW)

Disclosures:
None

Authors:
Kazuhiro Yoshiuchi, Department of Neurosciences, University of Medicine and Dentistry of New Jersey - New Jersey Medical School, East Orange, USA, and Department of Psychosomatic Medicine, Faculty of Medicine, The University of Tokyo, Tokyo, Japan Dane B. Cook, Department of Neurosciences, University of Medicine and Dentistry of New Jersey - New Jersey Medical School, East Orange, USA Kyoko Ohashi, Graduate School of Education, the University of Tokyo, Tokyo, Japan Yoshiharu Yamamoto, Graduate School of Education, the University of Tokyo, Tokyo, Japan Benjamin H. Natelson, Department of Neurosciences, University of Medicine and Dentistry of New Jersey - New Jersey Medical School, East Orange, USA

Abstract:
Background: Subjective symptoms are crucial in making the diagnosis and following the course of chronic fatigue syndrome (CFS). Recently, computerized electronic diaries (ED) using palmtop devices were developed for ecological momentary assessment (EMA) to solve the problems of paper-and-pencil diaries such as recall bias and fake compliance. However, these devices are not wearable, which leads to other problems such as forgetting to take it with her or putting it away in a place where its alarm cannot be heard. To solve these problems, we have developed the ECOLOG, a wearable watch-type computer device. Methods: We developed a watch type computer for ED, a continuous performance task (CPT) consisting of a one back memory task, and for measuring activity with an accelerometer continuously. The diary asks questions concerning physical symptoms and mood states approximately every four hours. Following diary entry, the CPT was administered. We collected data from CFS patients and healthy controls using the device one week before and two weeks after a maximal exercise test because CFS patients have the major complaint that symptoms get worse following exertion and little objective evidence to help understand this symptom exists. Analysis Plan: We will test the main effects of groups (controls vs CFS) and exercise (before and after), and the interaction between groups and exercise with mixed linear methods. Discussion: We have developed a new watch-type computer device for EMA and it appears useful in evaluating symptoms, cognitive function, and activity in CFS patients in the natural settings.

Funding Source:
NIH AI-32247

Disclosures:
Seiko Instruments Inc. (Kazuhiro Yoshiuchi)

Authors:
Karen K. Liu, Philip Davis, Rosalind W. Picard MIT Media Laboratory, Cambridge, USA

Abstract:
Background: Traditional ecological momentary assessment tools with random or fixed timing prompts are annoying and often miss key events. The disruptiveness of these devices hinders long-term acceptance of ubiquitous health systems as users eventually refuse to interact with these devices. We are building relational, wearable devices that sense physiological data and allow for more enjoyable real-time event and affect annotation. Methods: We use statistical machine learning, in tandem with user feedback, to learn minimally intrusive times to query the user. While this approach could potentially tailor interactions to the user based on previous user input and the current situation, it requires gathering and labeling huge amounts of training data, which can be an arduous and tedious process. We are developing algorithms that detect patterns of change in physiological signals to determine minimally intrusive interruption points from user feedback for continuous real-time learning of appropriate annotations. Using relational strategies such as empathy and social dialogue in conjunction with these algorithms, we aim to develop an enjoyable method for real-time event and affect annotation of physiological data. Analysis Plan: We plan on using measures such as willingness to use the device, amount of quality data collected, and the three-part working alliance measure (bond, task, goal) to evaluate the system. Discussion: Key challenges include crafting minimally annoying interactions, implementing machine learning for generalizing a small number of annotations to cover a large portion of the dataset, and also decision-making about when to interrupt the user and proactively solicit annotations.

Funding Source:
MIT Media Laboratory Things That Think Consortium

Disclosures:
None

Authors:
Mary Ann Parris Stephens & Erin M. Fekete, Kent State University, Kent, Ohio; Melissa M. Franks, Wayne State University, Detroit, Michigan; Anthony T. Harsch, ATH Consulting, Ann Arbor, Michigan

Abstract:
Traditional electronic ESM and EMA methods pose several unique problems for collecting data on older adults' behavioral and affective experiences. Many older adults suffer from impaired vision and manual dexterity. These physical limitations create difficulties for the use of popular ESM/EMA electronic methods relying on hand held computers with small display screens and keyboards. Additionally, most ESM/EMA methods have been designed to assess the experiences of one person at a time, and do not consider the simultaneous experiences of other family members (e.g., the spouse). We developed an easy to use electronic diary method for use with older medical patients and their spouse caregivers. Our method was also designed to address several potential pitfalls associated with collecting ESM/EMA data, including missed time assessments and inappropriate recording of data. Our data collection program captures daily and momentary experiences via lap-top computers, and allows the data to be viewed via Microsoft Excel or SPSS. Our new method allows older adults to view large font screens with as few as one questionnaire item per screen. Various modes can be used to enter data, including arrow keys, number keys, and mouse. The real-time and date stamp feature of the program records the date and clock times when participants start and finish each diary, and does not allow participants access to a given diary record before or after its scheduled assessment. The method also features separate diary programs for patients and spouses, with protected passwords to allow for independent and private responding in dyads.

Funding Source:
SUMMA Health System

Disclosures:
None

Authors:
Lawrence H. Cohen, University of Delaware, Newark, Delaware Andrew C. Butler, Beck Institute for Cognitive Therapy and Research, Bala Cynwyd, Pennsylvania Kathleen C. Gunthert, Department of Psychology, American University, Washington DC

Abstract:
Background: We are using a daily process design to assess depressed cognitive therapy (CT) outpatients' ability to regulate negative affect (NA) in the context of daily stress. We hypothesize that: (1) patients who begin treatment with relative strengths in NA regulation will show the most improvement in CT, independent of their initial level of depression; and (2) patients' NA regulation will improve during CT. Methods: Before their first treatment session (time 1), 60 depressed adults, who are receiving CT at the Beck Institute for Cognitive Therapy and Research (PA), complete nightly 8-minute automated and time-stamped telephone interviews for one week, administered via an interactive voice response (IVR) system (Healthcare Technology Systems). Items assess: (1) immediate mood and self-esteem; (2) negative and positive daily events; (3) the day's worst event; and (4) appraisals, automatic thoughts, and coping activities associated with the day's worst event. These nightly week-long interviews are readministered six weeks later (time 2). Patients also complete a depression inventory at every (weekly) treatment session. Analysis Plan: We will use hierarchical linear modeling to assess each patient's NA regulation (e.g., unique relationship between number of daily negative events and daily NA). We will use longitudinal growth modeling to establish each patient's change trajectory in treatment (based on weekly depression scores), and we will use the NA reactivity index to predict individual differences in those trajectories. To evaluate change during CT, we will conduct repeated measures (time 1, time 2) ANOVAs on the NA reactivity index. Discussion: Our project represents the first application of an IVR-based daily process methodology to research on psychotherapy. Our methodology has broad applicability to fundamental questions in psychotherapy and health intervention in general involving NA regulation.

Funding Source:
NIMH. 1 R21 MH067825-01. (Daily Process Design Applied to Cognitive Therapy) PI = Lawrence H. Cohen

Disclosures:
None

Authors:
Mark Muraven; University at Albany; Albany, NYR. Lorraine Collins; Research Institute on Addictions; Buffalo, NY Elizabeth T. Morsheimer; Research Institute on Addictions; Buffalo, NY Saul Shiffman; University of Pittsburgh; Pittsburgh, PA Jean A. Paty; University of Pittsburgh; Pittsburgh, PA

Abstract:
Background: Ecological Momentary Assessment (EMA) provides a methodology for examining the role of affect in drinking. Theories suggest that negative affective reactions to excessive drinking may promote dysregulation and further drinking. Drinkers may experience guilt if they violate a self-imposed limit on alcohol intake, especially if they attribute the violation to a personal weakness or failure. In turn, attempting to cope with guilt produced by excessive consumption may lead to more drinking in the future. Using EMA procedures, we examined the reciprocal relationship between negative affect and alcohol consumption in the daily lives of social drinkers. Methods: Two separate samples of social drinkers (total N = 144) monitored their alcohol intake and their cognitive and emotional response to that drinking on handheld computers for several weeks. Each day, participants completed four types of EMA interviews (random, before and after drinking, morning). Analysis Plan: Multilevel regressions were used to examine the within-person relationships among alcohol intake, reports of limit violations, and feeling bad about one's alcohol consumption. Results: Mornings after drinking, those who consumed more alcohol the previous day reported a stronger experience of a limit violation. Violating limits was associated with negative affect and guilt, even after controlling for hangover symptoms. Guilt over drinking predicted consumption later that day. Trait variables moderated these relationships. Discussion: Among social drinkers, violating a limit on alcohol intake produced guilt/negative affect. They responded to that guilt with increased drinking. EMA provided the best method for studying these processes as they unfolded over time.

Funding Source:
Grant AA07595 provided by the National Institute on Alcohol Abuse and Alcoholism of the National Institutes of Health awarded to R. Lorraine Collins. Grant AA12770 provided by the National Institute on Alcohol Abuse and Alcoholism of the National Institutes of Health awarded to Mark Muraven.

Disclosures:
After completion of the study, Saul Shiffman and Jean Paty founded invivodata, Inc., which provides support for electronic diary methods in clinical trials.

Authors:
Tracy Simpson, Daniel Kivlahan, Kristen Bush, Miles McFall all affiliated with VA Puget Sound Health Care System, Seattle, WA USA; Drs. Simpson, Kivlahan, and McFall also affiliated with University of Washington.

Abstract:
B. Research in progress. Daily monitoring studies have examined factors related to the maintenance of various problematic behaviors. Post Traumatic Stress Disorder and substance use problems frequently co-occur and many clients describe a functional, self-medication relationship between the two disorders. However, no reports of daily monitoring investigate this apparent relationship. This ongoing study examined the feasibility of using Interactive Voice Response to monitor PTSD, alcohol craving and use among patients in early recovery from an alcohol use disorder. Most participants were also diagnosed with PTSD (77%) and many were homeless (40%). To investigate measurement reactivity (i.e., that levels of symptom ratings would be influenced by the monitoring), we randomized 82 participants to: daily 2-3 minute monitoring for 28 days, weekly 6-8 minute monitoring for four weeks, or no-monitoring. The overall monitoring compliance rates for those in the monitoring conditions were comparable (daily 75.3%; weekly 71.9%). Excluding two participants who made no calls, those in the daily monitoring condition provided 81.8% of the possible data points over 28 days. One-month follow-up rates by group were: daily 82%; weekly 62%; no monitoring 90%. Comparisons between groups showed no evidence of significant measurement reactivity on key indicators at follow-up (PTSD severity, alcohol craving, days drinking, depression). These preliminary results indicate the feasibility of daily monitoring of alcohol craving and use and PTSD symptomatology even in an unstable treatment-seeking sample. In addition, the lack of detectable measurement reactivity suggests that this methodology may be used to examine temporal covariations between these symptoms without unduly influencing symptom ratings.

Funding Source:
1) VA Mental Illness Research, Education, and Clinical Center 2) University of Washington, Alcohol and Drug Abuse Institute

Disclosures:
None

Authors:
Marc I. Rosen, M.D. Joyce Cramer

Abstract:
Background: Our group has used electronic caps that record date and time of bottle opening in cue-dose training, an adherence-focused therapy that involves (a) review of electronic data to precisely identify missed doses and (b) encouraging patients to develop cues in their environment for taking prescribed medication. We will present data from 3 completed, published randomized controlled trials that examined whether cue-dose training, improves adherence. Methods: Three randomized controlled trials in which patients were randomly assigned to cue-dose training or control conditions were conducted - among patients prescribed psychotropic medication (Cramer and Rosenheck, Journal of Nervous and Mental Disorders, 1999), patients prescribed antiretroviral medication (Rigsby et al., Journal of General Internal Medicine, 2000), and patients prescribed the antihyperglycemic Metformin (Rosen et al., Behaviour Research and Therapy, In Press). Results: In the studies of patients prescribed psychotropic medication and patients prescribed metformin, cue-dose training was associated with greater adherence over time than the control condition. In patients prescribed antiretroviral medication, cue-dose training was associated with improved adherence only when it was coupled with monetary reinforcement. Discussion: These studies suggest that counseling based on review of electronically-generated adherence data can improve adherence to prescribed medication. We will further describe a theoretical framework for integrating electronically-generated adherence feedback into adherence counseling (Rosen et al., Behaviour Change, In Press) and into existing clinical care.

Funding Source:
R01-DA15215 (MIR)R01-MH61169 (MIR)

Disclosures:
Joyce Cramer is a consultant to AARDEX, Ltd.

Authors:
Tamlin C. Christensen and Lisa Feldman Barrett Boston College Chestnut Hill, MA, USA

Abstract:
Background. This poster addresses theoretical and practical issues in the measurement of momentary self-report data (ecological momentary assessment; EMA; also called experience-sampling). Early perspectives on self-reports conceived of EMA as the better measure of subjective experience compared to other forms of self-report. More recent perspectives have refined this view by suggesting that EMA captures a different type of knowledge for subjective experience (i.e., time-situated episodic knowledge rather than time-independent semantic knowledge). We extend this line of thinking by suggesting that EMA evokes a phenomenologically distinct state of conscious awareness in the respondent (i.e., time-situated autonoetic awareness; Tulving, 1985). Method/Analysis: Using a combined laboratory and computerized EMA study (PalmPilots running the Experience-Sampling Program: www2.bc.edu/~barretli/esp), we tested the implications and boundary conditions of a consciousness perspective. Results: We report two main findings. First, momentary self-report data can be influenced by factors (e.g., psychological defense) normally attributed to more traditional self-report procedures. This finding strengthens our understanding that real-time self-reports only yield information that a person is willing and able to represent in conscious awareness the moment a report is made. Second, we present evidence that EMA may not measure the same kind of conscious awareness for all individuals. For some, EMA may measure their time-situated autonoetic awareness, but for others, these procedures may measure a more time-independent belief-focused awareness. Discussion: Results have important implications for the measurement of health related experiences and, more generally, the science of self-report.

Funding Source:
Fetzer Institute grant to Lisa Feldman Barrett

Disclosures:
None