Failing to retain an adequate number of study participants in behavioral intervention trials poses a threat to interpretation of study results and its external validity. This qualitative investigation describes the retention strategies promoted by the recruitment and retention committee of the Behavior Change Consortium, a group of 15 university-based sites funded by the National Institutes of Health to implement studies targeted toward disease prevention through behavior change. During biannual meetings, focus groups were conducted with all sites to determine barriers encountered in retaining study participants and strategies employed to address these barriers. All of the retention strategies reported were combined into 8 thematic retention categories. Those categories perceived to be most effective for retaining study participants were summarized and consistencies noted among site populations across the life course (e.g., older adults, adults, children, and adolescents). Further, possible discrepancies between site populations of varying health statuses are discussed, and an ecological framework is proposed for use in future investigations on retention.

Innovative and effective health promotion interventions targeted on older adults within a public health framework will be increasingly important as the US population ages dramatically. The benefits of healthier lifestyles for older adults include increased functional ability and improved quality of life. The Study of Exercise and Nutrition in Older Rhode Islanders (SENIOR) Project is a multibehavioral health promotion intervention for community-dwelling older adults focusing on increasing exercise and fruit and vegetable consumption. Intervention materials are stage-tailored for each individual, and include manuals, newsletters, expert system assessments and reports, and telephone coaching. The primary purpose of the SENIOR Project is to investigate the relative effectiveness of a multiple-behavior intervention-based on a single theoretical framework-compared to single-behavior interventions. The secondary purposes are to investigate the intervention's effects on both functional ability and general health outcomes, and how older adults move along a continuum of changing health behaviors. The Transtheoretical Model of Health Behavior Change, the conceptual framework for the SENIOR Project, was chosen for the following reasons: performance potential with older adults, individual tailoring on a stage basis, technological features, and interdisciplinary research base and community partnership.

Presented are the basic design, methods, and baseline data analyses for the Study of Exercise and Nutrition in Older Rhode Islanders (SENIOR Project) an experimental study to investigate the relative effectiveness of a 12-month, stage of readiness to change-based multiple-behavior intervention (exercise and nutrition) compared to single-behavior interventions in a community-dwelling population of 1,277 older adults. Relationships between stage of readiness to change in the two target behaviors, as well as the relationship between stage of readiness and self-reported exercise levels and fruit and vegetable consumption, were examined using a combination of Pearson chi-squares, analyses of variance (ANOVA), and Spearman's rank order correlations. Stage of change (SOC) for fruit and vegetable consumption was significantly associated with the dietary measure, and SOC for exercise was associated with both the three physical activity measures and servings of fruits and vegetables per day. Overall, individual older adult's readiness to change seems largely to be behavior-specific.

This article discusses the process of developing collaborative relationships for a community-based health promotion project. A partnership was established among the university, the city where the intervention took place, and the community senior center. A community advisory board was created to identify the strengths, diversity, and needs of each partner. The community advisory board guided the partnership to recruit 1,277 older adults to participate in the intervention study. A sample was deemed representative after comparison with Census 2000 data, with gender and educational attainment being similar.