No abstract

This report examines the impact of individualized, population-based recruitment and retention approaches on the development of a subject pool, enrollment, and retention at 12 months of healthy, community-based women in three ethnic groups: African Americans, non-Hispanic European Americans, and Mexicans/Central Americans. Of 722 women contacted and screened, 346 (48%) were eligible and consented to participate. Attrition at 12 months was low (10%) compared with other published reports. The largest group of potential subjects was identified through broadcast media approaches, but this method produced the highest number of ineligible women and highest rate of attrition. Printed matter produced the next largest group of potential subjects, but ineligibility was high (53%). Face-to-face interactions enrolled the highest proportion of eligible women (84%) and lowest overall attrition (7%). Direct referral yielded fairly efficient enrollments (57%) and average attrition. Multiple approaches for recruitment can produce a diverse sample of healthy, community-based women. Face-to-face recruitment results in the highest yield of participants with the lowest attribution but is presumed to require more resources.

Four broad groups of factors have been linked with self-management behavior in type 2 diabetes over time: (1) characteristics of patients, (2) amount and management of stress, (3) characteristics of providers and provider-patient relationships, and (4) characteristics of the social network/context in which disease management takes place. Of these four, social network/context has received the least amount of study and has been described in terms not easily applicable to intervention. In this paper, we identified the social network/context of diabetes management as residing within the family. We defined the family for clinical purposes, reviewed the literature concerning what is known about the link between properties of the family context of care and outcomes in type 2 diabetes and other chronic diseases, and identified areas of family life that are relevant to diabetes management. This information was then used to demonstrate how a family context of care can serve as a clinical framework for integrating all four groups of factors that affect disease management. Implications of this approach for practice and research are described.

R E S U LT S -Both sex and the three domains of family life were related to disease management, but the results varied by ethnic group. For EA patients, sex, family world view, and family emotion management were related to disease management (scores for Family Cohere n c e w e re negatively associated with HbA 1 c level and depression, and poor scores for Conflict Resolution were linked with high depression); for Hispanic patients, sex and family s t ru c t u re / o rganization were related to disease management (high scores for Organized Cohesiveness were associated with good diet and exercise, and high scores for Family Sex-Role Tr aditionalism were related to high quality of life). No significant interactions with sex occurre d . C O N C L U S I O N S -Characteristics of the family setting in which disease management takes place are significantly linked to patient self-care behavior, and these linkages vary by patient e t h n i c i t y. A family' s multiple independent dimensions provide multiple targets for interv e n t i o n , and diff e rences in family norms, stru c t u res, and emotion management should be considere d to ensure that interventions are compatible with the setting of disease management.