Participation in HIV research: the importance of clinic contact factors.

Recruiting minority populations living with HIV to many types of clinic-based HIV research is a concern. This study examined an expanded range of predictors of HIV research participation (clinic contact, clinical, and personal characteristics) to investigate observed ethnocultural differences in HIV research participation. Research participation was defined as participation in any of diagnostic, pathogenesis, drug trial or survey research. Logistic regression modeling was used to predict research participation of 657 eligible patients (93% of the patient population) who began care between January 1997 and the end of September 2003 at a regional outpatient HIV care program in Calgary, Canada. Approximately one third (32%) were non-white, including 18% Aboriginal, 9% black, 4% Asian, and 1% Hispanic individuals. Twenty-nine percent (187/657) of the patients participated in at least one study of any kind. Multivariate analysis indicated that the strongest predictors of any research participation (including diagnostic, pathogenesis, drug trial, or survey studies) are clinical (including nadir CD4 count [odds ratio {OR} = 0.132, p < 0.05] and antiretroviral medication [OR = 3.041, p < 0.01]), and clinic contact characteristics (years in treatment [OR = 1.949, p < .05], and percentage of appointments kept [OR = 1.022, p < .05]). Few personal characteristics were related to research participation-only reported injection drug use (OR = 0.363, p < 0.05), and birth year (OR = 0.968, p = 0.054) were significantly or marginally related to research participation. There was no significant difference among ethnocultural groups (white, non-white). Results thus suggest that much of the variation in research participation associated with ethnocultural group or race is accounted for by clinic contact variables. This in turn suggests that it is the different patterns of clinical interactions and service use shown by these groups that may influence research participation. To attract under researched populations, attention should shift from the "who" of research participation to the "how" of clinical interactions.