398P
FACTORS RELATED TO PARTICIPATION IN COMMUNITY-BASED CLINICAL HIV RESEARCH
CA Worthington1, MJ Gill2
¹University of Calgary; 2Microbiology and Infectious Diseases, Faculty of Medicine, University of Calgary and Southern Alberta Clinic, Calgary, Alberta
Objectives: To describe participation in clinical HIV research, and examine predictors of research participation among clients of a regional sole HIV care provider.
Methods: Data were extracted from the databases of the Southern Alberta Clinic (SAC), in Calgary. Descriptive statistics were used to summarize patient characteristics, and univariate analyses and multiple logistic regression were used to examine relationships between research participation and socio-demographic, clinical, and clinic contact factors.
Results: 98% of eligible patients participated in a study that stored an extra tube of blood for resistance viral evolutionary studies. 29% of patients participated in at least 1 other study; 13% participated in 2 or more studies. Bivariate results suggest research participation (excluding viral evolutionary studies) was predicted by being male (p<.001), older (p<.001), white (p<.01), homosexual or bisexual (p<.001), having university education (p<.001), and with no history of injection drug use (p<.001). Patients with a lower nadir CD4 count, an AIDS diagnosis, those on antiretrovirals (ARV), a greater number of years in treatment, and a higher percentage of appointment kept (all p<.01) were more likely to participate in research studies. Multiple logistic regression indicated that clinical (nadir CD4 count OR=0.13, p<.05; ARV OR=3.0, p<05) and clinic contact characteristics (years in treatment OR=1.9, p<.05; percentage of appointments kept OR=1.02, p<.05) were the strongest predictors of research participation. Of the socio-demographic characteristics, only injection drug use history (OR=0.97, p<.05) remained a significant predictor in the multivariate model.
Conclusion: Participation in research is high when demands on patients are minimal (eg, a single tube of blood). For more burdensome studies, it appears that socio-demographic characteristics are markers for underlying clinical and clinic interaction factors that influence research participation. Thus, the clinical relationship and the manner that studies are structured are important factors to consider in planning research that adequately represents treatment populations.