Cost-effectiveness analysis of recruitment strategies in a community-based intervention study of HIV-infected persons

This paper describes and evaluates recruitment methods used to reach adults with HIV infection and to enroll eligible candidates in a randomized trial of aerobic exercise. Potential participants residing in the metropolitan Washington DC area were recruited from January 1994 to December 1996. The yield and associated cost of clinic centre site-visit (CSV) recruitment are compared to similar outcomes for community-based (CB) recruitment strategies, which consisted of presentations to local groups, mail/phone canvasses of caregivers, neighbourhood network promotion, public site postings and print media notices. Of 833 HIV infected adults ascertained during the recruitment phase as prospective study candidates, 66.7% were initially contacted during CSV recruitment. The remainder, 33.3% were CB recruits. The percentage of screened candidates who were subsequently enrolled in the study was 13.5% for CSV recruitment and 21% for CB recruitment. Ascertainment and screening costs combined were $158 per CB recruit compared to $232 per CSV recruit. Using multiple recruitment approaches we successfully achieved our enrollment goal of at least 100 volunteers from diverse populations by the end of the planned recruitment period.