Educational Services, Inc. (ESI)
4350 East West Highway, Suite 1100
April 28-29, 2009
Richard A. Jenkins, Ph.D.
Meeting Purpose and Intent:
This NIDA-sponsored consultation was inspired by investigators' observations that prevention and related epidemiology studies have had increasing difficulty recruiting drug-using men who have sex with men (MSM), particularly young men. A variety of hypotheses for this have been offered. These include generational shifts (e.g., a generation living with treatable HIV rather than seeing the deaths and declines of friends and loved ones), diminished sense of community (e.g., greater dispersion of gay male populations), "prevention fatigue", and the diminished importance of HIV as a uniting issue with the advent of highly active anti-retroviral treatment (HAART). Demographic changes in the epidemic, with increasing proportions of infections among racial/ethnic minority men, particularly African-Americans, mean consideration of cultural issues in recruitment and engagement of venues which only recently have been fully appreciated by the research community. There have been no recent summaries of research on recruitment of MSM populations and the purpose of this meeting was to summarize the current state of research in this area, with particular attention to drug using populations, and to make recommendations for future research.
Brief Summary & Discussion of Meeting Outcome:
The meeting reviewed investigators' experiences with common methods of recruiting drug-using MSM into studies, with particular attention to relatively emergent methods such as respondent-driven sampling (RDS) and internet based recruitment, and more established methods such as venue-day-time (VDT) sampling. Presentations reflected recruitment in current and past research studies. It was evident that although helping the community often motivated participation, altruism alone was insufficient. Experiences using RDS and related methods, such as participant referral to obtain a desired target sample yielded mixed results for meeting participants, with some efficiently accruing target samples, while others found these methods to be less effective. There was wide variation in accrual rates, but fairly consistent skew toward lower income populations over waves of RDS sampling. Intensive formative work, well-selected seeds, and training of participants in recruitment techniques seemed to yield more favorable outcomes. Internet methods were efficient although many "hits" were needed to accrue actual participants and conversion of internet participation to face-to-face follow-up is only beginning to be studied. VDT sampling appeared to pose fewer problems although venue selection (e.g., appropriateness of populations for target groups) and their spatial distribution (as a consideration for staffing) were identified as recurring problems for adequate sampling. Intervention studies, in particular, do not provide good opportunities for head-to-head comparisons of methods, and refinement of case study methods is, therefore, important. Newer methods such RDS and the internet provide certain advantages for studying the recruitment process in a case study format because data can be collected regarding different points between initial contact and actual participation. Additional considerations include institutional review board policy and practice, internal versus external validity considerations and the appropriateness of particular methods to populations in terms of culture, trust, social organization, and spatial distribution.
Recommendations centered on areas needing systematic evaluation, such as the utility of RDS for different settings and populations as well as practice considerations such as working with IRBs, further exploration of methods through case studies, utilizing multiple methods of recruitment for study retention, and methods for improving formative research. The participants plan to work toward to draft a manuscript summarizing findings from the presentations, related work by the participants, and relevant research from outside the participant group along with recommendations from the meeting.
Methodology-focused program announcements were provided to the participants, including the following:
OBSSR Methodology PAs (R01, R21, R03)
OBSSR Community Participation (Gen'l-R01; Medically Underserved R01, R21)
Participants (PDF Format, 56kb)
Agenda (PDF Format, 148kb)