Leland J. Yee, PhD, MPH, Dept. of Epidemiology, University of Pittsburgh, Graduate School of Public Health, 130 DeSoto Street, Pittsburgh, PA 15261, 626-203-3327, yeel@edc.pitt.edu and Scott D. Rhodes, PhD, MPH, CHES, Department of Public Health Sciences, Wake Forest University Health Sciences, Section on Social Sciences and Health Policy, Medical Center Boulevard, Winston-Salem, NC 27157-1063.

BACKGROUND: Stigmatized populations whose size and boundaries are unknown, such as many of those adversely affected by HIV (ie., drug users, gay men, and commercial sex workers), are called “hidden populations” (HPs). Effective sampling of HPs is challenging. To better inform valid sampling of HPs, we reviewed the literature to synthesize the current strategies used to reach HPs. METHODS: We systematically reviewed the literature and synthesized findings with respect to sampling methods for HPs and drew on our own experiences with sampling frame development, implementation and data interpretation. RESULTS: Several distinct approaches for targeting HPs were identified, including: 1) electronic mail/electronic-bulletin boards, 2) Internet-based web surveys, 3) traditional outreach sampling, including venue-based, time-space sampling, and 4) respondent-driven sampling (RDS). CONCLUSIONS: Methods have competing advantages and disadvantages. E-mail surveys may provide a known framework of who has been sampled, but have low response rates. Internet surveys only reach individuals with internet access and may not provide knowledge of who was surveyed. Outreach sampling captures only a visible portion of the target population who attend specific venues. Based on Markov chain theory, RDS relies on chain-referral sampling and may be more effective in providing a random sample and reach less visible members of the target populations, but more research on this method is needed. HIV infection disproportionately affects individuals in “hidden populations” and effective research and prevention in these groups—the most vulnerable members of our community—will rely on the utilization of sampling approaches derived from these empirical studies.

Learning Objectives: At the end of this presentation, individuals will be able to

• Define the term "hidden populations" and identify characteristics of hidden populations;
• Analyze various sampling methods in stigmatized and hard-to-define hidden populations;
• Explore strategies to better improve research and prevention in these communities; and,
• Examine the public health importance of combining knowledge learned from empirical studies aimed at developing better sampling strategies with ongoing research and prevention efforts in the HIV community.

Keywords: Research, HIV/AIDS

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.