Respondent-Driven Sampling and Time-Location Sampling: A Comparison of Implementation and Operational Challenges for HIV Behavioral Research

Marissa Hall, University of North Carolina at Chapel Hill
Clare Barrington, University of North Carolina at Chapel Hill
Sanny Y. Chen, Centers for Disease Control and Prevention (CDC)
Nelson Arambu, Universidad del Valle de Guatemala
Sonia Morales, Universidad del Valle de Guatemala
William M. Miller, University of North Carolina at Chapel Hill
Berta Alvarez, Universidad del Valle de Guatemala
Gabriela Paz-Bailey, Universidad del Valle de Guatemala

Respondent-driven sampling (RDS) and time-location sampling (TLS) are used to recruit men who have sex with men (MSM) for HIV behavioral research. Two cross-sectional surveys, one using RDS and the other TLS, were conducted simultaneously among MSM in Guatemala City in 2010. The purpose of this study is to analyze the strengths and challenges associated with implementing each method based on data obtained from key informant interviews (n=10) and one focus group with field staff. Both RDS and TLS successfully and efficiently recruited the target sample size. RDS offered greater privacy and safety, required fewer human and financial resources, and presented fewer logistical challenges. TLS led to a greater understanding of the context in which MSM socialize and meet sex partners, providing important information for prevention efforts and data interpretation. We conclude with concrete recommendations for improving RDS and TLS implementation.