• Lisa Avery and Michael Rotondi.
• Department of Mathematics and Statistics, University of Otago, Dunedin, New Zealand; Norman Bethune College, Keele Campus, York University, Keele Street, Toronto, Ontario M3J 1P3, Canada. Electronic address: lavery@maths.otago.ac.nz.
• J Clin Epidemiol. 2020 Jan 1; 117: 68-77.

BackgroundRespondent-driven sampling (RDS) is an increasingly popular method of recruiting participants from hard-to-reach populations and has traditionally been used to estimate the prevalence of HIV among marginalized people. The STROBE-RDS guidelines were published in 2015 to improve the reporting of these studies. We aim to determine the current applications of RDS and the quality of reporting of these studies.MethodsThe SCOPUS, PubMed, and CINAHL databases were used to find papers published in 2017, relating to RDS. Papers meeting the inclusion criteria of cross-sectional studies using RDS were classified according to the study outcome and target population. A random sample of 25 papers was selected to evaluate the quality of reporting using the STROBE-RDS guidelines.ResultsMen who have sex with men, people who inject drugs, and female sex workers were the most common populations for RDS studies; over half of the studies examined the HIV epidemic. Quality of reporting is good with respect to the original STROBE guidelines but is generally weaker with respect to RDS-specific aspects of the study, including recruitment and statistical analysis.ConclusionMost authors are using RDS appropriately and aware of the need for statistical adjustments to RDS data. Nonetheless, the STROBE-RDS guidelines should be more widely disseminated to promote better reporting of key aspects of RDS studies.Copyright © 2019 Elsevier Inc. All rights reserved.

22 keywords...

hide…