Lachowsky Nathan John, Sorge Justin Tyler, Raymond Henry Fisher, Cui Zishan, Sereda Paul, Rich Ashleigh, Roth Eric A, Hogg Robert S, Moore David M
Epidemiology & Population Health, British Columbia Centre for Excellence in HIV/AIDS, 608-1081 Burrard Street, Vancouver, V6T 1Y6, Canada.
School of Public Health and Social Policy, Faculty of Human and Social Development, University of Victoria, Victoria, Canada.
BMC Med Res Methodol. 2016 Nov 16;16(1):157. doi: 10.1186/s12874-016-0258-4.
Respondent-driven sampling (RDS) is an increasingly used peer chain-recruitment method to sample "hard-to-reach" populations for whom there are no reliable sampling frames. Implementation success of RDS varies; one potential negative factor being the number of seeds used.
We conducted a sensitivity analysis on estimates produced using data from an RDS study of gay, bisexual and other men who have sex with men (GBMSM) aged ≥16 years living in Vancouver, Canada. Participants completed a questionnaire on demographics, sexual behavior and substance use. For analysis, we used increasing seed exclusion criteria, starting with all participants and subsequently removing unproductive seeds, chains of ≤1 recruitment waves, and chains of ≤2 recruitment waves. We calculated estimates for three different outcomes (HIV serostatus, condomless anal intercourse with HIV discordant/unknown status partner, and injecting drugs) using three different RDS weighting procedures: RDS-I, RDS-II, and RDS-SS. We also assessed seed dependence with bottleneck analyses and convergence plots. Statistical differences between RDS estimators were assessed through simulation analysis.
Overall, 719 participants were recruited, which included 119 seeds and a maximum of 16 recruitment waves (mean chain length = 1.7). The sample of >0 recruitment waves removed unproductive seeds (n = 50/119, 42.0%), resulting in 69 chains (mean length = 3.0). The sample of >1 recruitment waves removed 125 seeds or recruits (17.4% of overall sample), resulting in 37 chains (mean length = 4.8). The final sample of >2 recruitment waves removed a further 182 seeds or recruits (25.3% of overall sample), resulting in 25 chains (mean length = 6.1). Convergence plots and bottleneck analyses of condomless anal intercourse with HIV discordant/unknown status partner and injecting drugs outcomes were satisfactory. For these two outcomes, regardless of seed exclusion criteria used, the crude proportions fell within 95% confidence intervals of all RDS-weighted estimates. Significant differences between the three RDS estimators were not observed.
Within a sample of GBMSM in Vancouver, Canada, this RDS study suggests that when equilibrium and homophily are met, although potentially costly and time consuming, analysis is not negatively affected by large numbers of unproductive or lowly productive seeds.
应答驱动抽样（RDS）是一种越来越常用的同伴链式招募方法，用于对没有可靠抽样框架的“难以接触到”的人群进行抽样。RDS的实施成功率各不相同；一个潜在的负面因素是所使用的种子数量。
我们对一项关于居住在加拿大温哥华年龄≥16岁的男同性恋、双性恋和其他与男性发生性行为的男性（GBMSM）的RDS研究数据所产生的估计值进行了敏感性分析。参与者完成了一份关于人口统计学、性行为和物质使用情况的问卷。为了进行分析，我们使用了越来越严格的种子排除标准，从所有参与者开始，随后去除无产出的种子、招募波次≤1的链以及招募波次≤2的链。我们使用三种不同的RDS加权程序（RDS - I、RDS - II和RDS - SS）计算了三种不同结果（HIV血清学状态、与HIV状态不一致/未知的伴侣进行无保护肛交以及注射毒品）的估计值。我们还通过瓶颈分析和收敛图评估了种子依赖性。通过模拟分析评估了RDS估计量之间的统计差异。
总体而言，招募了719名参与者，其中包括119名种子，最多有16个招募波次（平均链长 = 1.7）。招募波次>0的样本去除了无产出的种子（n = 50/119，42.0%），得到69条链（平均长度 = ）。招募波次>1的样本去除了125名种子或招募对象（占总样本的17.4%），得到37条链（平均长度 = 4.8）。最终招募波次>2的样本又去除了182名种子或招募对象（占总样本的25.3%），得到25条链（平均长度 = 6.1）。与HIV状态不一致/未知的伴侣进行无保护肛交以及注射毒品结果的收敛图和瓶颈分析结果令人满意。对于这两个结果，无论使用何种种子排除标准，粗略比例都落在所有RDS加权估计值的95%置信区间内。未观察到三种RDS估计量之间存在显著差异。
在加拿大温哥华的GBMSM样本中，这项RDS研究表明，当满足均衡性和同质性时，尽管可能成本高昂且耗时，但大量无产出或低产出种子不会对分析产生负面影响。
