Division of Epidemiology & Community Health, School of Public Health, University of Minnesota Twin Cities, 1300 South 2nd Street, Minneapolis, MN, 55454, USA.
Department of Biostatistics, University of North Carolina at Chapel Hill, 135 Dauer Drive, 3101 McGavran-Greenberg Hall, CB #7420, Chapel Hill, NC, 27599, USA.
BMC Infect Dis. 2019 Mar 5;19(1):213. doi: 10.1186/s12879-019-3700-5.
Latent class analyses (LCA) are increasingly being used to target specialized HIV interventions, but generalizability of emergent population structures across settings has yet to be considered. We compare LCA performed on two online samples of HIV negative Chinese men who have sex with men (MSM) to detect more generalizable latent class structures and to assess the extent to which sampling considerations impact the validity of LCA results.
LCAs were performed on an 1) nationwide online survey which involved no in-person contact with study staff and a 2) sentinel surveillance survey in which participants underwent HIV and syphilis testing in the city of Guangzhou, both conducted in 2014. Models for each sample were informed by risk factors for HIV acquisition in MSM that were common to both datasets.
An LCA of the Guangzhou sentinel surveillance data indicated the presence of two relatively similar classes, differing only by the greater tendency of one to report group sex. In contrast an LCA of the nationwide survey identified three classes, two of which shared many of the same features as those identified in the Guangzhou survey, including the fact that they were mainly distinguished by group sex behaviors. The final latent class in the nationwide survey was composed of members with notably few risk behaviors.
Comparisons of the latent class structures of each sample lead us to conclude that the nationwide online sample captured a larger, possibly more representative group of Chinese MSM comprised of a larger, higher risk group and a small yet distinct lower group with few reported behaviors. The absence of a lower risk group in the Guangzhou sentinel surveillance dataset suggests that MSM recruited into studies involving free HIV/STI testing may oversample MSM with higher risk behaviors and therefore greater risk perception. Lastly, two types of higher risk MSM were emergent across both samples distinguished largely by their recent group sex behaviors. Higher odds not only of self-reported HIV infection but also of closeted tendencies and gender fluid identities in this highest risk group suggest that interacting factors drive individual and structural facets of HIV risk.
潜类分析（LCA）越来越多地被用于针对特定的 HIV 干预措施，但新兴人群结构在不同环境下的可推广性尚未得到考虑。我们比较了在中国 HIV 阴性男男性行为者（MSM）的两个在线样本中进行的 LCA，以发现更具普遍性的潜在类别结构，并评估抽样考虑因素对 LCA 结果有效性的影响程度。
对 1）全国性在线调查进行 LCA，该调查不与研究人员进行任何面对面的接触，以及 2）在广州市进行的哨点监测调查，参与者在该市接受 HIV 和梅毒检测，这两项研究均于 2014 年进行。每个样本的模型均由两个数据集共有的 MSM 中 HIV 感染的风险因素提供信息。
广州市哨点监测数据的 LCA 表明，存在两种相对相似的类别，仅在一个类别的群体性行为倾向更大方面有所不同。相比之下，全国性调查的 LCA 确定了三个类别，其中两个类别具有许多相同的特征，与广州市调查中确定的类别相似，包括它们主要通过群体性行为来区分的事实。全国性调查中的最后一个潜在类别由行为风险较低的成员组成。
对每个样本的潜在类别结构的比较使我们得出结论，全国性在线样本捕获了更大、可能更具代表性的中国 MSM 群体，其中包括一个更大、风险更高的群体和一个规模较小但特征明显、行为风险较低的小群体。广州市哨点监测数据集中缺乏低风险群体表明，参与涉及免费 HIV/性传播感染检测的研究的 MSM 可能会过度抽样风险行为更高、风险感知更强的 MSM。最后，两个风险较高的 MSM 类型在两个样本中都出现了，主要区别在于他们最近的群体性行为。在这个最高风险群体中，自我报告的 HIV 感染、隐瞒倾向和性别流动身份的可能性更高，这表明相互作用的因素推动了 HIV 风险的个体和结构方面。
