Department of Economics, American University, Washington, DC, USA.
J Int AIDS Soc. 2013 Feb 12;16(1):17431. doi: 10.7448/IAS.16.1.17431.
This article explores three critical topics discussed in the recent debate over concurrency (overlapping sexual partnerships): measurement of the prevalence of concurrency, mathematical modelling of concurrency and HIV epidemic dynamics, and measuring the correlation between HIV and concurrency. The focus of the article is the concurrency hypothesis - the proposition that presumed high prevalence of concurrency explains sub-Saharan Africa's exceptionally high HIV prevalence. Recent surveys using improved questionnaire design show reported concurrency ranging from 0.8% to 7.6% in the region. Even after adjusting for plausible levels of reporting errors, appropriately parameterized sexual network models of HIV epidemics do not generate sustainable epidemic trajectories (avoid epidemic extinction) at levels of concurrency found in recent surveys in sub-Saharan Africa. Efforts to support the concurrency hypothesis with a statistical correlation between HIV incidence and concurrency prevalence are not yet successful. Two decades of efforts to find evidence in support of the concurrency hypothesis have failed to build a convincing case.
本文探讨了近期关于同时性（重叠性关系）争论中讨论的三个关键问题：同时性的流行率的测量、同时性与 HIV 流行动力学的数学建模，以及 HIV 和同时性之间相关性的测量。文章的重点是同时性假说，即假定的同时性高流行率解释了撒哈拉以南非洲异常高的 HIV 流行率。最近使用改进的问卷调查设计进行的调查显示，该地区报告的同时性发生率从 0.8%到 7.6%不等。即使在对合理的报告错误水平进行调整后，适当地参数化的 HIV 传染病的性网络模型也不能在撒哈拉以南非洲最近的调查中发现的同时性水平上产生可持续的传染病轨迹（避免传染病灭绝）。用 HIV 发病率和同时性流行率之间的统计相关性来支持同时性假说的努力尚未成功。二十年来，人们一直努力寻找支持同时性假说的证据，但都未能提出令人信服的案例。
