Neglected Issues and Hypotheses Regarding the Impact of Sexual Concurrency on HIV and Sexually Transmitted Infections

Excerpt

In the past few years, the relative contribution of concurrent sexual partnerships where partnerships overlap over time (i.e., concurrency) to the HIV epidemic in sub-Saharan Africa, and the need to scale up interventions to reduce concurrency, has been discussed extensively [1‐11]. Theoretically, concurrency enhances HIV transmission compared to serial monogamy because it shortens the time between sexual contacts with different partners and therefore reduces the window period between the acquisition of infection and transmission to other partners. It also removes the protective effect of sequential partnerships since the first partners of an individual cannot be infected by subsequent infected partners of this individual [6]. At the population level, concurrency increases network connectivity and creates larger components than serial monogamy, thus favouring infection transmission [6, 7, 12]. Currently, the evidence that concurrency facilitates HIV spread is mainly based on network modelling studies [12‐16]. The validity of the empirical evidence supporting concurrency as an independent risk factor for HIV transmission at the individual or population level has been highly criticised, especially in the African context [1, 2, 6, 8, 11, 17‐19]. Some empirical evidence on the concurrency and HIV association comes from studies on sexually transmitted infections (STIs) in the US [5, 6, 11, 20‐22] rather than for HIV and African countries. Therefore a pivotal question to the current debate is: “If concurrency is important why hasn’t it been demonstrated empirically yet?” Answering this question is obviously relevant to interpret the epidemiological patterns of HIV and to design effective intervention strategies. …