Understanding human sexual networks: a critique of the promiscuity paradigm

Abstract

Increases in sexually transmitted infection (STI) and HIV rates worldwide have prompted the dedication of research to identifying transmission co-factors, with one such co-factor being an individual's number of different sexual partners. Currently, the majority of STI/HIV transmission models are based on the assumptions that sexual networks have random distributions; whereas, in real-life, these assumptions have proven incorrect because group norms produce variations in sexual practices and differences in transmission co-factors (i.e. number, type, and timing of sexual contacts, use of protection, and genital co-infections). In fact, sexual groupings follow the distribution of the scale-free network. Because human sexual assemblages form scale-free networks, a large number of sexual partners does not necessarily mean that an individual is at risk for acquiring an STI, or conversely, that a small number of partners means that an individual is not at risk. Therefore, while an individual's number of sexual partners is important for population-based and case-management initiatives, it is impossible to determine group sexual norms, network location, and γ values at the individual level. This signifies that a reliance on individuals’ number of different sexual partners to determine their need for STI/HIV testing may be an unnecessarily invasive practice that negatively impacts on testing practices. Thus, it is important to be aware that it is not so much this number of contacts that is important, but rather what occurs during these connections at a network level, and how many concurrent connections exist across the group.