Article Text
Abstract
Background Places where people meet sex partners may play an important role in the propagation of sexually transmitted infections. Social network analysis (SNA) has the potential to quantify the role that places of social aggregation play in syphilis transmission based on a relational approach. The primary objective of this study was to explore the use of SNA as both an epidemiological and methodological tool to determine the relative importance of sex partner meeting venues to the transmission of syphilis, in a sustained infectious syphilis outbreak.
Methods In a network survey study, we identified and enrolled 52 cases and named contacts of infectious syphilis among individuals, aged 18−75 years at a Sexually Transmitted Disease clinic in Alberta Canada, during routine public health measures of infectious disease control between April and August, 2009. In addition to standard contact tracing information, participants were asked to list all venues attended in the last 6 months where sexual partnering may have occurred. We constructed a sexual affiliation network by linking together persons infected with syphilis, and their named sexual contacts, to sex partner meeting venues. By transposing the sexual affiliation matrix and applying matrix multiplication we created two separate networks; a network of persons connected by venues and a dual network of venues connected by persons. Hierarchal clustering was performed to model patterns of individual patronage of venues, and network algebraic measures of centrality and permutation statistical methods were used to determine what type of venue connected the most individuals infected with syphilis.
Results 77% of participants reported meeting a sex partner at a social venue in the last 6 months. We identified a densely connected sexual affiliation network of 94 men who have sex with men (MSM), comprised of 18 cases of infectious syphilis and 76 named sexual contacts connected by 21 venues. In the network of sex partner meeting venues, Internet venues had higher degree centrality than non-internet venues (p<0.05). In the network of men connected by venues, hierarchal clustering detected a cluster of 35 men linked together by their patronage of three Internet venues see Abstract P1-S4.10 Figure 1. These three Internet venues had the highest degree centrality in the network of sex partner meeting venues and connected two thirds of all infectious syphilis cases.
Conclusions To our knowledge, this is the first study to use SNA of a sexual affiliation network to quantify the importance of places in an outbreak of infectious syphilis. Network analysis allowed identification of three key venues that connected individuals who were infected with syphilis. These venues could provide public health officials with an epidemiologic target for primary and secondary prevention strategies to prevent further dissemination of disease.