Applying Social Network Interventions in Men Who Have Sex With Men: A Systematic Review
Communication 623 Final Paper
Presented in partial fulfillment of the requirements for Health Psychology Seminar at Annenberg School for Communication, University of Pennsylvania
by Qijia Chen May 2014
Instructor: Professor John B. Jemmott, Annenberg School for Communication
The current paper provides a comprehensive review of the literature on social network intervention in HIV prevention among MSM. I performed a systematic literature research in multi-lingual databases and located a relatively exhaustive collection of articles on social network interventions in MSM. There is ...view middle of the document...
Social network intervention has been successfully implemented in various target groups. In a systematic literature review, Wang et al. (2011) synthesized 11 network-based condom promotion intervention studies. All of the studies found substantial improvement in condom use or biological outcomes when comparing intervention and control conditions. Although only two out of the 11 included in Wang et al. (2011)’s review targeted MSM, the strong evidence of the effectiveness of social network interventions suggests that such interventions should have comparable, if not greater, effects among MSM. The adoption of network sciences in the public heath arena is relatively recent and multiple, often confusing, terms have been employed to refer to network constructs. To better explicate the scope and goals of the current analysis, I will first briefly clarify the definitions of some of the terms. A risk-potential network is defined as a pattern of multiple ties between two people that can spread infection if the infectious agent is present (Friedman, 1999). In the current analysis, risk-potential networks are conceptualized as a preexisting state determined by local
epidemiological and demographic features, rather than a vehicle for intervention. An egocentric network considers only the direct linages of a given person (“ego”). Operationally, an egocentric network relies solely on ego's self-reported network (“alters”) (Friedman, 2001). Therefore, by definition, an egocentric network might not include all the existing risk-potential linkages. Sociometric networks (or "social network") consist of a set of people and the entire pattern of linkages among them (Rothenberg et al., 1998). In sociometric networks, multidirectional linkages typically connect members (Wasserheit & Aral, 1996). The studies selected in the present paper assembled either egocentric or sociometric networks through volunteering naming by seeds or peer leaders or observation by ethnographic researchers. In some cases, researchers might name the egocentric networks identified by participants as "sociometric network", even though the networks might not reflect all necessary linkages between members. The current study focuses on social network intervention that includes all type of network “alters” (e.g., kin, friends, colleague’s, community members, or sex partners) rather than sexual relationship only. Although the latter might provide greater insights or better capture the riskpotential network, the emphasis of this paper is on social networks (egocentric and sociometric), particularly how researchers could employ the structural characteristics of such networks in HIV interventions with MSM. Network characteristics can include network size, network density, and perceived ego-alter closeness, and the effectiveness of social network interventions will be subject to the impact of interactions of these characteristics. The current analysis will focus on two forms of impact of social...