Elsevier

Drug and Alcohol Dependence

Volume 157, 1 December 2015, Pages 188-191
Drug and Alcohol Dependence

Short communication
Substance use network characteristics and drug and alcohol use behaviors among young men who have sex with men (YMSM)

https://doi.org/10.1016/j.drugalcdep.2015.10.003Get rights and content

Highlights

  • Substance use network characteristics are used to model substance use behavior.

  • Transitive networks are linked to higher probability and frequency of use.

  • Network effects remain after controlling for individual and dyadic factors.

Abstract

Background

Young men who have sex with men (YMSM) use alcohol and other drugs at rates higher than their heterosexual peers. While social networks of YMSM have been acknowledged as an important contextual influence on their health behavior, studies have largely focused on social and sexual networks rather than substance use networks, despite the potential importance of substance use alters in shaping substance use behavior.

Method

Using data collected from a diverse sample of YMSM (n = 156), two multilevel models examined the associations between network (e.g., degree and transitivity), dyadic (e.g., strength of relationship), and individual characteristics and two alter level dependent variables: recent drug use (versus no recent drug use) and frequency of substance use.

Results

Results indicated that transitivity was associated with both recent drug use (OR = 1.21, p = 0.012) and more frequent substance use (b = 0.08, p = 0.002). Degree was not significantly associated with either variable. Furthermore, participants were also less likely to have recently used drugs (OR = 0.93, p < 0.001) and tended to use substances less frequently (b = −0.01, p = 0.028) with older alters.

Conclusions

Despite substantial interest in network influences on health, much remains unknown about the impact of network structures on substance use. The current findings suggest that structural characteristics of substance use networks may provide important information regarding the drug and alcohol use behavior of YMSM even when controlling for dyadic and individual characteristics.

Introduction

Alcohol (Marshal et al., 2008) and drug use (Corliss et al., 2010, Marshal et al., 2009, Newcomb et al., 2014) are disproportionately higher among young men who have sex with men (YMSM) and other sexual minority adolescents than among their heterosexual peers. Alcohol and other drug use among YMSM has also been closely associated with a number of risk behaviors and negative health outcomes such as condomless sex (Bruce et al., 2012, Newcomb, 2013) and mental health problems such as post-traumatic stress disorder and depression (Halkitis et al., 2013). Given the heighted levels of substance use among YMSM and health burdens associated with this use, a growing body of research has examined predictors of alcohol and drug use (Newcomb et al., 2012, Thiede et al., 2003, Traube et al., 2012, Traube et al., 2013).

However, much remains unknown about contextual factors that may facilitate or reduce drug and alcohol use among YMSM. Social networks, or the structure of social ties between individuals, are an important determinant of substance use behavior among adolescents and young adults (Ennett et al., 2006, Schensul and Burkholder, 2005, Unger and Chen, 1999, Valente et al., 2004). Substance use networks (i.e., relationships between individuals that use drugs or alcohol together) are a subset of one's social network that may be especially important for understanding substance use in young adults. For example, previous studies indicate that members of an individual's social network, such as friends, are the most common source used to obtain drugs (Harrison et al., 2000, McCabe and Boyd, 2005) and, for individuals under the legal drinking age, alcohol (Harrison et al., 2000, Wagenaar et al., 1996). Given the reliance on networks to obtain these substances, the structure and composition of substance use networks may be especially important to understanding drug and alcohol use behaviors.

The aim of the current study is to examine the relationships between network (i.e., transitivity and network size), dyadic (e.g., age difference), and individual characteristics and drug and alcohol behavior with substance use alters to better understand the social and contextual factors associated with substance use behavior among YMSM. Specifically, we examined if these characteristics are associated with drug use and the frequency of substance use. Our analysis is novel as much of the previous research on substance use networks has almost exclusively focused on people who inject drugs (Latkin et al., 2010, Tobin et al., 2010), and most network studies of YMSM have largely focused on social and sexual networks (Birkett et al., 2015, McFadden et al., 2014, Mustanski et al., 2015, Tucker et al., 2012) rather than on substance use networks.

Section snippets

Participants

Cross-sectional egocentric network interviews were conducted with 175 YMSM (Birkett et al., 2015, Kuhns et al., 2015, Mustanski et al., 2015) taken from an ongoing longitudinal research study of YMSM recruited via a modified respondent driven sampling design. Participants whose visit coincided with the period of the network sub-study were eligible for the network interviews and all participants provided informed consent to be part of this study. A subsample of participants (n = 156) who indicated

Descriptive statistics

Participants reported an average of 12.27 substance use alters (SD = 8.78), with each alter having substance use ties with on average three other substance use alters (M = 3.00, SD = 2.91). The majority of substance use alters were identified as friends (61.6%) followed by other roles (15.3%; e.g., coworker, roommate, acquaintance), family (12.3%), serious partners (10.1%), and support (0.7%; e.g., minister, teacher). The majority of substance use alters were not sexual partners (73.2%), followed by

Discussion

Despite the abundant enthusiasm for utilizing network information to improve the implementation of health interventions (Valente, 2012), very little is known or understood about the specific network-level drivers of health, and how these specific drivers may differ in function by tie type, by population, and by health behavior. Therefore, to begin to tease apart the complex individual, dyadic, and network drivers of drug and alcohol use, this study provides the first detailed analysis of

Role of funding source

This research was partially supported by grants from the National Institute of Drug Abuse (Dual PIs: Mustanski & Garofalo, R01DA0255548; PI: Mustanski; R01DA025548-02S1; PI: Birkett, R03DA033906; PI: Birkett, K08DA037825).

Conflict of interest

No conflict declared.

Contributors

All authors contributed to the design, interpretation, and preparation of this manuscript.

References (27)

  • L. Kuhns et al.

    Modifying and adapting participant-aided sociograms for a study of social, sexual and substance-using networks among young men who have sex with men

    Connections

    (2015)
  • C.A. Latkin et al.

    Social norms, social networks, and HIV risk behavior among injection drug users

    AIDS Behav.

    (2010)
  • P.V. Marsden

    Network data and measurement

    Annu. Rev. Sociol.

    (1990)
  • Cited by (4)

    • Not getting high with a little help from your friends: Social versus drug network correlates of marijuana use among YMSM

      2019, Addictive Behaviors
      Citation Excerpt :

      For example, Barman-Adhikari et al. (2015) found strong associations between an individual's proximity to peer drug use and their own drug use behavior in a study of homeless youth. Similarly, a preliminary study found that an individual's drug use network density was associated with higher levels of drug and alcohol use (Janulis et al., 2015). Furthermore, one of the most common drug use behaviors in youth – marijuana use – has received less attention versus network correlates of injection drug risk behavior (De et al., 2007) or sexual risk behavior (Green et al., 2013; Latkin et al., 2003).

    View full text