Characterizing latent classes of social support among persons who inject drugs

Drug Alcohol Depend. 2020 Feb 1:207:107816. doi: 10.1016/j.drugalcdep.2019.107816. Epub 2019 Dec 23.

Abstract

Background: Social network members influence risk and health behaviors, yet little is known about the support that persons who inject drugs (PWID) receive from their closest ties.

Methods: 970 participants from the AIDS Linked to the IntraVenous Experience (ALIVE) study completed a social network survey between April 2016 and June 2017 about their five closest ties. Our analysis: 1) identified latent classes of support received by participants; 2) determined whether class membership differed by current (≤12 months) vs. former (>12 months) injection drug use; 3) compared classes of support by individual and network characteristics.

Results: 970 participants listed 3,388 network members. We identified three support classes: (1) Moderate (n = 249): probabilities of support <0.40; (2) High (n = 366): probabilities of support 0.58-0.82; (3) Very high (n = 355): probabilities of support 0.91-0.99. In adjusted analysis compared to moderate support, the odds of high and very high support increased as the mean age of network members increased (Adjusted Odds Ratio [AOR]:1.03; 95 % CI: 1.01, 1.05) and as the number of network members with whom they interacted daily increased (AOR-high: 1.33; 95 % CI:1.14, 1.56 and AOR-very high: 1.54; 95 % CI: 1.30, 1.83). While current injection drug use was associated with lower unadjusted odds of high and very high support, the associations were not statistically significant in adjusted analysis.

Conclusions: Support was higher among networks of older ties and more frequent interaction, but differences did not appear to be driven by injection drug use status. Findings point to the importance of the closest social ties.

Keywords: Latent class analysis; Persons who inject drugs; Social network; Social support.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Age Factors
  • Female
  • Humans
  • Latent Class Analysis*
  • Male
  • Middle Aged
  • Social Networking
  • Social Support*
  • Substance Abuse, Intravenous / psychology*
  • Surveys and Questionnaires
  • Time Factors