Multilevel Influences on HIV and Substance use in a YMSM Cohort

Funded by the National Institute of Drug Abuse, Award number 1U01DA036939-01.

Multilevel Influences on HIV and Substance use in a YMSM Cohort Young men who have sex with men (YMSM) ages 16-29 are disproportionately affected by HIV, with high prevalence and incidence. While all other risk groups have declining rates of infection, dramatic increases are occurring among YMSM. YMSM are understudied, but preliminary research demonstrates that the HIV epidemic in YMSM forms a syndemic with drug use, violence, and mental health problems. The study builds a dynamic dyadic-network cohort of 1,358 YMSM by hybridizing 2 existing cohorts of YMSM: Project Q2 (aged 20-24) and ChiGuys (aged 16-17). We will expand this unified cohort through the enrollment of serious sexual partners of core participants, thereby creating a dynamic dyadic network.

The overarching goal is to apply a “next generation” multilevel model (Network-Individual-Resources model) to understand the syndemic among a cohort of YMSM and inform “high-impact” HIV prevention with this vulnerable population.

We have four specific aims to achieve this goal:

Aim 1 is to understand how syndemics of substance use, HIV, STIs, mental disorders, and violence develop among YMSM using an accelerated longitudinal design. We will assess youth every 6 months during the transition from late adolescence to early adulthood, to describe the trajectories of syndemic development (including substance use) and use innovative statistical modeling to characterize different trajectories of syndemic development to predict youth most at risk of HIV acquisition.

Aim 2 is to determine how dyadic processes influence HIV risk behaviors and transmission among YMSM by enrolling their serious sexual partners into the cohort. We will utilize dyadic data to identify which sexual partner and relationship factors are longitudinally related to unprotected sex and HIV/STI transmission across development. Phylogenetic analyses will link HIV transmissions within dyads to determine what behavioral, partner and dyadic factors predict such transmission.

Aim 3 is to describe network and structural influences on syndemic development among YMSM. We will examine how sexual and drug use network structures explain racial disparities in HIV transmission, including network size, density, concurrency, and homophily. We will conduct ecological resource mapping to understand how access to resources impacts the HIV epidemic in YMSM, and how neighborhood resources and HIV prevalence impact HIV incidence.

Aim 4 is to determine if, and how, substance use increases HIV acquisition risk and viral set point. We hypothesized that substance use is associated with (a) acquiring HIV, (b) elevating viremia set point, and (c) evading host defenses early during incident HIV infections (HIV adaptation). We will build a repository of specimens/data before, during, and after incident HIV to facilitate collaborative studies.

Through these four aims, the multilevel data collected in this dynamic dyadic network of YMSM will provide an unprecedented window into HIV transmission dynamics.


Hogan, B., Melville, J. R., Philips, G. L., Janulis, P., Contractor, N., Mustanski, B. S., & Birkett, M. (2016). Evaluating the Paper-to-Screen Translation of Participant-Aided Sociograms with High-Risk Participants. Proceedings of CHI ’16, San Jose, CA.


Michelle BirkettNoshir ContractorPatrick Janulis