Developing measures of pathways that may link large-scale social structural factors with HIV epidemiology

Issues: HIV/AIDS researchers have called for new measurements to enable studies of how large-scale social structural factors may affect HIV epidemics. Few measures exist at the individual level that could be used to assess structural influences or potential effects of future structural interventions.

Description: Through extensive qualitative interviews of people who inject drugs (PWID), men-who-have-sex-with-men (MSM) and high-risk heterosexuals (HRH), and pilot testing, we developed a questionnaire to measure variables representing pathways that may link social structural factors with HIV risk behaviors and risk networks. We collected questionnaire data from 300 PWID in New York City during 2012-2013. We assessed validity using Pearson's correlations with criterion variables (e.g., income, syringe sharing, unprotected sex), and scale reliability using Cronbach's alpha.

Lessons Learned: The questionnaire incorporated novel individual items on topics including sex and injecting at different venues, membership in formal and informal groups, time-use, and 16 new scales on topics including norms about sex and drug injecting under different conditions, experiences helping others, experiences of denial of dignity, and attitudes regarding competitive, altruistic, solidaristic, and traditional values. Correlations with criterion variables ranged from 0.17 to 0.52. Cronbach's alpha values ranged from 0.62 to 0.91.

Recommendation: Most scales and items showed strong evidence of validity and reliability. We are currently collecting data from HRH, and will collect data from MSM in late 2014. Comparing results among PWID, MSM and HRH will let us assess generalizability, and compare convergent and divergent validity. Measures we develop will be made available to public health researchers.