While marked gender-based differences in drug-related risk and harm between men women who use drugs have been characterized to some extent, the complex relationship between gendered socioeconomic conditions, overdose risk, and drug use patterns and behaviours remains underexplored. We conducted gender-stratified repeated measures latent class analyses (RMLCA) with data from two ongoing cohorts of people who use drugs in Vancouver, Canada to identify discrete subgroups based on socioeconomic exposures. Multivariable generalized estimating equations models weighted by the respective posterior membership probabilities were applied to estimate the associations between socioeconomic class membership and non-fatal overdose. Sub-analyses investigated whether these relationships were explained by drug use-related variables. Between June 2014 and March 2020, 1723 participants provided 12120 observations. At study baseline, 1074 (62.3%) were men and 649 (37.7%) were women. In RMLCA, five distinct profiles of socioeconomic exposures emerged for men, and four emerged for women. Classes were presented in ascending order of socioeconomic disadvantage (i.e., ranging from Class 1 [least disadvantage] to Class 4/Class 5 [most disadvantage]) and were primarily distinguished by variations in income, material and housing in/security, participation in illegal/informal income generation, and criminal justice involvement. In analyses adjusting for demographic characteristics, for both men and women, progressive increases in exposure to multiple dimensions of disadvantage were positively and independently associated with non-fatal overdose when compared to the least socioeconomically disadvantaged subgroup. Subsequent analyses revealed that systematic differences in drug use patterns and behaviours across categories of socioeconomic disadvantage largely accounted for these relationships. Addressing multiple forms of mutually reinforcing health inequities may offer additional opportunity to mitigate non-fatal overdose in men and women who use drugs.
Read full abstract