AIM: For several decades, British Columbia (BC), Canada, has been experiencing a housing crisis marked by a shortage of safe and affordable housing, which coincides with a severe drug poisoning epidemic in the region. We explore the impact of housing instability on mortality (all-cause, drug-related) among a cohort of people with HIV (PWH) in BC. METHODS: Data are from the Longitudinal Investigation into Supportive and Ancillary Health Services (LISA) study (n = 997). Data from the cross-sectional LISA survey (2007-2010) are linked with administrative health data from Population Data BC and the BC Centre for Excellence in HIV/AIDS Drug Treatment Program (DTP) until March 31, 2020. We used inverse probability of participation weighting (IPPW) to address selection bias, introduced in LISA through oversampling of PWH marginalized by sociostructural inequities. We constructed participation weights using information from the DTP database, which includes all known PWH in BC accessing antiretrovirals via the DTP (including respondents and non-respondents to LISA). We estimated hazards of all-cause and drug-related mortality associated with housing instability using an adjusted, IPPW-weighted Cox proportional hazards model. RESULTS: In this sample, 317 (31.8%) people reported housing instability. Overall, 302 people (30.3%) died from any cause between the completion of the LISA survey until March 31, 2020
of those people, 138 (45.7%) experienced housing instability. Results suggest housing instability is associated with increased hazards of all-cause mortality (adjusted Hazards Ratio (aHR): 1.46
95% CI: 1.08-1.96). The association between housing instability and hazards for drug-related mortality include a range of values consistent with the null (aHR: 1.67
95% CI: 0.89-3.13). CONCLUSION: PWH experiencing housing instability may have greater hazards of all-cause mortality. Our findings add to the literature supporting a need to expand access to safe and affordable housing.