Abstract
Stimulant use among unstably housed individuals is associated with increased risks of psychiatric co-morbidity, violence, HIV transmission, and overdose. Due to a lack of highly effective treatments, evidence-based policies targeting the prevention of stimulant use disorder are of critical importance. However, little empirical evidence exists on risks associated with initiating or returning to stimulant use among at-risk populations. In a longitudinal cohort of unstably housed women in San Francisco (2016-2019), self-reported data on stimulant use, housing status, and mental health were collected monthly for up to 6 months, and factors associated with initiating stimulants after a period of non-use were identified through logistic regression. Among 245 participants, 42 (17.1%) started using cocaine and 46 (18.8%) started using methamphetamine. In analyses adjusting for demographics and socio-structural exposures over the preceding month, experiencing street homelessness was associated with initiating cocaine use (AOR: 2.10; 95% CI: 1.04, 4.25) and sheltered homelessness with initiating methamphetamine use (AOR: 2.57; 95% CI: 1.37, 4.79). Other factors-including race, income, unmet subsistence needs, mental health, and treatment adherence-did not reach levels of significance, suggesting the paramount importance of policies directed toward improving access to permanent supportive housing to prevent stimulant use among unstably housed women.