UC Berkeley

Background

Given the housing instability and frequent residential relocation (both volitional and hegemonic) of people who inject drugs, we sought to determine whether residential relocation (defined as sleeping in a different place in the past 30 days) is associated with health outcomes in a sample of people who inject drugs (PWID).

Methods

We recruited 601 PWID using targeted sampling and interviewed them between 2016 and 2018 in San Francisco and Los Angeles, CA about housing, drug use practices, and service utilization. We then developed multivariable regression models to investigate how residential relocation is associated with violence, health outcomes, and social service access. We analyzed our data between June 2018 and October 2019.

Results

Participants who relocated in the past 30 days had lower odds of being in substance use treatment (Adjusted Odds Ratio [AOR] = 0.62, 95% Confidence Interval [CI] = 0.42, 0.89) and higher odds of nonfatal overdose (AOR = 2.50, CI = 1.28, 4.90), receptive syringe sharing (AOR = 2.26, CI = 1.18, 4.32), severe food insecurity (AOR = 1.69, CI = 1.14, 2.50), having belongings stolen (AOR = 2.14, CI = 1.42, 3.21), experiencing physical assault (AOR = 1.58, CI = 1.03, 2.43), arrest (AOR = 1.64, CI = 1.02, 2.65), and jail (AOR = 1.90, CI = 1.16, 3.13) in the past 6 months when compared to those who did not relocate.

Conclusions

PWID who have relocated in the past 30 days have higher odds of experiencing violence and life- threatening adverse outcomes, and policies that disrupt living circumstances of PWID should be ended in favor of those that support housing stability.