Synergistic effects of exposure to multiple types of violence on non-fatal drug overdose among women who inject drugs in Indonesia

Background
While research has demonstrated associations between experiencing violence from intimate and non-intimate partners and non-fatal drug overdose among women who inject drugs, existing studies focus predominantly on the Global North and are analytically limited. Guided by syndemics theory, this study examined whether different forms of gender-based violence exert independent and interactive effects on non-fatal drug overdose among women who inject drugs in Indonesia.

Methods
We recruited 731 cisgender adult women who injected drugs in the preceding year via respondent-driven sampling. We used multivariate logistic regressions to examine associations between self-reported intimate partner violence (IPV), police sexual violence, and police extortion, and non-fatal drug overdose, with covariance adjustment for factors drawn from the risk environment. We tested for interaction effects among violence measures by calculating metrics for attributable proportion (AP), relative excess risk due to interaction (RERI), and synergy index (S).

Results
Experiencing IPV (AOR 2.5; 95 % CI 1.2, 5.1; p = 0.012), police extortion (AOR 2.2; 95 % CI 1.5, 3.2; p ≤ 0.001), and police sexual violence (AOR 3.7; 95 % CI 1.5, 9.4; p = 0.005) each independently predicted non-fatal overdose, after adjusting for potential confounders. A significant positive interaction was detected between IPV and police sexual violence on drug overdose (AP=0.6, p = 0.001; S = 3.8, p = 0.015) such that the joint effect of these two forms of violence was associated with a nearly fourfold increase in non-fatal overdose risk compared to the main effects of each violence exposure.

Conclusion
This is the first study to show that concurrent IPV and police sexual violence exert an amplifying effect on non-fatal overdose beyond the additive effects of each exposure. Supporting the value of gender-responsive harm reduction services that integrate violence and overdose responses, results suggest that eliminating one form of violence when multiple forms of GBV are present could magnify the expected reduction in overdose.