TABLE 5.

Multivariate GEE confounder model for the independent association between police-related barriers to harm reduction strategies and non-fatal overdose amongst women sex workers who use drugs in Metro Vancouver, Canada, 2010–2017 (N=624)

	Adjusted Odds Ratio (95% CI)	p - value
Police-related barriers to harm reduction strategies*†	2.15 (1.60 – 2.90)	<0.001

^*Time updated measure using the last six months as a reference

^†Adjusted for physical/sexual workplace violence*, frequency of injection drug use*, use of non-injection opioids*, and interview year. Other variables considered, but which fell out of the multivariable model, included place of service, mental health diagnoses, injection opioid use, and injection stimulant use.