Estimating the Impact of an Unsanctioned Safe Consumption Site in the United States: An Application of the Covariate Balancing Propensity Score

• Activity Number: 247 - Causal Inference and Statistical Learning of Intervention and Policy Effects
• Type: Contributed
• Date/Time: Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Health Policy Statistics Section
• Author(s): Erica Nicole Browne* and Barrot H Lambdin and Peter J. Davidson and Lynn H Wenger and Alex H Kral
• Companies: RTI International and RTI International and University of California San Diego, Department of Medicine and RTI International and RTI International
• Keywords: causal inference; propensity score; inverse probability of treatment weights; opioids; safe consumption site

Abstract

The public health impact of safe consumption sites (SCS, or facilities that provide a safe, clean space for people to inject or consume pre-obtained drugs under observation of trained staff) cannot currently be assessed in the US using the gold standard randomized controlled trial given their current lack of legal support. Yet, evidence specific to SCS’s impact in the US is considered necessary to warrant their adoption. In response to the opioid-related overdose epidemic, a SCS opened in 2014 in an urban area within the US without waiting for formal sanction. With data from an observational cohort study of 500 people who inject drugs residing near the SCS, we estimate the individual level causal effect of SCS use. We demonstrate the utility of applying the covariate balancing propensity score-based weighting technique to generate causal estimates [1] and offer strategies for how to present compelling results when key descriptives, like those which might be telling to the location of the unsanctioned SCS, need to be concealed.

1. Imai, K. and Ratkovic, M. (2014) Covariate balancing propensity score. J. R. Stat. Soc. B;76: 243-263.