Supervised consumption sites: a nuanced assessment of the causal evidence

Supervised consumption sites (SCS) operate in more than 10 countries. SCS have mostly emerged as a bottom-up response to crises, first to HIV/AIDS and now overdose deaths, in ways that make rigorous evaluation difficult. Opinions vary about how much favorable evidence must accumulate before implementation. Our aim was to assess the nature and quality of evidence on the consequences of implementing SCS.

We reviewed the higher-quality SCS literature, focusing on articles evaluating natural experiments and mathematical modeling studies that estimate costs and benefits. We discuss the evidence through the lens of three types of decision-makers and from three intellectual perspectives.

Millions of drug use episodes have been supervised at SCS with no reported overdose deaths; however, uncertainties remain concerning the magnitude of the population-level effects. The published literature on SCS is large and almost unanimous in its support, but limited in nature and the number of sites evaluated. It can also overlook four key distinctions: (1) between outcomes that occur within the facility and possible spillover effects on behavior outside the SCS; (2) between effects of supervising consumption and the effects of other services offered, such as syringe or naloxone distribution; (3) between association and causation; and (4) between effectiveness and the cost-effectiveness of SCS compared to other interventions.

The causal evidence for favorable outcomes of supervised consumption sites is minimal, but there appears to be little basis for concern about adverse effects. This raises the question of how context and priors can affect how high the bar is set when deciding whether to endorse supervised consumption sites. The literature also understates distinctions and nuances that need to be appreciated to gain a rich understanding of how a range of stakeholders should interpret and apply that evidence to a variety of decisions.