Health, Costs, and Injection-Related Infections at a Hypothetical Overdose Prevention Center

Abstract

Importance: While the impact of overdose prevention centers (OPCs) on fatal overdose, HIV, and hepatitis C is well-characterized, the long-term clinical and cost impact on injection-related infections is unknown, and empirical data from the US are limited.

Objective: To estimate the 10-year impact of a hypothetical OPC in Denver, Colorado, on injection-related infections, hospitalizations, mortality, and associated costs from a payer perspective.

Design, setting, and participants: This decision analytical modeling study used a calibrated and validated Monte Carlo microsimulation model of the natural history of injection drug use. Data included people who inject drugs (PWID) in Denver, Colorado, simulated over 10 years, from 2023 to 2032. Input data were largely collected from the 2022 National HIV Behavioral Surveillance survey and published research on currently operating OPCs.

Exposure: Treatment service model for PWID; the status quo, in which 3 syringe service programs are operating, was compared with counterfactual scenarios in which a single OPC also operated and served between 10% and 70% of Denver's population of PWID.

Main outcomes and measures: Incidence of serious injection-related infections (SIRIs; infective endocarditis [IE] and skin and soft tissue infections [SSTIs]), hospitalizations, mortality, and associated costs from a health care payer perspective.

Results: The modeled population of 9697 PWID had an input mean (SD) age of 41.3 (1.9) years and included 74.1% male participants. Over a 10-year period, the status quo resulted in approximately 15 400 SIRIs, 15 000 hospitalizations for SIRIs and overdose, and 2400 deaths among Denver's population of PWID. Compared with the status quo scenario, a hypothetical OPC decreased SSTI incidence by up to 11.5% (95% credible interval [CrI], -16.8% to -6.4%), IE incidence by up to 22.0% (95% CrI, -27.6% to -6.1%), hospitalizations by up to 8.5% (95% CrI, -14.0% to -2.6%), and all-cause mortality by up to 5.8% (95% CrI, -12.4% to 4.9%), and saved between $7 million and $46 million from a payer perspective over 10 years, depending on the program reach. Findings were robust in sensitivity analyses.

Conclusions and relevance: In this simulation modeling study of the effect of a hypothetical OPC, implementing a single OPC in Denver was an effective intervention to decrease SIRIs, drug use-related hospitalizations, costs, and deaths among PWID.

Figure 1.. Harm Reduction Service Utilization Among PWID Under Various Scenarios Involving OPC Implementation

OPC indicates overdose prevention center; SSP, syringe service program; PWID, people who inject drugs.

Figure 2.. Ten-Year Clinical Outcomes (2022-2032) Among PWID by Scenario

IE indicates infective endocarditis; OPC, overdose prevention center; PWID, people who inject drugs; SSTI, skin and soft tissue infection. Error bars represent 95% credible intervals (CrIs) for each clinical outcome based on outputs of 1000 independent runs, each drawing unique sets of uncertain parameters (eg, the weekly probability of an individual with an SSTI linking to inpatient care) from given distributions. Error bars between scenarios may overlap, even when the 95% CrI for the change in an outcome between scenarios (reported in the text) does not include 0%. This is because the reported outcome of percentage change in a given clinical outcome between 2 scenarios was calculated assuming that the probabilistically drawn set of uncertain input parameters is the same in both scenarios being compared (ie, scenarios are not independent of one another).

Figure 3.. Total Costs to Health Care Payers Over 10 Years (2023-2032) Under Various Scenarios Involving OPC Implementation

OPC indicates overdose prevention center.