Back

Prescription drug monitoring programs increase racial/ethnic inequities in unmet demand for substance use disorder treatment among people who inject drugs. A repeated cross-sectional analysis of people who inject drugs in 19 US metro areas in 2012, 2015, 2018, and 2022

Ibragimov, U.; Beane, S.; Haardörfer, R.; Haley, D. F.; Yarbrough, C. R.; Linton, S.; Beletsky, L.; Cooper, H. L.

2025-06-11 public and global health
10.1101/2025.06.10.25328800 medRxiv
Show abstract

BackgroundEvidence indicates that prescription drug monitoring programs (PDMPs) reduce demand for substance use disorder (SUD) treatment among the general population, perhaps by minimizing the risk of SUD onset through limiting access to prescribed opioids. Little is known about PDMP effects on SUD treatment among people who inject drugs (PWID), a population at high overdose risk. MethodsUsing four waves (2012, 2015, 2018, and 2022) of National HIV Behavioral Surveillance (NHBS), we conducted two-way fixed-effect modelling of associations of state-level "mandated review" PDMP policies and individual-level (1) SUD treatment utilization, and (2) unmet demand for SUD treatment among 24,518 PWID in 13 states. We tested effect modification by race/ethnicity. ResultsPDMPs were associated with an 8 percentage-point increase in the probability of unmet demand for SUD treatment in the sample as a whole (95% CI: 3.0, 12.0). PDMP implementation was also associated with an increased Black, Indigenous, Latinx, and other people of color (BILPOC) vs. White gap in the probability of unmet demand, from a 3.0 percentage-point gap in non-PDMP states (95% CI: 1.0, 5.0) to a 9 percentage-point gap in PDMP states (95% CI: 7.0, 11.0). ConclusionsPDMPs may increase racial/ethnic inequities in SUD treatment access. To strengthen PDMP effectiveness, supply reduction policies must be accompanied by enhanced access to SUD treatment and other services for PWID, particularly among BILPOC PWID.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.