Variations in prescription drug monitoring program use by prescriber specialty

Background: Although prescription drug monitoring programs (PDMPs) have been widely implemented to potentially reduce abuse of prescription opioids, there is limited data on variations in PDMP use by prescriber specialty. Such knowledge may guide targeted interventions to improve PDMP use. Methods: Using data from Washington state Medicaid program, we performed a retrospective cohort study of opioid prescribers and their PDMP queries between Nov 1, 2013 and Oct 31, 2014. PDMP registration was mandatory for emergency physicians, but not for other providers. The unit of analysis was the prescriber. The primary outcome was any prescriber queries of the PDMP. We used multivariate regression models to identify variations in PDMP queries by prescriber specialty, as well as to explore explanatory pathways for observed variations. Results: We studied 17,390 providers who prescribed opioids, including 8718 (50%) who were not registered with PDMP, 4767 (27%) who were registered but had no recorded use of the PDMP, and 3905 (23%) PDMP users (queries/user: median 18, IQR 5–64). Compared to general medicine physicians, PDMP use was higher for emergency physicians (OR 1.4, 95%CI: 1.2–1.7), and lower for surgical specialists (OR 0.1, 95%CI: 0.08–0.1), obstetrician-gynecologists (OR 0.2, 95%CI: 0.1–0.2) and dentists (OR 0.4, 95%CI: 0.4–0.5). Higher use by emergency physicians appeared to be mediated by higher registration rates, rather than by provider level predilection to use the PDMP. Conclusions: A minority of opioid prescribers to Medicaid beneficiaries used the PDMP. We identified variations in PDMP use by prescriber specialty. Interventions to increase PDMP queries should target both PDMP registration and PDMP use after registration, as well as specialties with current low use rates.