Impact of Prescription Drug Monitoring Program Implementation on Rates and Characteristics of People Seeing Multiple Prescribers in Primary Care: A Controlled Interrupted Time-Series Analysis.

Abstract

To examine changes in rates of primary care patients seeing multiple prescribers and characteristics of patients who ceased seeing multiple prescribers for monitored medicines after voluntary implementation of the Victorian prescription drug monitoring program (PDMP). Controlled interrupted time series analysis of primary care electronic medical records. A total of 562 general practices across three Victorian healthcare networks (Monash Health, Peninsula Health, Eastern Health). People prescribed at least one PDMP-monitored medicine (e.g., opioids, benzodiazepines) and/or non-monitored psychotropic medicines (e.g., antidepressants, antipsychotics) between 1 January 2017 and 30 June 2023. Voluntary (1 April 2019) and mandatory (1 April 2020) implementation of the Victorian PDMP. Changes in the monthly rate of people seeing multiple prescribers (defined as four or more prescribers) following PDMP implementation for monitored medicines, with non-monitored medicines used as a control; characteristics of people who ceased seeing multiple prescribers for monitored medicines following PDMP implementation. Following voluntary PDMP implementation (1 April 2019), there was a significant reduction in the differential step and trend changes in the rates of seeing multiple prescribers between people prescribed monitored and non-monitored medicines (differential step change: β, -3.55 [95% confidence interval (CI), -5.08 to -2.03]; differential trend change: β, -0.29 [95% CI, -0.46 to -0.12]). Following mandatory PDMP implementation (1 April 2020), there was no significant step change difference. However, there was an increase in the differential trend change in the rate of seeing multiple prescribers between those prescribed monitored and non-monitored medicines (differential trend change: β, 0.21 [95% CI, 0.05-0.37]; p = 0.009). Logistic regression revealed that older age (95% CI, 1.39-1.75), male gender (95% CI, 1.09-1.25), metropolitan residence (95% CI, 1.04 and 1.23) and substance use disorder diagnosis (95% CI, 1.07-1.28) were associated with significantly higher odds of seeing multiple prescribers before PDMP implementation. Implementation of the PDMP was associated with meaningful reductions in people accessing monitored medicines from four or more prescribers.