Evaluation of Washington State Prescription Monitoring Program — 2011

BACKGROUND: Drug poisoning deaths have surpassed motor-vehicle traffic crashes as the leading acute cause of death nationally. Approximately 40% of all drug poisonings deaths are attributable to prescription drugs. State prescription monitoring programs (PMPs) aim to reduce prescription drug abuse by collecting and monitoring data regarding dispensed controlled substances. Washington state launched its PMP in October 2011 to capture all Schedule II–V controlled substances dispensed in Washington or shipped to a Washington resident.

METHODS: By following CDC’s “Updated Guidelines for Evaluating Public Health Surveillance Systems,” we evaluated Washington’s PMP to understand usefulness and assess system attributes. Interviews were conducted with Washington Department of Health staff administrating PMP, subject-matter experts, and public insurers (e.g., Medicaid and worker compensation). We assessed PMP’s usefulness in clinical practice through an online survey sent to  ~7,000 registered providers in September 2012. We examined PMP’s simplicity, flexibility, data quality, acceptability, sensitivity, predictive value positive, representativeness, timeliness, and stability for collecting information on dispensed controlled substances. Additionally, we evaluated the sensitivity, predictive value positive, and representativeness of detecting prescription drug abuse.

RESULTS: Of ~1,600 providers answering the online survey, >82% indicated the PMP was useful for monitoring patients’ controlled substance use. Medicaid reported the PMP identified >400 cases of suspected abuse or insurance fraud; workers’ compensation used PMP to better coordinate care for workers who filed claims. PMP’s strengths include simplicity (data are only collected from pharmacies); data quality (only 3% of records submitted each month contain errors); and sensitivity, predictive value positive, and representativeness of the system for capturing controlled substances dispensed in Washington. Weaknesses include inflexibility (to revise data collection fields); acceptability (only 21% of 35,000 eligible providers are registered with the system to use PMP); timeliness (data are inaccessible until ~7–10 days after dispensing); and stability (funding will terminate in June 2013). Data were unavailable for evaluating sensitivity, predictive value positive, and representativeness to detect prescription drug abuse, but concerns exist that the system incorrectly flags certain persons (e.g., those with chronic cancer pain) as possible abusers and does not capture persons who use medication dispensed to another person.

CONCLUSIONS: Washington’s PMP is useful for patient care and detecting fraud, but more providers will need to use the system for it to help curb prescription drug abuse and overdose deaths. Recommendations include decreasing the time from dispensing a prescription to data being accessible and securing permanent funding to guarantee the system’s future.