Prescription opioids have been linked to over half of the 28,000 opioid overdose deaths in 2014. High rates of prescription opioid non-medical use have continued despite nearly all states implementing large-scale prescription drug monitoring programs (PDMP), which points to the need to examine the impact of state PDMP's on curbing inappropriate opioid prescribing. In the short-term, PDMPs have been associated with short-term prescribing declines. Yet little is known about how such policies differentially impact patient subgroups or are interpreted by prescribing providers. Our objective was to compare volumes of prescribed opioids before and after Indiana implemented opioid prescribing emergency rules and stratify the changes in opioid prescribing by patient and provider subgroups.
An interrupted time series analysis was conducted using data obtained from the Indiana PDMP. Prescription level data was merged with census data to characterize patient socioeconomic status. Analyses were stratified by patients' gender, age, opioid dosage, and payer. The primary outcome indicator was the total morphine equivalent dose (MED) of dispensed opioids per day in the state of Indiana. Also considered were number of unique patients, unique providers, and prescriptions; MED per transaction and per day; and number of days supplied.
After controlling for time trends, we found that total MED for opioids decreased after implementing the new emergency rules, differing by patient gender, age, and payer. The effect was larger for males than females and almost 10 times larger for 0-20 year olds as compared to the 60+ age range. Medicare and Medicaid patients experienced more decline in prescribing than patients with private insurance. Patients with prescriptions paid for by workers' comp experienced the most significant decline. The emergency rules were associated with decline in both the number of prescribers and the number of day supply.
Although the Indiana opioid prescribing emergency rules impacted statewide prescribing behavior across all individual patient and provider characteristics, the emergency rules' effect was not consistent across patient characteristics. Further studies are needed to assess how individual patient characteristics influence the interpretation and application of state policies on opioid prescribing.