Background: Little is known about the prevalence of opioid shoppers in clinical practices and the relation between prescriber characteristics and the risk of having opioid shoppers.
Objective: Describe the prevalence of opioid shoppers in prescribers' practices. Assess the relation between prescribers' characteristics and patient opioid shopping behavior.
Design: Retrospective cohort study using a large US retail prescription database.
Patients, participants: Patients with ≥1 opioid dispensing were followed 18 months. These patients' prescribers are the focus of the study.
Main outcome measures: A patient was a "shopper" if he or she had opioid prescriptions written by ≥1 prescriber with ≥1 day of overlap filled at ≥3 pharmacies and a "heavy shopper" if he or she had ≥5 shopping episodes. The proportions of shoppers by prescriber and the proportion of prescribers with ≥1 shopper or heavy shopper were calculated.
Results: Among 858,290 opioid prescribers, most (87 percent) had no shoppers and 98 percent had no heavy shoppers. Prescribers who were aged 70-79 years, male, or who prescribed schedule II opioids had an increased likelihood of having shoppers. As the number of patients for whom a prescriber prescribed opioids increased, the proportion of shoppers also increased. Prescribers with 66 or more patients receiving opioids, who represented 25 percent of prescribers, prescribed for 82 percent of all shoppers.
Conclusion: The great majority of opioid prescribers appear to have no shoppers in their practice. Any educational program will be more cost effective if targeted to prescribers of schedule II opioids with a large volume of patients requiring opioids.