Opioid Prescription Dispensing Patterns in Patients with Bipolar Disorder: Real-World Evidence from the IBM Market Scan Research Databases

Abstract

Objective

Prescription opioid dispensing patterns over time were assessed for individuals with bipolar disorder (BD) vs matched controls.

Methods

Health insurance claims data from the IBM MarketScan Commercial Database and Multi-State Medicaid Database were analyzed. Individuals aged 18 to 64 with ≥1 inpatient or ≥2 outpatient claims for BD during the year preceding the analysis year (2015-2019) were included, with age- and sex-matched controls. Baseline demographic and clinical characteristics were evaluated. Opioid dispensing during each analysis year was defined as either chronic (coverage for ≥70 days in any 90-day period, or ≥6 prescriptions dispensed during analysis year) or nonchronic (≥1 prescription dispensed, not meeting chronic definition).

Results

BD patients had a higher prevalence of medical and psychiatric comorbidities, including pain diagnoses, vs controls. Among patients with BD in the Commercial database, chronic opioid dispensing decreased from 11% (controls: 3%) in 2015 to 6% (controls: 2%) in 2019, and in the Medicaid database, from 27% (controls: 12%) to 12% (controls: 5%). Among patients with BD in the Commercial database, nonchronic dispensing decreased from 26% (controls: 17%) in 2015 to 20% (controls: 12%) in 2019, and from 32% (controls: 26%) to 25% (controls: 14%) in the Medicaid database.

Conclusion

Between 2015 and 2019, there was a significant decrease in chronic and nonchronic prescription opioid dispensing among BD patients and controls across both the Commercial and Medicaid databases. Despite this finding, it is important to note that both chronic and nonchronic opioid dispensing was consistently higher for BD patients vs controls over time, across both databases.

Funding

Alkermes, Inc.