Senior Actuarial Analyst Milliman, Inc. WIndsor, United States
Background: While increasing drug overdose deaths related to opioids have been recognized as a Public Health Emergency since 2017, increasing overdoses associated with other categories of drugs have generally received less recognition.
Objectives: To describe the prevalence of psychiatric prescription drug utilization and related overdose rates in the U.S., and how overdose rates vary with pharmacy fill patterns.
Methods: We summarized healthcare claims data from Milliman’s Consolidated Health Cost GuidelinesTM Sources Database for commercially insured members continuously enrolled in 2020. We identified psychiatric medications as those with Generic Product Identifiers with a therapeutic class of “antianxiety agents” or “antidepressants” as defined by Medi-Span®. We identified overdoses based on the presence of an inpatient claim with an ICD-10-CM diagnosis code beginning with T43 (poisoning by, but excluding those related to assault, adverse effect of, and underdosing of, psychotropic drugs).
Results: We identified 1,600,481 commercially insured members with any prescription fill for a psychiatric medication. We found the rate of overdoses related to psychiatric medications (regardless of the presence of a prescription fill) in the inpatient setting of 9.0 per 100,000 members in 2020. Six out of 10 overdoses were among members with an identifiable prescription fill. The rate of overdose increased with the number of pharmacies at which members filled psychiatric prescriptions, from 0.03% for those filling at 1 pharmacy, to 0.09% for 2 pharmacies, 0.19% for 3 pharmacies, and 0.56% for 4 or more pharmacies. We also found that the overdose rate among members that filled a psychiatric prescription(s) at 4 or more pharmacies was higher for those with a recorded diagnosis of depression and anxiety at 1.53% as compared to 0.10% for those who did not have either diagnosis recorded.
Conclusions: Our initial findings on patterns in psychiatric medication overdoses and pharmacy fill patterns suggests that additional research on potential risk factors, such as concomitant opioid use, may be warranted.
Differences between our findings and those for other populations depend on the extent that the data analyzed represents those populations, including members who use alternative means than the single health insurance plan analyzed to obtain psychiatric medications.
This analysis has been prepared for the specific purpose of describing healthcare utilization for psychiatric medication use and related overdoses. This information may not be appropriate, and should not be used, for any other purpose. Milliman does not intend to benefit or create a legal duty to any third party recipient of its work even if we permit the distribution of our work product to such third party.
