The effect of the Medicare Part D benzodiazepine exclusion on the utilization patterns of benzodiazepines and substitute medications.

BACKGROUND

Although the benzodiazepine exclusion policy in the U.S. Medicare Part D drug coverage program has been studied, little information is available on individual use and switching patterns between benzodiazepines and substitute medications. Patients voluntarily were continuing or stopping benzodiazepines or switching to substitute medications. These individual-level outcomes can provide information beneficial to providers and policymakers to better understand the intended and unintended consequences of exclusion policies.

OBJECTIVE

The objective was to determine the effect of the Medicare Part D benzodiazepine exclusion on the utilization patterns of benzodiazepines and substitute medications by a select group of Medicare beneficiaries for a year following implementation of the exclusion. This research focused on the examination of the within-person patterns of benzodiazepine use and factors associated with these patterns.

METHODS

A quasi-experimental, comparative study was used to analyze prescription patterns and multinomial regression models were applied to investigate factors predicting different benzodiazepine use patterns. Pharmacy dispensing data for continuously eligible Medicare beneficiaries with at least one benzodiazepine fill in 2005 were reduced to a comparison group of 216 individuals with continual coverage and an intervention group of 250 individuals who lost coverage for benzodiazepines. Four individual patients' drug use patterns, continuation, switch, fluid movement, and cessation were identified by sorting and arraying pharmacy dispensing data to apply systematic drug file review. Multinomial regression models were used to examine the impact of coverage, demographic, medical, economic, and pharmaceutical factors.

RESULTS

Significantly more Medicare seniors who lost benzodiazepine coverage switched to potential substitute medications than those who continued to have coverage. Interestingly, 12 percent of affected seniors and 6 percent of unaffected seniors switched from and back to benzodiazepines (fluid movement). Zolpidem was the most popular substitute agent despite being an expensive brand-name drug. Regression models revealed that affected individuals had nearly two times the odds of engaging in switch-related patterns than those who had continuous coverage. Also, women were twice as likely to discontinue benzodiazepines as men.

CONCLUSIONS

More seniors who lost benzodiazepine coverage engaged in medication switching, and women were more likely to stop benzodiazepines after the implementation of the exclusion policy.