OBJECTIVES: Measure opioid exposure using combined data from household surveys and Medicare Part D claims and provide national prevalence estimates of exposure to opioids among the elderly Medicare population with the revised measure. Secondary objective is to report the gap between measurements based on Medicare Part D claims and survey reports of opioid use.

METHODS: Secondary analyses of data from the Medicare Current Beneficiary Survey, collected in year 2013, a nationally representative database of Medicare beneficiaries. Study population was limited to community dwellers (n=10,311).

RESULTS: Nearly 22% of the study population was exposed to opioids. Females, blacks, and Native Americans were more likely to be exposed. Beneficiaries under age 65 and over age 74 were more likely to use opioids, and lower income was associated with increased use. The most frequently prescribed drugs were tramadol and oxycodone, each representing one third of all prescriptions. Other substantial contributors were morphine, codeine, and fentanyl representing more than 5% each. More than one third of all opioid prescriptions were not identified by Medicare claims data.

CONCLUSIONS: This paper provides an update on estimates of opioid use among the elderly and discovers at minimum a 14% gap between prescriptions reported by survey versus Medicare claims. This information can be used by health service providers and policymakers when addressing the issue of problematic opioid use among the elderly.