Panel Paper: Health Insurance, Price Changes, and the Demand for Pain Relief Drugs: Evidence from Medicare Part D

Saturday, November 10, 2018
Wilson A - Mezz Level (Marriott Wardman Park)

*Names in bold indicate Presenter

Aparna Soni, Indiana University


This paper evaluates how prices affect the utilization of opioids and other pain relief drugs among the Medicare population. Opioid overdoses have increased dramatically over the past two decades, and the increased prevalence of prescription opioids is a key driver of the overdose epidemic. While past policies have focused on restricting the supply of opioids, policymakers have recently proposed market-based solutions, such as opioid taxes and removing opioids from insurance formularies. In order to assess the potential impact of these policies, it is important to understand how opioid utilization responds to price changes. I accomplish this by estimating price elasticities of demand for different types of prescription painkillers.

My identification comes from the introduction of Medicare Part D in 2006, which provided publicly subsidized prescription drug coverage to Medicare enrollees. The policy led to an exogenous reduction in out-of-pocket (OOP) drug prices faced by seniors. I obtain data on individuals’ detailed drug utilization and expenditures from the Medical Expenditure Panel Survey (MEPS), 1996-2015. I estimate difference-in-differences models, comparing out-of-pocket painkiller prices and purchases among individuals who were affected by Part D (those aged 65 to 74) with those who were not affected (those aged 55 to 64 and not on Medicare), before and after the introduction of the policy in 2006.

I find that Part D led to a $0.38 decrease in the average OOP price per dose of prescription painkillers (p<0.01), or a 28% decrease from pre-2006 levels. The quantity purchased (measured in units of “days supplied”) increased by 48.1 per year (p<0.01) or a 74% increase from pre-2006. This implies a price elasticity of demand for prescription painkillers of -2.67. Further analysis suggests that most of the quantity increase came from new opioid users, as opposed to those who had used opioids even before 2006. The demand for opioids appears to be highly price elastic, at least for new users, which suggests that market-based solutions, such as taxes, may be effective in curbing opioid use.

Full Paper: