Auditing the Prescription Drug Consumer Price Index in a Changing Marketplace

The Bureau of Labor Statistics' Prescription Drug Consumer Price Index (CPI-Rx) has been the focus of considerable controversy in recent years. The CPI-Rx limits its sampling frame to transactions in retail outpatient outlets, and excludes prescription pharmaceuticals dispensed in hospitals, physician/clinic outpatient facilities, and nursing homes. Thus the CPI-Rx overlooks the increasingly important specialty pharmaceuticals dispensed in hospitals, outpatient clinics, and physician offices, whose transactions are instead captured in the overall hospital or professional services component of the medical CPI. Specialty drugs now account for 55% of all U.S. drug spending, double the amount of a decade earlier. To the extent specialty drug growth differs from that of traditional pharmaceuticals, the CPI-Rx could provide an inaccurate measure of overall drug price inflation. Using data from the IBM MarketScan database for the years 2010-2019 and IQVIA-designated specialty drugs, we estimate that by not sampling specialty drugs in non-retail settings, the CPI-Rx has understated overall U.S. prescription drug price inflation by just under 75 basis points annually. We discuss implications for health care policy, and suggest the BLS examine the feasibility of publishing an overall pharmaceutical price index incorporating both traditional and specialty pharmaceuticals dispensed in retail and non-retail settings.