Regional Price Level Estimates for Medical Services in the United States

To estimate regional price levels for medical services in the United States by type of service and in aggregate. To compare medical and non-medical price variation, examine the relationship between prices and spending, and develop a deflator-based utilization measure.

I measure state-level medical price variation using hedonic regression models that control for differences in service mix and patient characteristics. I estimate separate models for inpatient, outpatient, and professional services, and compute expenditure-weighted aggregate price levels. The results are used to construct new utilization measures, quantify the share of spending variation explained by price levels, and examine the relationship between medical and non-medical price levels using price parity estimates from the BEA.

I use commercial health care claims from the Health Care Cost Institute (HCCI) database and the Merative MarketScan database from 2018 to 2022.

Medical prices are 70%-80% higher in the most expensive states than in the least expensive states. Alaska, Wyoming, Wisconsin, Oregon, and California tend to have the highest medical prices, while Alabama, Arkansas, Kentucky, Michigan, and Louisiana tend to have the lowest, although there is considerable heterogeneity across service categories. Medical prices are significantly more disperse than non-medical prices, and the correlation between the two is weak across states (0.27). Price variation explains about one-half of the variation in health care spending per beneficiary. The MarketScan and HCCI databases yield similar estimates.

Commercial medical prices vary considerably across states, and this variation is not strongly correlated with non-medical price levels. This suggests that market forces governing health care prices are only weakly related to those affecting non-medical goods and services prices. Additionally, price variation is a significant driver of spending variation, implying that policies to reduce prices in expensive states could significantly reduce spending.