Estimating Primary Care Spending in the United States: Toward a Common Method.

Abstract

Background: The lack of common definition and methods, coupled with the scarcity of suitable data sources, have impeded efforts to track primary care spending in the United States.

Objectives: Explore consistent approaches to estimating primary care spending.

Research design: A recently developed framework for primary care services was applied to 2 datasets: the Medical Expenditure Panel Survey (MEPS), a survey of noninstitutionalized individuals and their families in the United States, and MarketScan, a database containing health insurance claims of employees and their dependents for a selection of major US companies, to estimate primary care spending per-person-per-year (PPPY) and as percentage of total health care spending (PTHS) covering 2010-2021. Cross-sectional and trend analyses were conducted, and key methodological issues were explored.

Results: In 2019, average primary care spending was $504 PPPY (95% CI: $469-$539), accounting for 8.07% PTHS (95% CI: 7.56%-8.58%), based on MEPS, and $378 PPPY (95% CI: $377-$379), accounting for 6.30% PTHS (95% CI: 6.28%-6.32%), based on MarketScan. There were steady increases between 2010 and 2021 in PPPY primary care spending (from $309 to $639 based on MEPS and from $343 to $433 based on MarketScan), but small fluctuations in PHTS primary care spending (between 6% and 9%). Misalignments between the definitions and the data were identified, and standard errors for the estimates were calculated.

Conclusions: With explicit definitions, transparent methodologies and appropriate quantification of estimation uncertainty, comparable and reproducible estimates can be obtained to assess and track primary care spending in the United States.