Customize your JAMA Network experience by selecting one or more topics from the list below.
Reiff J, Brennan N, Fuglesten Biniek J. Primary Care Spending in the Commercially Insured Population. JAMA. 2019;322(22):2244–2245. doi:10.1001/jama.2019.16058
Efforts to increase the value of health care by allocating more resources to primary care have used the share of total health care spending attributed to primary care as a measure of success. A 2019 study found that primary care represented 2% to 5% of total spending among Medicare fee-for-service beneficiaries in 2015.1 We assessed the share among individuals younger than 65 years covered by employer-sponsored insurance from 2013 to 2017.
Using Health Care Cost Institute data from 3 national payers, representing 26% of US individuals covered by employer-sponsored insurance, 3 annual measures were calculated. First, the share of total spending on services rendered by primary care clinicians (PCCs) was calculated (broad definition). Similar to previous studies,1,2 PCCs included family practice, geriatric medicine, gynecology, internal medicine, or pediatric physicians; physician assistants; or nurse practitioners on more than 50% of professional claims. Hospitalists were excluded. Second, the share of spending on primary care services rendered by PCCs, defined by Current Procedural Terminology codes, including evaluation and management visits, vaccinations, care planning, and other related services, was calculated (narrow definition). Third, utilization was calculated as the share of individuals who received at least 1 service from a PCC. The measures were assessed in the overall sample and in subgroups by age. Spending was defined as the total amount paid by the insurer and individual. Individuals with 12 months of medical and prescription drug coverage and positive total spending, including medical care and prescription drugs, in a calendar year were included. Spending was inflation-adjusted to 2017 US dollars using the Consumer Price Index.
To determine differences between 2013 and 2017, Wilcoxon signed-rank tests and logistic regressions were calculated for the share of PCC spending and utilization, respectively. Statistical significance was defined as a 2-sided P < .05. Analyses were conducted using SAS, version 9.4 (SAS Institute).
Under the broad definition, mean primary care spending increased from $511 among 11 406 520 individuals in 2013 to $538 among 11 608 038 individuals in 2017 (Table 1), but declined as a share of total spending from 8.97% to 8.04% (difference, −0.93% [95% CI, −0.95% to −0.91%]; P < .001) (Table 2). Mean total spending increased from $5701 to $6688. Children had the highest primary care spending as a share of their total health care spending, with 20.33% in 2013 and 19.54% in 2017 (P < .001), and individuals aged 55 to 64 years had the lowest, with 7.25% in 2013 and 6.33% in 2017 (P < .001).
Under the narrow definition, the primary care spending share declined from 4.60% to 4.35% (difference, −0.25% [95% CI, −0.27% to −0.23%]; P < .001). This decline was accounted for by children, for whom the share decreased from 13.68% to 12.51% (P < .001). The primary care spending share under this definition did not change substantially for any other age group. The share of individuals utilizing a PCC increased from 78.35% in 2013 to 79.65% in 2017 (difference, 1.30% [95% CI, 1.27%-1.34%]; P < .001) and varied across age groups (Table 2).
From 2013 to 2017, the share of total spending attributed to primary care declined among individuals covered by employer-sponsored insurance despite an increase in PCC utilization and spending on primary care because total spending grew more quickly. Primary care may be both a substitute for and complement to non–primary care services.3 As a substitute, primary care may decrease spending for specialty and inpatient care, where services are more expensive.4 As a complement, it may increase utilization of more expensive care because patients are referred to a broader network of clinicians.5 A better understanding of the relationship between primary care and specialty utilization and spending is needed.
The estimates of primary care spending share are higher than estimates among Medicare fee-for-service beneficiaries,1 and fall between other estimates of individuals covered by employer-sponsored insurance using a convenience sample of insurers2 and Medical Expenditure Panel data.6
Factors affecting primary care spending, such as patient and PCC demographics and insurance plan benefit design, were not studied. The data may not be representative of the entire employer-sponsored insurance population.
Accepted for Publication: September 12, 2019.
Corresponding Author: Julie Reiff, BA, Health Care Cost Institute, 1100 G St NW, Ste 600, Washington, DC 20005 (firstname.lastname@example.org).
Author Contributions: Ms Reiff and Dr Biniek had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: All authors.
Acquisition, analysis, or interpretation of data: Reiff, Fuglesten Biniek.
Drafting of the manuscript: Reiff, Fuglesten Biniek.
Critical revision of the manuscript for important intellectual content: Brennan, Fuglesten Biniek.
Statistical analysis: Reiff.
Administrative, technical, or material support: Reiff, Brennan.
Supervision: Brennan, Fuglesten Biniek.
Conflict of Interest Disclosures: None reported.
Funding/Support: Funding was contributed to the Health Care Cost Institute by Aetna, UnitedHealthcare, Humana, and Kaiser Permanente.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Information: Data were contributed by Aetna, UnitedHealthcare, and Humana.
Create a personal account or sign in to: