Report refutes primary assumption
It sounds simple: to improve health outcomes, increase the number of primary-care doctors and make sure people have access to them. But it turns out there's not a simple relationship between the number of primary-care providers in a region, the use of primary care by patients, and the health outcomes of those patients, according to a recent report from the Dartmouth Atlas Project.
Level: A compilation of findings already published in peer-reviewed journals, the report used Medicare data from 2003 to 2007 to analyze regional and racial variations in primary-care supply, use, and quality. The number of primary-care providers in a community probably does influence outcomes at a "very fine, local level," says lead author David Goodman, M.D. But at the regional level, where most workforce policies are set, he found no correlation.
The report drew several surprising conclusions. First, the overall supply of primary-care clinicians in a region is not related to how frequently Medicare beneficiaries use primary care. For example, in some regions, the supply of primary-care physicians was low, yet a relatively high proportion of beneficiaries had at least one visit per year with a primary-care provider. But in some regions with a higher supply of providers, fewer beneficiaries had a primary-care visit.
Blacks were less likely than whites to see a primary-care clinician.
Second, access to a primary-care clinician (as measured by having at least one annual visit) is by itself no guarantee that patients will receive recommended care. For example, there was no relationship between the percentage of beneficiaries in a region having at least one annual visit and several indicators of quality care—such as the percentage of diabetic beneficiaries receiving an annual eye exam, the rate of leg amputations among diabetic beneficiaries, and the rate of hospitalizations that result from poorly controlled chronic conditions.
Care: Third, in general within a region, blacks were much less likely than whites to see a primary-care clinician and much more likely to be hospitalized. Given previous research on racial disparities in care, that may not be too surprising. More surprising was that where beneficiaries lived had more influence on their health outcomes than the color of their skin. Differences among regions were much greater than differences between blacks and whites. For example, blacks were four times as likely as whites to have a leg amputated as a result of poorly controlled diabetes or peripheral vascular disease, but there was a tenfold regional variation in leg amputations.
These findings complicate the argument that simply producing more primary-care physicians will improve the nation's health. Primary-care supply can be "influential at a very local level," says Goodman. But "physicians don't tend to settle where needs are greater," he adds. "What doctors and nurses do . . . makes a huge difference, but the sheer number of physicians is not very powerful."
If you'd like to offer feedback about this article, we'd welcome getting your comments at DartMed@Dartmouth.edu.
This article may not be reproduced or reposted without permission. To inquire about permission, contact DartMed@Dartmouth.edu.