Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2009 Feb 17;150(4):236-42.
doi: 10.7326/0003-4819-150-4-200902170-00004.

Primary Care Physicians' Links to Other Physicians Through Medicare Patients: The Scope of Care Coordination

Affiliations
Free PMC article

Primary Care Physicians' Links to Other Physicians Through Medicare Patients: The Scope of Care Coordination

Hoangmai H Pham et al. Ann Intern Med. .
Free PMC article

Abstract

Background: Primary care physicians are expected to coordinate care for their patients.

Objective: To assess the number of physician peers providing care to the Medicare patients of a primary care physician.

Design: Cross-sectional analysis of claims data.

Setting: Fee-for-service Medicare in 2005.

Participants: 2284 primary care physicians who responded to the 2004 to 2005 Community Tracking Study Physician Survey.

Measurements: Primary patients for each physician were defined as beneficiaries for whom the physician billed for more evaluation and management visits than any other physician in 2005. The number of physician peers for each physician was the sum of other unique physicians that the index physician's primary patients visited plus other unique physicians who served as the primary physician for each of the index physician's nonprimary patients during 2005.

Results: The typical primary care physician has 229 (interquartile range, 125 to 340) other physicians working in 117 (interquartile range, 66 to 175) practices with which care must be coordinated, equivalent to an additional 99 physicians and 53 practices for every 100 Medicare beneficiaries managed by the primary care physician. When only the 31% of a primary care physician's primary patients who had 4 or more chronic conditions was considered, the median number of peers involved was still substantial (86 physicians in 36 practices). The number of peers varied with geographic region, practice type, and reliance on Medicaid revenues.

Limitations: Estimates are based only on fee-for-service Medicare patients and physician peers, and the number of peers is therefore probably an underestimate. The modest response rate of the Community Tracking Study Physician Survey may bias results in unpredictable directions.

Conclusion: In caring for his or her own primary and nonprimary patients during a single year, each primary care physician potentially must coordinate with a large number of individual physician colleagues who also provide care to these patients.

Funding: National Institute on Aging, American Medical Group Association, and the Robert Wood Johnson Foundation.

Figures

Figure 1
Figure 1. Calculating the Number of Physician Peers
Solid arrows represent a physician being linked to his or her “medical home” patients based on having billed for the plurality of that patient's evaluation and management visits, while dashed arrows represent linkages to a physician's other, “non-medical home” patients. In this example, we calculated the number of peers for Physician A by counting Physicians B, C, and D each only once, and did not include Physician A in that count, resulting in a total of three peers.

Comment in

  • Primary care: too important to fail.
    Meyers DS, Clancy CM. Meyers DS, et al. Ann Intern Med. 2009 Feb 17;150(4):272-3. doi: 10.7326/0003-4819-150-4-200902170-00009. Ann Intern Med. 2009. PMID: 19221380 No abstract available.

Summary for patients in

Similar articles

See all similar articles

Cited by 65 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback