VOL. 7, NO. 3 THE AMERICAN JOURNAL OF MANAGED CARE 261 . . . COSTS OF ILLNESS . . . Specialty of Principal Care Physician and Medicare Expenditures in Patients with Coronary Artery Disease: Impact of Comorbidity and Severity Robert L. McNamara, MD, MHS; Neil R. Powe, MD, MPH, MBA; David R. Thiemann, MD; Thomas Shaffer, MHS; Wendy Weller, MHS; and Gerard Anderson, PhD M anaged care organizations often use general- ists as gatekeepers in an attempt to control costs by limiting access to specialists. Although this system often conflicts with a patient’s desire to choose his or her own physicians 1-3 and with increasing evidence of better outcomes with specialist care in certain patient groups, 4-6 there is a perception that specialists provide care that is more resource intensive and thus more expensive than care provided by generalists. However, evidence about the actual cost of care is mixed. 7-9 To determine whether specialty care for elderly patients with coronary artery disease (CAD) is more resource intensive for fee-for-service patients, we examined Medicare expenditures. We divided the patients into a cardiologist group and a generalist group (incorporating general practitioners, family physicians, and internists) based on the specialty with the most outpatient claims for an individual patient. To examine the potential influence of patient mix, we evaluated expenditures stratified in 2 ways: by comorbidity from illness other than coro- Objective: To explore differences in expenditures for elder- ly patients with acute and chronic coronary artery disease according to the specialty of the principal care physician. Study Design: Retrospective analysis of Medicare claims. Patients and Methods: A total of 250,514 patients with coronary artery disease (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 410-414) were drawn from a national random sample of 1992 Medicare expenditures. Patients were classified by the physician type with the highest number of Medicare Part B outpatient claims into a cardiologist group and a generalist group. The outcome was mean total expenditures, stratifying (1) by comorbidity as measured by the modified Charlson Index and (2) by severity defined as the proportion of patients with acute myocardial infarction or unstable angina. Results: Those patients in the cardiologist group had lower comorbidity and higher severity than those in the generalist group. Overall mean expenditures were significantly higher for the cardiologist group than for the generalist group ($7658 vs $6047; P < .001). These differences in mean expenditures were evident at each level of comorbidity. However, when stratified by severity of diagnosis, differences were seen pre- dominantly in those with acute diagnoses. For those with either acute myocardial infarction or unstable angina, the mean expenditures were higher for the cardiologist group than for the combined generalist group ($15,378 vs $12,260; P < .001); however, the mean expenditures for those with only chronic conditions were similar ($4856 vs $4745; P = .53). Conclusion: Expenditures were higher when cardiologists were the principal care physicians treating patients with acute disease but not chronic disease. (Am J Manag Care 2001;7:261-266) From the Department of Epidemiology (RLM, NRP), the Department of Medicine (NRP, DRT, GA), and the Department of Health Policy and Management (NRP, WW, GA), The Johns Hopkins Medical Institutions, Baltimore, MD; the Chinle Comprehensive Health Care Facility, Indian Health Service, Chinle, AZ (RLM); and the Agency for Healthcare Research and Quality, Rockville, MD (TS). This project was supported by the Commonwealth Fund grant #96169. An abstract of this work was presented at the 47th Annual Scientific Session of the American College of Cardiology, Atlanta, GA, March 29-April 2, 1998. Address correspondence to: Robert L. McNamara, MD, MHS, Chinle Comprehensive Health Care Facility, PO Box PH, Chinle, AZ 86503. E-mail: rmcnamar@jhsph.edu.