Geographic variation in resource use for coronary artery bypass surgery

OBJECTIVES. The purpose of this study-was to examine the national variability in patient-level cost and length of stay for coronary artery bypass grafting (CABG) in Medicare patients. METHODS. Retrospective multivariate regression analysis was done using Medicare administrative files and American Hospital Association files. Patients in the study had an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) procedure code for CABG, with accompanying 1990 procedure date, in the Medicare Provider Analysis and Review File (n=92,449). RESULTS. Outcome measures used were inpatient cost (exclusive of professional fees) and inpatient length of stay associated with bypass admission. The national average cost of bypass surgery was $22,847 (median $18,783), with an accompanying average length of stay of 16 days (median 13 days). Multivariate regression analysis revealed that patient-level cost and length of stay were related to clinical, demographic, hospital, and regional characteristics (R(2) = 25% and 16%, respectively). After accounting for these characteristics at the patient level, considerable variation among states persisted in both cost and length of stay. In addition, states with similar adjusted lengths of stay varied widely with respect to adjusted cost. No relation was found at the state level between level of resource use and either procedural mortality or 60-day mortality/readmission rates. CONCLUSIONS. Considerable variability exists among states in patient-level cost and length of stay for CABG surgery, after adjusting to the extent possible for clinical, demographic, hospital, and regional characteristics. The lack of association at the state level between resource use and rates of mortality and hospital readmission suggests that costs could be reduced in many areas of the United States without compromising quality of care.