Cardiac specialty hospitals assert better patient outcomes and efficiency, whereas general hospitals contend they attract healthier patients.
To ascertain whether increased cardiac specialization of a hospital's services is associated with improved outcomes for coronary artery bypass graft (CABG) surgery and whether patients with and without comorbid disease have equivalent outcomes to their counterparts at general hospitals.
We conducted a retrospective cohort study of a 5% sample of Medicare beneficiaries undergoing CABG procedures from 2001 to 2003. Using multivariate analyses, we assessed the hospital degree of cardiac specialization on mortality, length of stay, hospital readmissions, and emergency room visits leading to rehospitalization for all patients and those with more comorbidities.
Patients at cardiac specialty hospitals had less comorbid disease (29.7% with Charlson scores ≥2) than those at moderately (37.2%) and least specialized hospitals (36.6%, P = 0.001). Overall, CABG outcomes in all 3 groups were similar for inpatient mortality (P = 0.78), 30-day postdischarge mortality (P = 0.69), emergency room visit leading to rehospitalization (P = 0.35), and hospital readmission within 30 days postdischarge (P = 0.70). However, for patients with greater comorbidity, 30-day postdischarge mortality was worse at cardiac specialty hospitals compared with least specialized hospitals from across the United States (adjusted odds ratio, 1.71; 95% confidence interval, 1.26–2.32; P = 0.001). Results were robust when hospitals were compared within metropolitan statistical areas.
Favorable patient selection may occur at cardiac specialty hospitals. Although healthier patients fare comparably across types of hospitals, patients with greater comorbid disease seem to experience worse 30-day postdischarge mortality at specialty hospitals.
From the *Robert Wood Johnson Clinical Scholars Program; †Department of Medicine, Johns Hopkins School of Medicine; ‡Department of Health Policy and Management; §Department of Biostatistics; and ¶Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland.
Supported by the Robert Wood Johnson Foundation under its Clinical Scholars Program.
Reprints: Constance W. Hwang, MD, MPH, C/O, Neil R. Powe, Welch Center for Prevention, Epidemiology and Clinical Research, Johns Hopkins Medical Institutions, 2024 E. Monument St., Suite 2-600, Baltimore, MD 21205. E-mail: email@example.com.