There are many factors hypothesized as contributing to overuse of percutaneous coronary intervention (PCI) in the United States, including financial ties between physicians and hospitals, but empirical data are lacking. We examined PCI indications in not-for-profit (NFP), major teaching, for-profit (FP), and physician-owned specialty hospitals.

A retrospective cohort study of 1,113,554 patients who underwent PCI in 694 hospitals (NFP 471, teaching 131, FP 79, specialty 13) participating in the CathPCI Registry® between January 1, 2004, and December 31, 2007. Percutaneous coronary intervention indications derived from American College of Cardiology Guidelines were classified as survival benefit (patients with primary reperfusion for ST-elevation myocardial infarction), potential quality of life benefit (patients with non–ST-elevation myocardial infarction, acute coronary syndrome (ACS), positive stress test, or chest pain), or unclear indications (patients receiving PCI without an obvious potential survival or quality of life benefit).

The percentage of PCI performed for unclear indications was somewhat higher for specialty hospitals (5.1% of all procedures) as compared with other hospital categories (FP 4.7%, NFP 4.2%, major teaching 4.5%; P < .001). Overall, 17% of hospitals had ≥20% of their total PCI procedures performed for unclear indications, but the proportion of FP, NFP, major teaching, and specialty hospitals reaching this threshold was not statistically different (20%, 16%, 17%, and 15%, respectively; P = .84).

A small proportion of PCI procedures were performed in patients with unclear indications, but there was wide variation across hospitals. On average, specialty hospitals performed more PCIs for unclear indications. Efforts to reduce variability should be pursued.