The objective of this study was to test two hypotheses. First, six-month revascularization rates among acute myocardial infarction patients would be higher if patients were admitted to a university hospital, and second, the rates would be higher if patients were admitted to a hospital with the capacity to perform invasive procedures.

Data for this study were derived from the Hospital Medical Records Institute in Ontario. All patients admitted to a hospital from April 1, 1991 through September 30, 1991 with a diagnosis of myocardial infarction were identified. Inclusion criteria were: Ontario residence, age greater than 35 years, discharged home alive with length of stay greater than 4 days, and index admission not resulting from a hospital transfer. Among those included, patients who received coronary angioplasty and/or coronary artery bypass grafting (CABG) surgery within 6 months were identified. Each admission was evaluated rather than each patient, thus duplicate admissions for the same patients were counted more than once. Since patients were generally readmitted to the same facility, the authors concluded this would not bias the results. Covariates included were age, gender, acute myocardial infarction complication, distance to nearest catheterization facility, hospital size (number of beds), length of stay in the hospital and comorbidity.

Overall, more patients (79.7%) were treated in non-teaching hospitals than teaching hospitals, and more patients (87.4%) were treated in hospitals without rather than with revascularization facilities. The study hypotheses were supported. That is, patients treated in teaching hospitals were significantly more likely to receive revascularization, even if the hospital did not have revascularization facilities, compared with patients treated in non-teaching hospitals with similar treatment facilities. Also, patients treated in hospitals with revascularization facilities were more likely to undergo these invasive procedures than patients treated in other hospitals, regardless of hospital's teaching status. These patterns were the same even after adjusting for demographic and clinical variables, except the availability of revascularization facilities no longer resulted in statistically significant differences in the odds of invasive procedures for patients treated in community hospitals.

Because the more liberal use of revascularization procedures contrasts the conservative use of other types of invasive procedures in teaching hospitals, the authors suggest as an explanation that teaching hospitals have two competing roles: one as a site for critical thinking and evidence-based practice, leading to skepticism about technology and invasive management, and another as a site for innovation and adoption of cutting-edge technology.

The study was unable to identify reasons for these differences, however it is possible that the clinical differences in patients referred to teaching versus non-teaching hospitals were not fully taken into account in these analyses.