The effect of nonteaching services on the distribution of inpatient cases for internal medicine residents.

PURPOSE: Nonteaching services (NTSs) are becoming increasingly prevalent in academic hospitals. This study was designed to determine whether the presence of an NTS is associated with higher acuity and altered case mix on the teaching service.

METHOD: The authors carried out a retrospective, cross-sectional analysis of data about all general medical admissions between January 1, 2005 and June 30, 2005 to either of two teaching hospitals in Rochester, New York. A total of 6,907 inpatients were studied, of whom 1,976 (29%) were admitted to medicine resident services and 4,931 (71%) were admitted to NTSs. Hospital billing databases were used to determine patient demographics, ICD-9 diagnoses, Charlson Comorbidity Index scores, and patient disposition.

RESULTS

Compared with NTS patients, patients on resident services had higher median Charlson Comorbidity Index scores (3.0 versus 2.0, P < .001) and numbers of comorbidities (9.0 versus 8.0, P < .001) and were more likely to require intensive care (15.5% versus 7.6%, P < .001) and to die in the hospital (8.2% versus 4.5%, P < .001). Patients on the resident services were more likely to have acute renal failure, respiratory failure, septicemia, and HIV. Residents were less likely to care for patients with primary diagnoses of chest pain, cellulitis, alcohol withdrawal, and sickle cell crisis. The differences in patients' conditions between resident services and NTSs were similar in the two hospitals and also among patients who had not received intensive care.

CONCLUSIONS

Patients on resident services may be more medically complex and more likely to have high-acuity diagnoses than patients on NTSs. How these differences affect residents' education, residents' career decisions, and practice styles deserves further study.