Do hospitalists or physicians with greater inpatient HIV experience improve HIV care in the era of highly active antiretroviral therapy? Results from a multicenter trial of academic hospitalists

Background. Little is known about the effect of provider type and experience on outcomes, resource use, and processes of care of hospitalized patients with human immunodeficiency virus (HIV) infection. Hospitalists are caring for this population with increasing frequency. Methods. Data from a natural experiment in which patients were assigned to physicians on the basis of call cycle was used to study the effects of provider type - that is, hospitalist versus nonhospitalist - and HIV-specific inpatient experience on resource use, outcomes, and selected measures of processes of care at 6 academic institutions. Administrative data, inpatient interviews, 30-day follow-up interviews, and the National Death Index were used to measure outcomes. Results. A total of 1207 patients were included in the analysis. There were few differences in resource use, outcomes, and processes of care by provider type and experience with HIV-infected inpatients. Patients who received hospitalist care demonstrated a trend toward increased length of hospital stay compared with patients who did not receive hospitalist care (6.0 days vs. 5.2 days; P = .13). Inpatient providers with moderate experience with HIV-infected patients were more likely to coordinate care with outpatient providers (odds ratio, 2.40; P = .05) than were those with the least experience with HIV-infected patients, but this pattern did not extend to providers with the highest level of experience. Conclusion. Provider type and attending physician experience with HIV-infected inpatients had minimal effect on the quality of care of HIV-infected inpatients. Approaches other than provider experience, such as the use of multidisciplinary inpatient teams, may be better targets for future studies of the outcomes, processes of care, and resource use of HIV-infected inpatients.