We studied 118 patients with prolonged critical illness to determine if there would be a difference in outcome between patients managed by university-based (UB) or community-based (CB) intensivists. Patients consecutively admitted to a long-term acute care hospital were assigned in an arbitrary manner to the UB service or CB service. Patient survival, length of stay, and success and length of time required for liberation from mechanical ventilation (MV) were compared using survival analysis, logistic regression, and analysis of variance techniques. Patients on the UB service were liberated from MV in 32% fewer days (39 versus 57 d, p = 0.02) and were marginally more likely to be liberated from MV (46% versus 30%, p = 0.14). UB physicians were more likely to write do not resuscitate orders (59% versus 33%, p < 0.01) and to withdraw life-sustaining therapy (12% versus 2%, p = 0.09). There were no detectable differences in survival between the two groups. Estimated reimbursement for CB physicians ($6,797/patient) was 46% greater than for UB physicians ($4,651/patient) for discharged patients (p = 0.03). We conclude that patients experiencing prolonged critical illness may experience different outcomes based on their physician provider. In our study, patients were liberated more quickly from MV, were withdrawn from life support more readily, and were managed at lower cost by UB intensivists than by CB intensivists.