Functional outcomes of posthospital care for stroke and hip fracture patients under Medicare

BACKGROUND: Medicare's introduction of the Prospective Payment System for hospitals has led to tremendous growth in ways of providing posthospital care. Despite substantial differences in costs per episode of care, the type of posthospital care that produces the best results for specific types of patients is not clear. This study analyzed the outcomes of different types of posthospital care for a cohort of older Medicare patients (who had diagnoses associated with the use of a range of posthospital care modalities) for up to a year after hospital discharge. METHODS: Medicare patients hospitalized with strokes and hip fractures were enrolled consecutively just before discharge from 52 hospitals in three cities in 1988-1989. These diagnosis-related groups were chosen because patients were discharged to all three major types of Medicare-supported posthospital care. Patients were interviewed in-person before discharge and again at 6 weeks, 6 months, and 1 year after discharge. The functional outcomes of posthospital care were evaluated by the instrumental variables estimation approach to correct for selection bias caused by nonrandom treatment assignment The impacts of discharge locations on the functional outcomes were examined by one-way analyses of variance (ANOVA). RESULTS: In general, the more disabled patients went to nursing homes and rehabilitation, but the overlap in distribution was sufficient to conduct the analyses. Stroke patients discharged to nursing homes had the highest mortality rate (P < .01). Stroke patients discharged to home health had the lowest rehospitalization rates (P < .05) Hip fracture discharged to home health care had the highest adjusted rehospitalization rate, whereas those discharged to nursing homes had the lowest adjusted rehospitalization rate (P < .05). For stroke patients, posthospital care in rehabilitation facilities or home health care was associated with significantly better functional improvement compared with stroke patients discharged elsewhere. However, functional outcomes deteriorated by 1 year posthospitalization among stroke patients who received their posthospital care at nursing homes or received no formal posthospital care. For hip fracture patients, all four types of posthospital care were associated with functional improvement, but patients discharged to rehabilitation facilities experienced the most functional improvement. CONCLUSIONS: The choice of posthospital care can influence the course of Medicare patients. Careful attention should be paid to how hospital discharge decisions are made and to the financial incentives for different types of posthospital care provided under the current payment system. The current supply of nursing homes is not well suited to the demands of posthospital care.