Objective: To assess whether outcomes and costs differ for elderly patients admitted to rehabilitation hospitals, subacute nursing homes, and traditional nursing homes.
Design: Inception cohort stratified by provider type and followed prospectively for 6 months.
Setting: A total of 92 hospital-based units and freestanding facilities from 17 states.
Patients: A total of 518 randomly selected patients with hip fracture and 485 stroke patients admitted from November 1991 to February 1994.
Main outcome measures: At 6 months comparing community residence, recovery to premorbid levels in 5 activities of daily living (ADLs), Medicare costs, and the number of therapy and physician visits. Outcomes were adjusted for premorbid residence and function, caregiver availability, comorbid illness, admission function, cognition, depression, sensory deficits, and mobility impairments.
Results: On admission, rehabilitation hospital patients were more likely (P<.001) to have caregivers and better cognitive and physical function. Hip fracture patients admitted to rehabilitation hospitals did not differ from patients admitted to nursing homes in returning to the community (adjusted odds ratio [OR], 1.3; 95% confidence interval [CI], 0.6-2.6) or in the number of ADLs recovered to premorbid level (difference, 0.09 ADL; 95% CI, -0.27-0.44), but stroke patients admitted to rehabilitation hospitals were more likely to return to the community (adjusted OR, 3.3; 95% CI, 1.5-7.2) and recover ADLs (difference, 0.63 ADL; 95% CI, 0.20-1.07). Subacute nursing home patients with stroke were more likely than traditional nursing home patients to return to the community (adjusted OR, 6.8; 95% CI, 2.2-21.4), there was no difference in return to the community for patients with hip fracture (adjusted OR, 1.6; 95% CI, 0.7-3.6), and there were no differences in recovery of ADLs for either condition. Medicare costs were greater (P<.001) for rehabilitation hospital patients than for subacute nursing home patients, and the costs for subacute nursing home patients were greater (P=.03 for stroke and .009 for hip fracture) than for traditional nursing home patients.
Conclusions: Study findings are consistent with enhanced outcomes for elderly patients with stroke treated in rehabilitation hospitals but not for patients with hip fracture. Subacute nursing homes were more effective than traditional nursing homes in returning patients with stroke to the community, despite comparable functional outcomes.