Outcome and utilization differences for older persons with stroke in HMO and fee-for-service systems.

OBJECTIVES: To compare treatment and outcomes for older persons with stroke in Medicare health maintenance organizations (HMOs) and fee-for-service (FFS) systems. DESIGN: Inception cohort stratified by payer and followed for 1 year. SETTING: Six HMOs and five FFS systems with large Medicare populations in the West, Midwest, and Eastern United States. PARTICIPANTS: A total of 429 randomly selected stroke patients receiving rehabilitation in nursing homes or rehabilitation hospitals (RHs) from June 1993 to June 1995. MEASUREMENTS: Improvement in activities of daily living (ADLs) during rehabilitation, and ADL recovery, community residence, and utilization until 12 months after stroke. Outcomes were adjusted for premorbid function, marital status, comorbid illness, posthospital function, cognition, psychological problems, and stroke deficits. RESULTS: At baseline, HMO patients were more likely to be married, and less likely to be blind or have psychiatric diagnoses. HMO patients had shorter hospitalizations (P < .001), were less likely to be admitted to RHs (13% vs 85%, P < .001), and received fewer therapy and physician specialist visits (P < .001) but more home health visits (P < .001). During rehabilitation, FFS patients made greater improvement in ADLs (difference, 0.73 ADLs; 95% CI, .37-1.09). At 1 year, there was no difference in ADL recovery (difference, −0.24 ADL; 95% CI, −0.64-0.16), but FFS patients were more likely to reside in the community (adjusted OR, 1.8; 95% CI, 1.1–3.1), and HMO patients were more likely to reside in nursing homes (adjusted OR, 2.4; 95% CI, 1.1–5.5). CONCLUSION: Study findings suggest that short-term functional outcomes and eventual community residence rates are poorer for Medicare HMO patients with stroke than for stroke patients receiving FFS care, consistent with the lower intensity of rehabilitation (in nursing homes vs RHs) and less specialty physician care. J Am Geriatr Soc 48:726–734, 2000.