Background: Stroke survivors have identified “home-time” (days spent at home alive and outside of a hospital) as a meaningful outcome. However, home-time has not been well studied as an outcome of interest outside of clinical trials. We aimed to evaluate home-time as a patient-centered outcome in Medicare beneficiaries with ischemic stroke in comparison with modified Rankin Score (mRS) at 90 days and 1 year post event.
Methods: Patients ≥65 years old enrolled in GWTG-Stroke and AVAIL (Adherence eValuation After Ischemic Stroke Longitudinal) registry were linked to Medicare claims to ascertain home-time, defined as time spent alive and out of a hospital or skilled nursing facility, at 90 days and 1 year after admission for ischemic stroke. The correlation of home-time with mRS at 90 days and 1 year from admission were evaluated graphically and by Pearson correlation coefficients. In addition, to assess the level of agreement between measures, the c-index of 90 day and 1 year home-time predicting 90 day and 1 year mRS 0-2 were determined.
Results: There were 815 ischemic stroke patients (age median 76 years [IQR 70-82], 46% female, NIHSS median 4 [IQR 2-7]) from 88 participating hospitals. The 90 day and 1 year median home-times were 79 days (IQR 52-86) and 349 days (IQR 303-360). The 90 day and 1 year median mRS were 2 (IQR 1-4) and 2 (IQR 1-4). The Pearson correlation coefficient between 90 day mRS and home-time within 90 days was -0.731 (p-value<.0001) (Figure). The c-index of 90 days home-time predicting 90 day mRS 0-2 was 0.837. The Pearson correlation coefficient between 1 year mRS and home-time within 1 year was -0.713 (p-value<.0001). The c-index of 1 year home-time predicting 12 month mRS 0-2 was 0.828.
Conclusions: In a population of older ischemic stroke patients, home-time was correlated with and showed a strong level of agreement with mRS at 90 days and 1 year. Home-time may be able to serve as a novel, patient-centered, outcome measure for hospital/transitional stroke care.