Abstract Body: Introduction: Physical and occupational therapy are a standard of care for in-hospital stroke patients. The Activity Measure for Post-Acute Care (AMPAC) daily activity and basic mobility scores, a measure of functional ability, are used to assess the need for rehab time while hospitalized. Previous studies demonstrate that AMPAC predicts a patients discharge disposition (DD). This study aims to evaluate whether NIHSS, AMPAC or total rehab duration while hospitalized predicts DD or length of stay (LOS).
Methods: Data from adult patients who discharged from two medical centers in Portland, OR in 2019 or between January 2022 and May 2023 were abstracted from electronic medical records. Inclusion criteria was a diagnosis of ischemic stroke or intracerebral hemorrhage (ICH). Patients were excluded from the analysis if they:
Had a diagnosis of subarachnoid hemorrhage or traumatic ICH.
Died in-hospital or entered hospice within 7 days of arrival.
Had concurrent acute disease at arrival that was determined to impact rehab.
The primary outcomes were DD and LOS. We dichotomized DD as “good” (home, inpatient rehab) or “poor” (hospice, left against medical advice, acute care facility, intermediate care, and skilled nursing facility). Generalized linear models were used to test if cohort year, arrival NIHSS, 1^st AMPAC score, and rehabilitation time associated with DD and LOS. Two-tailed p value of less than 0.05 was considered statistically significant.
Results: More rehab minutes were documented for the 2019 cohort [178 (105, 320) vs. 115 (78, 195) Table 1]. We found no differences between NIHSS at arrival or 1^st AMPAC score. Our model showed that NIHSS at admit (OR = 0.94, CL: 0.90-0.99, Table 2) and 1^st AMPAC score (OR = 1.2, CL: 1.2-1.3, Table 2) were significant predictors of a good DD. 1^st AMPAC was also a predictor of shorter LOS (β = -3.8, CL: -5.8 to -1.9, Table 2).
Conclusions: AMPAC was the most predictive measure of a good discharge disposition. These findings highlight the continued importance of AMPAC in predicting discharge outcomes and suggest that variations in rehabilitation minutes and hospital stay do not substantially impact discharge disposition.