A Comparison of Acute Ischemic Stroke Patients Discharged to Inpatient Rehabilitation vs a Skilled Nursing Facility: The Paul Coverdell National Acute Stroke Program.

OBJECTIVE: To compare the sociodemographic, clinical, and hospital related factors associated with discharge of acute ischemic stroke (AIS) survivors to inpatient rehabilitation (IRF) and skilled nursing facility (SNF) rehabilitation services. DESIGN: Retrospective descriptive study from the Paul Coverdell National Acute Stroke Program (PCNASP) participating hospitals during 2016 to 2019. SETTING: 9 Participating states from PCNASP in United States. PARTICIPANTS: 130,988 patients with AIS from 569 hospitals (N=337,857). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURE: Discharge to IRF and SNF. RESULTS: Patients discharged to a SNF had longer length of hospital stay, more comorbidities, and higher modified Rankin scores compared with patients discharged to an IRF. Nine characteristics were associated with being less likely to be discharged to an IRF than an SNF: older age (85+ years old, adjusted odds ratio [AOR]=0.20 [confidence interval [CI]=0.18-0.21]), identifying as non-Hispanic Black (AOR=0.85 [CI=0.81-0.89]), identifying as Hispanic (AOR=0.80 [CI=0.74-0.87]), having Medicaid or Medicare (AOR=0.73 [CI=0.70-0.77]), being able to ambulate with assistance from another person (AOR=0.93 [CI=0.89-0.97]), being unable to ambulate (AOR=0.73 [CI=0.62-0.87]) and having comorbidities, prior stroke (AOR=0.69 [CI=0.66-0.73]), diabetes (AOR=0.85 [CI=0.82-0.88]), and myocardial infraction or coronary artery disease (AOR=0.94 [CI=0.90-0.97]). Four characteristics were associated with being more likely to be discharged to an IRF than an SNF: being a man (AOR=1.20 [CI=1.16-1.24]), and having a slight disability (Rankin Score 2) (AOR=1.41 [CI=1.29-1.54]), being at larger hospitals (200-399 beds: AOR=1.31 [CI=1.23-1.40]; 400+ beds: AOR=1.29 [CI=1.20-1.38]), and being at a hospital with stroke unit (AOR=1.12 [CI=1.07-1.17]). CONCLUSION: This study found differences in demographic, clinical, and hospital characteristics of AIS patients discharged for rehabilitation to an IRF vs SNF. The characteristics of patients receiving rehabilitation services may be helpful for researchers and hospitals making policies related to stroke discharge and practices that optimize patient outcomes. Populations experiencing inequities in access to rehabilitation services should be identified, and those who qualify for rehabilitation in IRF should receive this care in preference to rehabilitation in SNF.