Early hospital readmission following stroke: the Florida Stroke Registry.

Background: Hospital readmission is an important indicator of poor transition of care post-stroke. Data on characteristics of patients at highest risk for readmission is limited and necessary to inform effective interventions. The goal is to identify risk factors at hospital discharge that predict 30-day readmission in the Florida Stroke Registry (FSR). Methods: The study population included 45,877 patients discharged home or to rehabilitation with an ischemic stroke or intracerebral hemorrhage in the FSR between 2017 and 2019. The FSR is a voluntary statewide registry of stroke patients from 167 hospitals using data from Get With the Guideline-Stroke. Readmissions were ascertained by propensity matching FSR with the Florida Agency for Healthcare Administration dataset, which includes all hospital admissions in Florida. The primary outcome was 30-day hospital readmission for any cause, and secondary outcomes were vascular-related and stroke readmissions specifically. Multivariable logistic regression models identified patient characteristics that independently predicted 30-day readmissions, including sociodemographics, stroke clinical characteristics, in-hospital treatment, medical history, discharge status, and hospital characteristics. Results: A hospital readmission within 30 days was experienced in 12% of cases; 6% had a vascular-related readmission, and 3% a recurrent stroke. The following characteristics were independently associated with an increased risk of all-cause readmission: Medicare or Medicaid insurance, large artery atherosclerosis as the stroke mechanism, increased stroke severity, diabetes, atrial fibrillation, peripheral vascular disease, coronary artery disease, prior stroke, chronic renal insufficiency, and depression. The following characteristics were independently associated with a decreased risk of all-cause readmission: ambulation, treated dyslipidemia, tPA treatment, discharge mRS 0-2, and treatment at a comprehensive stroke center. Conclusions: The risk of 30-day hospital readmission was substantial, modifiable, and impacted by insurance status, medical history, stroke etiology and severity, stroke care, and functional status at discharge. These findings can inform strategies to target high-risk patients who can benefit from interventions to improve transitions of care post-stroke.