Related Experiment Video
Updated: Jun 13, 2026
09:10
Determining the Functional Status of the Corticospinal Tract Within One Week of Stroke
Published on: February 22, 2020
Predicting 30-Day Readmission After Stroke: A Systematic Review and Meta-Analysis to Inform Predictor Selection
Saurabh Kalra1,2, Farya Fakoori1, Mohammad Nafeli Shahrestani1
1Department of Neurology, University of Miami Miller School of Medicine, Miami, FL 33136, USA.
Diagnostics (Basel, Switzerland)
|June 12, 2026
Summary
Thirty-day stroke readmissions are common at 12.9%. Current prediction models show moderate accuracy, highlighting a need for improved stroke care transition and patient context factors in future models.
Area of Science:
- Neurology
- Public Health
- Health Services Research
Background:
- Thirty-day readmission after stroke is a significant concern, with limited recent synthesis of current rates, prediction model performance, and key predictor domains.
- Understanding readmission patterns is crucial for improving post-stroke care and patient outcomes.
Purpose of the Study:
- To comprehensively synthesize contemporary 30-day readmission rates after stroke.
- To evaluate the performance of existing prediction models for stroke readmissions.
- To identify and summarize predictor domains used in stroke readmission models.
Main Methods:
- A systematic literature search was conducted across major databases (PubMed, Embase, Web of Science, Scopus, Google Scholar) for studies published between January 2021 and October 2025, adhering to PRISMA guidelines.
- Random-effects meta-analysis was used to pool readmission proportions and model discrimination (Area Under the Receiver Operating Characteristic Curve - AUC).
- Heterogeneity was assessed using I-squared statistics, and predictors were categorized by domain.
Main Results:
- Twenty studies were included, with 15 contributing quantitative data from 358,434 patients.
- The pooled 30-day readmission proportion was 12.9% (all-cause: 14.2%; stroke-specific: 3.6%).
- Pooled model discrimination (AUC) was 0.69, indicating moderate predictive ability, with substantial heterogeneity (I² > 98%) across studies. Predictor selection was inconsistent and data-driven.
Conclusions:
- Thirty-day stroke readmissions remain a prevalent issue, with current predictive models demonstrating only modest discrimination.
- The development of high-performing, generalizable stroke readmission prediction models is still needed.
- Future models require broader predictors encompassing stroke severity, care transitions, follow-up, and patient context, alongside external validation and clinical integration.