Readmission Risk Prediction Model for Patients with Chronic Heart Failure: A Systematic Review and Meta-Analysis

Area of Science:

• Cardiology
• Medical Informatics
• Health Services Research

Background:

• Heart failure (CHF) readmissions pose a significant clinical and economic burden.
• Accurate risk prediction models are crucial for managing CHF patients and reducing readmissions.
• Existing models require systematic evaluation to assess their predictive performance.

Purpose of the Study:

• To systematically review and meta-analyze existing risk prediction models for CHF patient readmission.
• To identify the most effective predictors for CHF readmission risk.

Main Methods:

• A comprehensive literature search was conducted across multiple databases (PubMed, Embase, Web of Science, etc.) up to July 2021.
• Nine studies (2 prospective, 7 retrospective) met the selection criteria.
• Study quality was assessed using the CHARMS checklist, and prediction models were systematically evaluated.

Main Results:

• All evaluated models demonstrated predictive capability with an area under the receiver operating characteristic curve > 0.63.
• Commonly identified predictors included B-type natriuretic peptide (BNP) or N-terminal pro-brain BNP (OR 4.35), renal insufficiency (OR 1.60), comorbidities, and prior hospitalization history.
• The predictive accuracy of models was enhanced by non-parametric statistical methods and large electronic datasets.

Conclusions:

• Risk prediction models for CHF readmission show promise, with specific biomarkers and clinical factors being significant predictors.
• Calibration and external validation are essential for clinical implementation of these models.
• Combining parametric and non-parametric methods may lead to more precise predictive models for clinical use.