Comparison of a Frailty Risk Score and Comorbidity Indices for Hospital Readmission Using Electronic Health Record...

Area of Science:

• Gerontology
• Health Informatics
• Predictive Analytics

Background:

• Hospital readmissions among older adults (≥50 years) pose a significant healthcare challenge.
• Existing predictive models often utilize frailty risk scores (FRS) and comorbidity indices, but their comparative effectiveness requires further investigation.

Purpose of the Study:

• To evaluate the predictive performance of five FRS definitions and three comorbidity indices for 3-, 7-, and 30-day hospital readmissions.
• To identify an optimal combined model for FRS and comorbidity in predicting readmissions.

Main Methods:

• Retrospective analysis of electronic health records (EHRs) from 55,778 hospitalized adults aged ≥50 years.
• Multivariable logistic regression and area under the curve (AUC) analysis were employed to assess predictive accuracy.
• Evaluation of individual and combined FRS and comorbidity measures.

Main Results:

• Both FRS and comorbidity were independently associated with hospital readmission.
• Combined FRS and comorbidity models demonstrated strong predictive accuracy, with AUCs ranging from 0.75–0.77 for 30-day and 0.84–0.85 for 3-day readmissions.
• FRS measures showed a stronger association with 30-day readmissions compared to shorter intervals.

Conclusions:

• Frailty risk scores and comorbidity indices are valuable predictors of hospital readmission in older adults.
• Combined FRS and comorbidity models offer enhanced predictive capabilities for readmission risk.
• These findings support the integration of FRS and comorbidity data into EHR systems for improved patient management and readmission reduction strategies.