Admission Data Predict High Hospital Readmission Risk.

Area of Science:

• Healthcare Management
• Clinical Informatics
• Patient Outcomes

Background:

• Hospital readmissions represent a significant burden on healthcare systems.
• Predicting readmission risk is crucial for resource allocation and patient care optimization.

Purpose of the Study:

• To identify key data available at hospital admission that accurately predict the risk of 30-day readmission.
• To develop a predictive model for hospital readmissions using readily available admission data.

Main Methods:

• Retrospective analysis of 958 adult, nonpregnant patients admitted to a Family Medicine Service.
• Data abstracted from administrative sources and electronic medical records.
• Logistic regression analysis to identify predictors of 30-day readmission, including polypharmacy and Charlson comorbidity index.

Main Results:

• Patients had a 14% readmission risk; polypharmacy and high Charlson scores (≥5) significantly increased this risk.
• A logistic model incorporating both factors showed an odds ratio of 1.7 for high Charlson scores and 2.1 for polypharmacy.
• The model accurately predicted readmission risk, with an area under the receiver operating characteristics curve of 85%.

Conclusions:

• Polypharmacy and a high Charlson score at admission are strong predictors of hospital readmission.
• This predictive model can help conserve resources and target interventions for high-risk patients.
• The findings suggest a more efficient approach to managing readmission risk compared to existing models.