Development and Validation of a NEWS2-Enhanced Multivariable Prediction Model for Clinical Deterioration and...

Area of Science:

• Internal Medicine
• Clinical Risk Stratification
• Predictive Analytics

Background:

• Early identification of clinical deterioration is crucial for patient outcomes in general medicine.
• The National Early Warning Score 2 (NEWS2) aids in predicting patient worsening.
• Integrating NEWS2 with other clinical data can improve predictive accuracy.

Purpose of the Study:

• To develop and validate a predictive model for high clinical risk (HCR) and mortality.
• To assess the added value of combining NEWS2 with clinical and demographic data.
• To enhance early risk stratification for timely clinical decisions.

Main Methods:

• Retrospective cohort study of 2108 general medicine patients.
• Logistic regression models incorporating NEWS2, age, central venous catheter (CVC), and Barthel Index.
• Area under the ROC curve (AUC) used for model performance assessment.

Main Results:

• 29% of patients developed high clinical risk status.
• Older age, CVC presence, lower Barthel Index, and higher NEWS2 scores predicted HCR and mortality.
• The integrated model achieved AUC of 0.798 for HCR and 0.716 for mortality.

Conclusions:

• Combining NEWS2 with electronic health record data creates a more sophisticated risk stratification tool.
• This enhanced tool can support timely interventions and optimized therapeutic management.
• Prospective studies are needed to confirm the impact on patient outcomes and nurse workload.