This study evaluates whether a hospital’s modified early warning score (Chonnam Modified Early Warning Score, C-MEWS) accurately predicts various clinical outcomes.
It was conducted to compare and analyze the validity of the previously well-known sc...
This study evaluates whether a hospital’s modified early warning score (Chonnam Modified Early Warning Score, C-MEWS) accurately predicts various clinical outcomes.
It was conducted to compare and analyze the validity of the previously well-known scoring indices, such as the National Early Warning Score (NEWS) and Modified Early Warning Score (MEWS), and propose an early warning scoring index with the highest predictive power.
From September to December 2021, data was collected on 464 patients, aged 18 or older, who were admitted to the cardiology department at C University Hospital, located in Gwangju Metropolitan City. As clinical outcome indicators, ICU admission, CPR occurrence, in-hospital mortality, and any poor clinical outcome were included, and the accuracy (sensitivity, specificity, positive predictive value, negative predictive value, and Youden index) of each of the three tools was identified. In addition, the area under the receiver operating characteristic curve (AUCs) for poor clinical outcomes of the three tools was compared, and the most accurate predictive instrument and cut-off value were determined.
The results of this study are as follows:
1. According to NEWS, MEWS, and C-MEWS, 14.7%, 7.5%, and 30% of the 464 patients included in this study were classified as high-risk, respectively.
2. For ICU admission, NEWS showed the highest sensitivity with 78.8%, a specificity of 90.3%, and a Youden index of 0.69; C-MEWS performed next best with a sensitivity of 81.8%, a specificity of 74.0%, and a Youden index of 0.55; and MEWS exhibited a sensitivity of 54.5%, a specificity of 96.1%, and the lowest Youden index with 0.50.
3. NEWS showed a sensitivity of 45.5%, a specificity of 86.1%, and a Youden index of 0.31 for predicting CPR occurrence; C-MEWS was the most accurate with a sensitivity of 63.6%, a specificity of 70.9%, and a Youden index of 0.34; MEWS was the least accurate with a sensitivity of 36.4%, a specificity of 93.2%, and a Youden index of 0.29.
4. For in-hospital deaths, NEWS demonstrated the best predictability with a 68.6% sensitivity, 89.7% specificity, and a Youden index of 0.57; C-MEWS showed 80% sensitivity, 74.1% specificity, and a Youden index of 0.54, and MEWS showed the least accuracy with 54.3% sensitivity, 96.3% specificity, and a Youden index of 0.50.
5. For clinical deterioration involving any of the three indicators(intensive care unit admission, CPR, and in-hospital mortality), NEWS had the highest sensitivity of 77.6%, a specificity of 92.8%, and a Youden index of 0.70; C-MEWS showed a sensitivity of 83.7%, a specificity of 76.4%, and a Youden index of 0.60; and MEWS showed the least predictive ability with 53.1% sensitivity , 97.8% specificity, and a Youden index of 0.50.
6. The AUC of the NEWS tool for predicting clinical outcomes was .899(.847 to .951), and the highest cut-off point was 6.5, with a Youden index of 0.704.
In conclusion, based on the validation of the three scoring indices for predicting various clinical outcomes in patients with heart disease, C-MEWS was determined to be a tool with high sensitivity but low specificity and positive predictive value, whereas the NEWS tool best predicted the clinical outcomes. For the efficient operation of the rapid response team in the future, it is necessary to accurately predict poor clinical outcomes by applying a tool with high sensitivity and specificity.
In conclusion, as a result of a validity study on various clinical outcomes in patients with heart disease, C-MEWS was found to be a tool with high sensitivity but low specificity and positive predictive value, and the tool that best predicted poor clinical outcomes was the NEWS score of 7. For the efficient operation of the rapid response team in the future, it is necessary to evaluate the predictability by applying a tool with high sensitivity and specificity.