IS in Healthcare
Loading...
Paper Type
Short
Paper Number
2398
Description
Due to the importance of intensive care units (ICUs) in hospitals, many researchers have developed new predictive models to assess the patients’ illness severity and estimate their mortality in ICUs. In this study, we assess the predictive power of the Acute Physiology and Chronic Health Evaluation IV (APACHE IV) used in the eICU database. Based on the performed explanatory analysis, we found that APACHE IV is not predictive for all diseases. Receiver operating characteristic (ROC) curve analysis showed that, in some diseases, the AUC is as low as 0.75. Moreover, by showing the insufficiency of 0.5 as a threshold, we estimated an optimal threshold for predicted mortality for each disease. Based on the new threshold, the accuracy of APACHE IV was low (approximately 0.7). We further examined the role of APACHE score in predicting the length of stay and ventilation days by calculating the Pearson correlation coefficients (PCCs).
Recommended Citation
Rajabalizadeh, Atefeh; Wang, Shaodong; Javadi, Mahdi; Safaei, Nima; Talafidaryani, Mojtaba; Li, Qing; Zhang, Wenli; and Moqri, Mahdi, "In-depth Evaluation of APACHE Scoring System Using eICU Database" (2020). ICIS 2020 Proceedings. 17.
https://aisel.aisnet.org/icis2020/is_health/is_health/17
In-depth Evaluation of APACHE Scoring System Using eICU Database
Due to the importance of intensive care units (ICUs) in hospitals, many researchers have developed new predictive models to assess the patients’ illness severity and estimate their mortality in ICUs. In this study, we assess the predictive power of the Acute Physiology and Chronic Health Evaluation IV (APACHE IV) used in the eICU database. Based on the performed explanatory analysis, we found that APACHE IV is not predictive for all diseases. Receiver operating characteristic (ROC) curve analysis showed that, in some diseases, the AUC is as low as 0.75. Moreover, by showing the insufficiency of 0.5 as a threshold, we estimated an optimal threshold for predicted mortality for each disease. Based on the new threshold, the accuracy of APACHE IV was low (approximately 0.7). We further examined the role of APACHE score in predicting the length of stay and ventilation days by calculating the Pearson correlation coefficients (PCCs).
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
5-Health