Introduction
The Simplified Acute Physiology Score (SAPS) III has been validated as the Korea SAPSIII model in Korea. However, the Acute Physiology and Chronic Health Evaluation (APACHE) IV model has not yet been validated in Korea provincial intensive care units (ICUs).
Objectives
The aim of this study was to compare the ability of the APACHE IV with Korean SAPS 3 in predicting hospital mortality in a provincial ICU population.
Methods
The study was designed as a prospective cohort study for patients admitted to the nine intensive care units in the four provincial academic medical centers from September 2013 to February 2014. Validation and comparison were conducted to measure discrimination and calibration with using the area under the receiver operating characteristic curve (AUC) and the Hosmer-Lemeshow test, respectively.
Results
Among 733 enrolled patients, 34.4% (252/733) were surgical patients. The hospital mortality was 23.5%. The median APACHE IV score was 71(Standard deviation ± 33.8), and the predicted death rate was 25.6% respectively. The observed hospital mortality was 23.5%. The discriminative powers of two models were similar. The AUCs were 0.753 (95% confidence intervals (CI) : 0.711-0.794) for APAPCH IV and 0.768 (95% CI : 0.727-0.809) for Korean SAPS III respectively. Hosmer-Lemeshow C and H statics showed good calibration for both models, (H = 8.69, p = 0.370; C = 128.17, p = 0.824 for APACHE IV, and H = 8.40, p = 0.396; C = 178.98, p < 0.001 for Korean SAPS III respectively).
Conclusions
For Korea regional ICU patients, the APACHE IV and Korea SAPS III showed good discrimination and good calibration for hospital mortality. Therefore, the APACHE IV prognostic model might be applied to predict mortality in Korea regional ICUs.
References
- 1.So Yeon Lim, et al. Validation of SAPS3 Admission Score and Its Customization for Use in Korean ICU Patients: A Prospective Multicenter Study. Respirology. 2013;18(6):989–995. doi: 10.1111/resp.12115. [DOI] [PubMed] [Google Scholar]
- 2.Jae Yeol Kim, et al. External Validation of the Acute Physiology and Chronic Health Evaluation II in Korean Intensive Care Units. Yonsei Med J. 2013;54(2):425–431. doi: 10.3349/ymj.2013.54.2.425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hosmer DW, Lemeshow S. Confidence interval estimates of an index of quality performance based on logistic regression models. Stat Med. 1995;14:2161–2172. doi: 10.1002/sim.4780141909. [DOI] [PubMed] [Google Scholar]
- 4.Hannah Lee, et al. Validation of the APACHE IV model and its comparison with the APACHE II, SAPS 3, and Korean SAPS 3 models for the prediction of hospital mortality in a Korean surgical intensive care unit. Korean J Anesthesiol. 2014;67:115–122. doi: 10.4097/kjae.2014.67.2.115. [DOI] [PMC free article] [PubMed] [Google Scholar]