Zimmerman JE, Kramer AA, McNair DS, et al. Acute Physiology and Chronic Health Evaluation (APACHE) IV: Hospital mortality assessment for today’s critically ill patients

George Washington University, Washington, Washington, D.C., United States
Critical Care Medicine (Impact Factor: 6.31). 06/2006; 34(5):1297-310. DOI: 10.1097/01.CCM.0000215112.84523.F0
Source: PubMed


To improve the accuracy of the Acute Physiology and Chronic Health Evaluation (APACHE) method for predicting hospital mortality among critically ill adults and to evaluate changes in the accuracy of earlier APACHE models.
: Observational cohort study.
A total of 104 intensive care units (ICUs) in 45 U.S. hospitals.
A total of 131,618 consecutive ICU admissions during 2002 and 2003, of which 110,558 met inclusion criteria and had complete data.
We developed APACHE IV using ICU day 1 information and a multivariate logistic regression procedure to estimate the probability of hospital death for randomly selected patients who comprised 60% of the database. Predictor variables were similar to those in APACHE III, but new variables were added and different statistical modeling used. We assessed the accuracy of APACHE IV predictions by comparing observed and predicted hospital mortality for the excluded patients (validation set). We tested discrimination and used multiple tests of calibration in aggregate and for patient subgroups. APACHE IV had good discrimination (area under the receiver operating characteristic curve = 0.88) and calibration (Hosmer-Lemeshow C statistic = 16.9, p = .08). For 90% of 116 ICU admission diagnoses, the ratio of observed to predicted mortality was not significantly different from 1.0. We also used the validation data set to compare the accuracy of APACHE IV predictions to those using APACHE III versions developed 7 and 14 yrs previously. There was little change in discrimination, but aggregate mortality was systematically overestimated as model age increased. When examined across disease, predictive accuracy was maintained for some diagnoses but for others seemed to reflect changes in practice or therapy.
APACHE IV predictions of hospital mortality have good discrimination and calibration and should be useful for benchmarking performance in U.S. ICUs. The accuracy of predictive models is dynamic and should be periodically retested. When accuracy deteriorates they should be revised and updated.

Download full-text


Available from: Andrew A. Kramer
  • Source
    • ", Zimmerman JE 2006 [70] "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Previous studies have demonstrated that ICU (intensive care unit) survivors have decreased long-term survival rates compared to the general population. However, knowledge about how to identify ICU survivors with higher risk of death and the adjustable factors associated with mortality is still lacking. Methods and Design The FROG-ICU (the French and European Outcome Registry in Intensive Care Units) study is a prospective, observational, multicenter cohort study where ICU survivors are followed up to one year after ICU discharge. Beside one year survival, the study is designed to assess incidence and identifying risk factors for mortality over the year following discharge from the ICU. All consecutive patients admitted in ICU to the 28 participating centers during the study period will be included. Every subject will undergo an evaluation at admission, throughout the ICU stay and at ICU discharge. The global, especially cardiovascular, assessment of each subject will be performed through a complete clinical exam, instrumental tests (electrocardiogram, echocardiogram) and biological parameters. Blood and urine samples will be collected at admission and at discharge with the primary goal to assess effectiveness of routine and novel cardiovascular, inflammatory and renal biomarkers, with potential interest in risk stratification for patients who survive an ICU stay. The follow up will include a careful tracking of patients through telephone calls and questionnaires at 3, 6 and 12 months after ICU discharge. FROG-ICU aims to identify the clinical and biological phenotype of patients with different levels of probability of death in the year after ICU discharge. Discussion FROG-ICU has been designed to better understand long term outcome after ICU discharge as well as risk factors for all-cause and cardiovascular morbidity and associated mortality. It is a large prospective multicenter cohort with a biological (on plasma and urine) collection and one-year follow-up of ICU patients. FROG ICU will allow performing a risk stratification of ICU survivors as to recognize the subset of patients who may benefit from an early intervention to allow decreased cardiovascular morbidity and related mortality. Trial registration NCT01367093.
    Full-text · Article · Dec 2015 · BMC Anesthesiology
  • Source
    • "Multiple scoring systems, including the APS, APACHE II, and APACHE IV, have been used widely in clinical practice to predict outcome in ICU patients [21, 22]. We used the APS, APACHE II, and APACHE IV to evaluate prognoses and analyzed the correlation between these clinical variables and blood physicochemical parameters. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: To determine the influence of physicochemical parameters on survival in metabolic acidosis (MA) and acute kidney injury (AKI) patients. Materials and methods: Seventy-eight MA patients were collected and assigned to AKI or non-AKI group. We analyzed the physiochemical parameters on survival at 24 h, 72 h, 1 week, 1 month, and 3 months after AKI. Results: Mortality rate was higher in the AKI group. AKI group had higher anion gap (AG), strong ion gap (SIG), and apparent strong ion difference (SIDa) values than non-AKI group. SIG value was higher in the AKI survivors than nonsurvivors and this value was correlated serum creatinine, phosphate, albumin, and chloride levels. SIG and serum albumin are negatively correlated with Acute Physiology and Chronic Health Evaluation IV scores. AG was associated with mortality at 1 and 3 months post-AKI, whereas SIG value was associated with mortality at 24 h, 72 h, 1 week, 1 month, and 3 months post-AKI. Conclusions: Whether high or low SIG values correlate with mortality in MA patients with AKI depends on its correlation with serum creatinine, chloride, albumin, and phosphate (P) levels. AG predicts short-term mortality and SIG value predicts both short- and long-term mortality among MA patients with AKI.
    Full-text · Article · Aug 2014 · BioMed Research International
  • Source
    • "However, Zimmerman et al. [5] suggested that APACHE IV has better accuracy than the previous systems, and older models should not be used. The APACHE IV model showed good discrimination and calibration in the United States where the model was developed [5]. Outside the United States, recent studies have demonstrated that the discriminatory performance of APACHE IV was good [8,9,10,11,12]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background The Acute Physiology and Chronic Health Evaluation (APACHE) IV model has not yet been validated in Korea. The aim of this study was to compare the ability of the APACHE IV with those of APACHE II, Simplified Acute Physiology Score (SAPS) 3, and Korean SAPS 3 in predicting hospital mortality in a surgical intensive care unit (SICU) population. Methods We retrospectively reviewed electronic medical records for patients admitted to the SICU from March 2011 to February 2012 in a university hospital. Measurements of discrimination and calibration were performed using the area under the receiver operating characteristic curve (AUC) and the Hosmer-Lemeshow test, respectively. We calculated the standardized mortality ratio (SMR, actual mortality predicted mortality) for the four models. Results The study included 1,314 patients. The hospital mortality rate was 3.3%. The discriminative powers of all models were similar and very reliable. The AUCs were 0.80 for APACHE IV, 0.85 for APACHE II, 0.86 for SAPS 3, and 0.86 for Korean SAPS 3. Hosmer and Lemeshow C and H statistics showed poor calibration for all of the models (P < 0.05). The SMRs of APACHE IV, APACHE II, SAPS 3, and Korean SAPS 3 were 0.21, 0.11 0.23, 0.34, and 0.25, respectively. Conclusions The APACHE IV revealed good discrimination but poor calibration. The overall discrimination and calibration of APACHE IV were similar to those of APACHE II, SAPS 3, and Korean SAPS 3 in this study. A high level of customization is required to improve calibration in this study setting.
    Full-text · Article · Aug 2014 · Korean journal of anesthesiology
Show more