190 likes | 328 Views
Dichotomization of ICU length of stay based on model calibration. Marion Verduijn , Niels Peek, Frans Voorbraak, Evert de Jonge, Bas de Mol. ICU length of stay (LOS). Important outcome after cardiac surgery Predictive models for identification of high risk patients
E N D
Dichotomization of ICU length of stay based on model calibration Marion Verduijn, Niels Peek, Frans Voorbraak, Evert de Jonge, Bas de Mol
ICU length of stay (LOS) Important outcome after cardiac surgery Predictive models for • identification of high risk patients • case load planning and resource allocation
Main objective Development of a predictive model to estimate the risk of long ICU LOS using the method of class probability trees
Data 2063 patients (Academic Medical Center, Amsterdam, 1997-2001) • preoperative (e.g., age, gender) • operative (e.g., surgery type, duration) • first 24h ICU stay (e.g., blood pressure, temperature) (122 patients died: 5.2%)
Problem of outcome definition How should we define the outcome ‘long ICU LOS’? Literature: outcome dichotomized based on threshold values of 2-10 days without motivation or based on simple statistics
Objective of this study Selection of the threshold value to dichotomize ICU LOS in a structured fashion based on data analysis
Approach • Development of tree models for outcomes defined with different threshold values • Calculation of the model performance • Selection of the best model
First results † patients with ICU LOS higher than the threshold value of death ‡ determined using 10-fold cross-validation
ALOR distance Distance between two probabilities for a given x Absolute Log-Odds Ratio
Distances between probabilities II Property of ALOR: approximate proportional equivalence
MALOR statistic Distance measure for all elementsin F Mean value of the Absolute Log-Odds Ratio (MALOR) quantifies model calibration
Procedure of threshold selection 1) define a set of possible threshold values T 2) for all threshold values t in T do a) define the dichotomized outcome Yt using threshold t b) build a predictive model Mtfor outcome Yt c) compute DMALOR(Mt , Pt) 3) select threshold value with minimal MALOR statistic
Additional results † patients with ICU LOS higher than the threshold value of death ‡ determined using 10-fold cross-validation
Discussion and conclusions • Class probability trees to identify high risk groups • Performance measure should be insensitive to class unbalance when comparing models for different prediction problems