Mortality prediction ability of phycians in intensive care units of Turkey (MOPAP)

Abstract

Introduction: Intensive care physicians are increasingly involved in decision making about the prognosis of intensive care unit ICU patients. With this study; we aimed to evaluate the power of clinician foresight at prediction of mortality in patient at triage to intensive care and patient follow-up. Materials and Methods: This study was conducted in ICUs located in various geographical regions of Turkey between January 1, 2017-April 30, 2017. The clinical research was planned as observational, multicenter, cross-sectional. Results: A total of 1169 intubated patients were followed in 37 different ICU. At the beginning of the follow-up we asked the physician who will follow the patient in the ICU to give a score for the probability of survival of the patients. Scoring included a total of 6 scores from 0 to 5, with the "0" the worst probability "5" being the best. According to this distribution, only 1 (0.9%) of 113 patients who were given 0 points survived. Three (6.1%) of 49 with the best score of 5 died. Survival rates were significantly different in each score group (r: -0.488; p< 0.001). After the combined mortality estimation scores based on the clinical observations of the physicians (0 and 1 point score was combined as non-survive, 4 and 5 score was combined as survived) 320 of the 545 patients were estimated to be dead and 225 were predicted survival. Sensitivity and spesifity of scoring system to predict mortality was 91.56% (95% Cl: 87.96-94.37), 76.89% (95% CI: 70.82-82.23) respectively. Conclusion: In this study, we concluded that the physicians who follow the patients in the ICU can predict the poor prognosis at the time of admission and the high mortality rate. The physician's opinion on mortality estimation should be considered in intensive care mortality scoring in addition to other laboratory and clinical parameters.