Predictors of mortality in the intensive care unit for adult patients admitted on mechanical ventilation: admission profile

Abstract

Evaluating risk factors for mortality in local populations such as adult patients admitted on mechanical ventilation in intensive care units (ICU) may provide support for the management and improvement of outcomes in these units. The inclusion of the workload of professionals in these models has offered a different view of predictors. The aim of this study was to evaluate whether Nursing workload assessed by the Nursing Activities Score (NAS), predictors of mortality (APACHEII and SAPS3) and some additional admission variables for patients admitted on mechanical ventilation in an ICU are predictors of death. We evaluated 194 patients who remained on mechanical ventilation for 48 hours before or after admission in one ICU, in a university hospital of high complexity. The clinical and socio-demographic profile, the NAS of admission and some admission variables were evaluated. The outcome discharge or death in the ICU was evaluated for all patients, and from simple or multiple logistic regression models, risk or protective factors for death in the ICU were obtained. Individually, only SAPS3 was significant for prediction of death (OR = 1.03; CI95%: 1.01; 1.05), while the APACHEII and the NAS of admission was not able to predict ICU mortality. In the multiple model, the only risk factors for ICU mortality were the presence of chronic obstructive pulmonary disease (OR = 8.82; CI95%: 1.82; 42.70), having thyroid diseases (OR = 5.98; CI95%: 1.15; 31.22) and the increase in the level of urea in the blood (OR = 1.01; CI95%: 1,002; 1.02). The admission variables of this population were more effective in predicting ICU mortality than the predictors of mortality evaluated here.