Key performance indicators in intensive care medicine. A retrospective matched cohort study.

Expert panel consensus was used to develop evidence-based process indicators that were independent risk factors for the main clinical outcome parameters of length of stay in the intensive care unit (ICU) and mortality. In a retrospective, matched data analysis of patients from five ICUs at a tertiary university hospital, agreed process indicators (sedation monitoring, pain monitoring, mean arterial pressure [MAP] >or= 60 mmHg, tidal volume [TV] <or= 6 ml/kg body weight, peak inspiratory pressure [PIP] <or= 35 cmH(2)O and blood glucose [BG] >or= 80 and <or= 130 mg/dl) were validated using a prospective dataset of 4445 consecutive patients. After matching for age, sex and ICU, 634 patients were analysed. Logistic regression of the 634 patients showed that monitoring analgesia and sedation, MAP >or= 60 mmHg and BG >or= 80 mg/dl were relevant for survival. Linear regression of the 634 patients showed that analgesia monitoring, PIP <or= 35 cmH(2)O and TV <or= 6 ml/kg were associated with reduced length of ICU stay. Linear regression on all 4445 patients showed analgesia, sedation monitoring, MAP >or= 60 mmHg, BG >or= 80 mg/dl and <or= 130 mg/dl, PIP <or= 35 cmH(2)O and TV <or= 6 ml/kg were associated with reduced length of ICU stay, indicating that adherence to evidence-based key process indicators may reduce mortality and length of ICU stay.