How objective is the observed mortality following critical care?

Excerpt

Hospital mortality is frequently quoted as an outcome measure in intensive care medicine. It is valued by patients, families, health workers, policy makers and all of society, although long-term survival is probably the most important outcome for the patient and family. Hospital mortality per se is not an accurate reflection of ICU performance because ICUs admit heterogeneous groups of patients in terms of age, co-morbidities and acute health status, but it is one of the most readily available parameters. The development of mortality prediction models that can account for most of the previously mentioned confounders has led to the introduction of the standardised mortality ratio (SMR), which is the ratio between observed and predicted deaths of ICU admitted patients in a given period of time [1]. Not surprisingly, the report as well as analysis of SMR is one of the indicators identified by the Task Force on Safety and Quality of the European Society of Intensive Care Medicine [2]. The use of SMR requires on one hand an appropriately calibrated severity of illness score to predict deaths and on the other the in-hospital mortality to calculate the actual deaths. It may improve intensive care quality and allow the comparison of a given ICU against others in a benchmarking project. Thus, ICU SMR and benchmarking are increasingly reported in PubMed articles as shown in Fig. 1.

…