In a recently published meta-epidemiological study of 604 randomised clinical trials (RCTs) published between 1977 and 2018 from 53 Cochrane systematic reviews in critical care, less than 7% of the RCTs had overall low risk of bias, i.e. had high methodological quality, and the proportion did not increase over time.

Also, the median sample size of the trials did not seem to increase over time.

As trials with lower methodological quality and higher risk of bias are associated with inflated effect estimates, we need to critically appraise and interpret critical care trials considering their methodological quality.

One important methodological characteristic to consider is clinical heterogeneity.

It has been argued that critically ill patients constitute the most heterogeneous population in the hospital and that no two patients are the same.

Clustering of critically ill patients into sub-phenotypes based on clinical parameters reflecting disease mechanisms and/or treatment responses, rather than admission diagnoses has therefore been proposed.

Read More