11 - Trials

The application of evidence-based medicine is a skill fundamental to the practice of critical care medicine. The best evidence to assess the risks and benefits of both new and currently used therapies comes from randomised clinical trials (RCTs). In order to conclude that the results of an RCT are reliable, it is important that bias is minimised. Key steps in minimising bias are randomisation that maintains allocation concealment, blinding whenever possible, control of confounding by assessing baseline balance and use of concomitant therapies, and complete follow up. It is also important that methods of analysis and reporting are robust and transparent. RCTs are expensive and may take many years to plan and conduct, recent innovations that seek to improve the efficiency of the clinical trials enterprise include adaptive, platform and basket trials. At the same time the use of Bayesian rather than frequentist statistics is becoming more widespread with its advocates claiming it avoids dichotomisation and helps embrace uncertainty, providing clinicians with nuanced evidence. There are some questions faced by clinicians that cannot be addressed by RCTs, and under these circumstances other study designs, such as cohort studies or case control studies, may provide the best available, although imperfect, evidence to guide care.