28 September 2023 - A randomised clinical trial is frequently the preferred research design for testing new medical treatments. Randomisation helps to ensure that the participants in the treatment groups are similar in the distribution of prognostic factors. 

This minimises bias in statistical comparisons of patient outcomes and allows differences to be interpreted as the causal effect of treatment assignment. However, there are situations in which randomising participants to a control treatment or placebo within a randomised clinical trial may not be practical or ethical, eg, if a comparison with placebo is desired but an effective treatment already exists. In such cases, researchers might use data from participants who had received the intended control treatment in a prior study, termed historical controls, to estimate the benefit of the new treatment.

Read JAMA Viewpoint