What is it about?

It is shown analytically and justified by simulation that if balance is defined in terms of its effect or otherwise on the validity of statistical inference it does not increase with increasing sample size. This is because standard statistical methods make an implicit allowance for imbalance and the smaller the trial the larger the allowance that is made.

Featured Image

Why is it important?

A simple analytical formula is provided and used to interpret previous simulate results coming to a different conclusion, one which leaves room for trials with more modest numbers of patients.


I have been thinking about this topic for many years and am convinced that balance is something of an unnecessary obsession in discussion of randomised trials.

Professor Stephen J Senn
Consultant Statistician

Read the Original

This page is a summary of: Empirical studies of balance do not justify a requirement for 1,000 patients per trial, Journal of Clinical Epidemiology, August 2022, Elsevier, DOI: 10.1016/j.jclinepi.2022.02.010.
You can read the full text:




The following have contributed to this page