What is it about?

A simple model shows that between study variation can be important when using historical controls. The TARGET study is used to illustrate this. Some implications for epidemiological studies are discussed. Recommendations for planning and reporting studies are made.

Featured Image

Why is it important?

Researchers are encouraged to think about using historical data in a way that confronts the problems from the beginning. This should encourage more realistic approaches to using historical controls


My involvement in the TARGET study some years ago as a member of a data safety monitoring has an important effect on my thinking. The study was run in two sub studies. At the sub-study level patients were randomised within centre between treatments but at the study level allocation was between centre. There were striking baseline differences between sub-studies but not within.

Professor Stephen J Senn
Consultant Statistician

Read the Original

This page is a summary of: Clustered allocation as a way of understanding historical controls: Components of variation and regulatory considerations, Statistical Methods in Medical Research, October 2019, SAGE Publications,
DOI: 10.1177/0962280219880213.
You can read the full text:




The following have contributed to this page