An introduction to randomization for clinical trials 1

Published on September 29, 2016   34 min

Other Talks in the Series: Design and Analysis of Randomized Clinical Trials

Other Talks in the Series: The Risk of Bias in Randomized Clinical Trials

This is a Henry Stewart talk on an introduction to randomization for clinical trials. My name is William F. Rosenberger. I'm University Professor and Chairman of the Department of Statistics at George Mason University. I also have written two books on the subject of randomization.
Randomization is a time-tested concept that has appeared in clinical trials since the 1950s. I'm going to discuss different randomization procedures, and then I'm going to compare them, and then I'm going to talk about using the randomization itself as a basis for inference. I'll discuss stratification in clinical trials and then draw some conclusions. We begin with an introduction.
Clinical trials had been the gold standard in medical research for the past 70 years. Clinical trials began after the Second World War, and in fact, some of the great cryptographers in the war later became biostatisticians at the National Institutes of Health and the Medical Research Council in Britain following the war. And there we have the development of modern biostatistics, or in the UK often called medical statistics. In this talk, I'm going to focus on a particular area of statistical methodology that really forms the backbone of clinical trials, and that is the concept of randomization. Initially, randomization was developed in the context of agricultural studies. Sir R. A. Fisher was one of the leading proponents but randomization has a natural home in the context of clinical trials.

An introduction to randomization for clinical trials 1

Embed in course/own notes