My earlier post Clinical trial design — for beginners has been one of my more popular ones. People who’ve read that and would like to know more, can download an Adobe pdf file, Inferential Statistics, Descriptive Statistics and The Analysis Plan.
I wrote it as Chapter 14 of the first edition of Steven E. Linberg’s book, Expediting Drug and Biologics Development, Parexel International, 1995. My chapter is an introduction to basic statistics in clinical trials for non-statisticians. Although not very technical, it’s more formal and more comprehensive than Clinical trial design — for beginners.
The theme of Steve’s book was “Target-oriented design” of clinical research. The idea is to start with the end: What are you trying to prove? What do you want the FDA to agree to? Figure out what you’d need to demonstrate in order to get that. Then figure out the scientific step before that . . . and so on, until you work your way back to the starting point where you are now.
This is a great approach — one I still use with my clients and that I use in evaluating research proposals I’m asked to review. I start first by understanding the aims, and then I ask if the plans are adequate to get there. People tend start their thinking with where they’d like to go next, and then bumble along to the next step . . . and they perhaps wind up disappointed not to get where they wanted to in the end. I encourage them to think about goals right away.
A further continuation is also available as a pdf presentation download, geared to students in a graduate course in clinical trial management. Statistical Design: Clinical Development of Drugs and Biologics.
A more elementary discussion is in the entry Clinical trial design — for beginners.