Chapter 7

Statistical analysis of recurrent adverse events

Liqun Diao Ph.D.

Liqun Diao Ph.D.

Department of Statistics and Actuarial Science, University of Waterloo, 200 University Avenue West, Waterloo, ON, Canada N2L 3G1

Search for more papers by this author
Richard J. Cook Ph.D.

Richard J. Cook Ph.D.

Department of Statistics and Actuarial Science, University of Waterloo, 200 University Avenue West, Waterloo, ON, Canada N2L 3G1

Search for more papers by this author
Ker-Ai Lee M.Sc.

Ker-Ai Lee M.Sc.

Department of Statistics and Actuarial Science, University of Waterloo, 200 University Avenue West, Waterloo, ON, Canada N2L 3G1

Search for more papers by this author
First published: 12 December 2014

Summary

In many clinical trials, adverse events may occur repeatedly over the course of treatment and follow-up. This chapter focuses primarily on the setting of transient adverse events for which it may be sensible to count the number of occurrences and make comparisons between groups on the basis of these counts. This is often reasonable when individuals are followed for the same length of time and there is little interest in when the events occur. The chapter defines notation and discusses general models for recurrent events. Simple regression models are appealing when comparing treatment groups and the chapter describes one such regression approach here. It is increasingly common for the Food and Drug Administration to recommend conduct of large Phase 4 trials to facilitate the collection of more extensive adverse event data in a sample of individuals treated under the standard of care.

The full text of this article hosted at iucr.org is unavailable due to technical difficulties.