2003 DesignandAnalysisofCrossoverTri

Jump to: navigation, search

Subject Headings: Cross-Over Trial, Cross-Over Trial Evaluation, Cross-Over Trial Evaluation Algorithm, Direct-by-Period Interaction


Cited By



This book is concerned with a particular sort of comparative trial known as the cross-over trial in which subjects receive different sequences of treatments. Such trials are widely used in clinical and medical research and in other diverse areas such as veterinary science, psychology, sports science and agriculture. The first edition of this book, which appeared in 1989, was the first to be wholly devoted to the subject. We remarked in the preface to the first edition that there existed a "large and growing literature." This growth has continued during the intervening years, and includes the appearance of three other books devoted to the topic. Naturally the newer developments have not been spread uniformly across the subject, but have reflected both the areas where the design has remained in widespread use and new areas where it has grown in importance. Equally naturally, some of the literature also reflects the particular interests and idiosyncracies of key researchers. This new edition of the book reflects those areas of development that we regard as important from a practical perspective, but where material is as appropriate and relevant now as it was before, we have kept the previous structure more or less intact.

In the first edition we started with a chapter wholly devoted to the two-period two-treatment design. The aim was that this should be, to a large extent, self-contained and mostly at a level that was accessible to the less statistically experienced. It was intended that this chapter alone would be sufficient for those who had no need to venture beyond this simple design. We have kept this structure, and aim, for the second edition but have enlarged the chapter somewhat, both to incorporate intervening developments, and to increase the degree of self-sufficiency by including more on nonparametric analyses and simple analyses for binary data.

One major change over the last 14 years has been the development of very general tools for the analysis both for continuous and discrete dependent data that are available in widely used computer packages. This has meant that we are now able to bring virtually all the analyses in the book into a small number of general frameworks, and in turn, this has allowed a more coherent development of the methods for analysis. This is reflected in two new chapters (5 and 6) on the analysis of continuous and categorical cross-over data, respectively. These have absorbed the previous chapters and sections on analysis that were scattered throughout the first edition of the book. Our approach to the analysis of categorical data in particular has been much affected by the many developments of the past 14 years. The wide availability of the necessary software tools has also meant that we have been able to provide the required statements to allow the reader to apply the methods described in these chapters.

The two-treatment cross-over trials have become increasingly widely used in recent years in bio-equivalence trials. The special nature of the analyses associated with such trials has evolved to the point that we have included an entirely new chapter on this topic. The methods used there rest heavily on the framework developed in Chapter 5.



 AuthorvolumeDate ValuetitletypejournaltitleUrldoinoteyear
2003 DesignandAnalysisofCrossoverTriByron Jones
Michael G Kenward
Design and Analysis of Cross over Trials2003
AuthorByron Jones + and Michael G Kenward +
titleDesign and Analysis of Cross over Trials +
year2003 +