Statistical considerations in the design and analysis of cross-over trials

Cross-over designs are frequently used in clinical trials, usually to compare two treatments. It has long been known that the simple two-period cross-over is adversely affected by the presence of carry-over effects, but it is nevertheless still commonly used. This thesis examines the problems caused by carry-over effects in detail and considers ways of overcoming these problems, either by using a more complex cross-over design, or by carrying out a more sophisticated statistical analysis. Only designs for comparing two treatments are considered, but analysis for continuous and binary variables is covered.

Name dc:type.qualificationname

PhD

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

De Montfort University

Year dc:date.issued

1991

Author dc:creator

• Morrey, Gilbert Heneage

Licence

• https://dora.dmu.ac.uk/bitstreams/4a58e4de-d0cc-4bcc-83c0-820784c493fc/download