Design aspects of individually randomised trials, cluster randomised trials and stepped wedge designs


Course title

Design aspects of individually randomised trials, cluster randomised trials and stepped wedge designs.



Mike Campbell () and Erik Cobo ().

M. J. Campbell is Emeritus Professor of Medical Statistics at the University of Sheffield. He is the author (co-author) of seven books about the role of Statistics in Medical Research and 240 publications in refered journals. The last book was published in 2014 by Wiley-Blackwell, and its title is: How to design, analyse and report cluster randomized trials in medicine and health related research.

Erik Cobo: Graduate, 1980, and phD 1987, in Medicine from the University of Barcelona, DEA in Biometry and Epidemiology, 1986, at the University of Paris VI, Diploma in Social Science Data Analysis at the University of Essex, 1988. Assistant Professor of the Dept. d'Estadística i Investigació Operativa at UPC.


Course Language



Course Schedule

July 4 to 6: 9:00am to 1:00pm.

July 7: 9:00am to 12:00am.



The purpose of this course is to introduce the randomised controlled trial and its extensions into cluster randomisation and stepped wedge design from a statistical point of view. The design aspects will be considered, especially the size of the trial. The analysis and reporting will also be discussed, using CONSORT statements for each design (the CONSORT statement for stepped wedge is being developed)  . The package R will be used for analysis. The course will be evaluated using a Multiple Choice Questionnaire.


Lecture plan

  • Day 1 4hrs (with Dr E Cobo Valeri)

Individually Randomised Trials

History, purpose of Randomisation, (ABC- Allocation at random, Blindness and Control) considerations of ICHE9, Design Parallel group, Factorial, Cross-Over, CONSORT, sample size.

  • Day 2 4 hrs

Cluster trials

Design and sample size,   CONSORT statement.

Reading a paper on a cluster trial.

  • Day 3 4 hrs

Cluster trials

Analysis and reporting, ethics.

  • Day 4 3 hrs

Stepped wedge designs

Purpose, analysi,s sample size CONSORT.