Open AccessPitfalls of and Controversies in Cluster Randomization Trials
Author(s) -
Allan Donner,
Neil Klar
Publication year - 2004
Publication title -
american journal of public health
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.284
H-Index - 264
eISSN - 1541-0048
pISSN - 0090-0036
DOI - 10.2105/ajph.94.3.416
Subject(s) - randomization , inference , cluster (spacecraft) , randomized controlled trial , multitude , research design , clinical trial , selection (genetic algorithm) , cluster randomised controlled trial , causal inference , medicine , computer science , statistics , artificial intelligence , pathology , mathematics , political science , law , programming language
It is now well known that standard statistical procedures become invalidated when applied to cluster randomized trials in which the unit of inference is the individual. A resulting consequence is that researchers conducting such trials are faced with a multitude of design choices, including selection of the primary unit of inference, the degree to which clusters should be matched or stratified by prognostic factors at baseline, and decisions related to cluster subsampling. Moreover, application of ethical principles developed for individually randomized trials may also require modification. We discuss several topics related to these issues, with emphasis on the choices that must be made in the planning stages of a trial and on some potential pitfalls to be avoided.