Connect

Member Research and Reports

Member Research and Reports

UAB Discusses Design, Analysis, and Interpretation of Cluster Randomized Controlled Trials

With growing interest in community-based, public health, and policy interventions to reduce obesity or to improve nutrition, the use of cluster randomized controlled trials (cRCTs) has increased. Also known as group randomized trials and community-randomized trials, cRCTs are experiments in which groups of individuals — such as families, classrooms, schools, worksites, or counties — are randomly assigned to experimental conditions, whereas outcomes are recorded at the individual level. Errors in the design, analysis, and interpretation of cRCTs are unfortunately all too common. This situation seems to stem in part from investigator confusion about how the unit of randomization affects causal inferences and the statistical procedures required for the valid estimation and testing of effects.

BrownA_UAB ASPPH LiP_UAB_ASPPH
[Photo: Dr. Andrew W. Brown (left) and Dr. Peng Li]

In a recent article, Dr. Andrew W. Brown, scientist in the Nutrition Obesity Research Center (NORC) and office of energetics, and Dr. Peng Li, statistician in the office of energetics, at the University of Alabama at Birmingham, provide a brief introduction and overview of the importance of cRCTs and highlight and explain important considerations for the design, analysis, and reporting of cRCTs by using published examples. “cRCTs are invaluable for studying interventions that must be given to groups,” Dr. Brown said, “but they can take more planning and different statistical analyses than simple RCTs. We hope our article will help make investigators aware of some of these challenges, particularly in the realms of nutrition and obesity.”

Co-authors are Dr. Kathryn Kaiser, instructor in UAB’s office of energetics; Dr. Michelle M. Bohan Brown, consultant in UAB’s department of nutrition sciences and assistant professor at Clemson University; Dr. David B. Allison, distinguished professor and director of UAB’s NORC and office of energetics; UAB alum Dr. Scott W. Keith, at Thomas Jefferson University; and Dr. J. Michael Oakes, at the University of Minnesota.

“Best (But Often Forgotten) Practices: Designing, Analyzing, and Reporting Cluster Randomized Controlled Trials” was published in May in The American Journal of Clinical Nutrition.

Journal article: http://ajcn.nutrition.org/content/early/2015/05/27/ajcn.114.105072.abstract