Skip to content


  • Poster presentation
  • Open Access

Analysis and reporting of individually randomised controlled trials with clustering in one arm only: how should we do it?

  • 1,
  • 1,
  • 1,
  • 1,
  • 1 and
  • 1
Trials201516 (Suppl 2) :P139

  • Published:


  • Public Health
  • Statistical Method
  • Random Effect
  • Back Pain
  • Random Effect Model

In an individually randomised controlled trial (iRCT) where the intervention is delivered by a health professional/therapist it seems likely that the effectiveness of the intervention could depend on the skill of the therapist delivering it. This leads to a potential clustering of the outcomes for the patients treated by the same therapist.

If outcomes are clustered then the usual statistical methods for analysing RCT data may not be appropriate as they assume that outcomes observed on different participants are independent.

In some iRCTs there is clustering in only one arm; where the therapist only treats participants in one arm of the trial but there is no equivalent clustering in the control arm.

Several strategies have been proposed for this problem: ignoring the clustering; imposing clusters in the control arm and fitting a fixed or random effects model, or using a partially clustered approach where only the clustering in the treatment arm is modelled.

This talk will describe and compare the statistical methods for analysing continuous outcomes from an iRCT with some element of clustering in one arm. Four confirmatory case studies (with a combined 1,167 randomised participants) will be analysed using each of the methods: specialist clinics for the treatment of venous leg ulcers (Morrell et al. 1998), acupuncture for low back pain (Thomas et al. 2006), cost-effectiveness of community postnatal support workers (CPSW) (Morrell et al. 2000), and Putting Life in Years (PLINY) (Mountain et al. 2014). Recommendations for the best and most practical approaches will be made.

Authors’ Affiliations

University of Sheffield, Sheffield, UK


© Walters et al. 2015

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate. Please note that comments may be removed without notice if they are flagged by another user or do not comply with our community guidelines.