Skip to content


  • Poster presentation
  • Open Access

Making core outcomes more clinically meaningful: 2-for-the-price-of-1 with a distributional approach

Trials201516 (Suppl 1) :P16

  • Published:


  • Public Health
  • Standard Deviation
  • High Risk
  • Lung Function
  • Multiple Testing


In RCTs with a continuous outcome, it is easier to interpret a difference in proportions at high risk than the difference in means. For example when comparing lung function in two groups, a difference in proportion <10th centile may be more clinically meaningful than the difference in means. However choosing a dichotomised continuous outcome at the outset requires a larger sample than using the raw data and may also distort the true difference since data are discarded on dichotomisation.

Materials and methods

We have derived a dual-outcome solution where the difference in means and difference in proportion at high risk are equivalent inference-wise [1]. The distributional approach treats the proportion below a cut-point as a function of the mean and standard deviation of the distribution, assuming the data are Normal or can be transformed to Normal. The method is robust in situations commonly encountered [2]. We illustrate here with lung function (LF) outcome data [3].


This trial compared LF in adolescent children. They had been randomised to one of two types of ventilation at birth. The primary LF outcome forced expiratory flow 75% z-score (FEF75) differed significantly between the groups (difference in z-scores 0.23; 95% CI: 0.02, 0.45, p=0.04). The difference in percentage <10th centile-for-normal was calculated using the distributional approach as: 10.3%; 95% CI: 0.7, 20.0%, p=0.04.


Using the distributional approach an outcome can be presented as both a difference in means and difference in proportions at high risk without penalty due to loss of power or multiple testing since the test of the means is equivalent to the test of the proportion at high risk.

Authors’ Affiliations

Division of Health and Social Care Research, King’s College London, London, SE1 3QD, UK
NIHR Biomedical Research Centre at Guy’s and St Thomas’ NHS Foundation Trust and King's College London, London, SE1 3QD, UK
AG Epidemiologie & International Public Health, Bielefeld University, Postfach 10 01 31, D-33501 Bielefeld, Germany


  1. Peacock JL, Sauzet O, Ewings SM, Kerry SM: Dichotomizing continuous data while retaining statistical power using a distributional approach. Statist Med. 2012, 31: 3089-3103. 10.1002/sim.5354.View ArticleGoogle Scholar
  2. Sauzet O, Peacock JL: Estimating dichotomised outcomes in two groups with unequal variances: a distributional approach. Statist Med. 2014, 33: 4547-4559. 10.1002/sim.6255.View ArticleGoogle Scholar
  3. Zivanovic S, Peacock JL, Alcazar-Paris M, Lo JW, Lunt A, Marlow N, Calvert SA, Greenough A: Late Outcomes of a Randomized Neonatal Trial of High Frequency Oscillation. N Engl J Med. 2014, 370: 1121-1130. 10.1056/NEJMoa1309220.View ArticlePubMedPubMed CentralGoogle Scholar


© Peacock et al; licensee BioMed Central Ltd. 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.