Efficient statistical analysis of trial designs: win ratio and related approaches for composite outcomes

Abstract In randomized controlled clinical trials, composite outcomes are often used to study treatment effects. This approach is popular because it increases the number of observed events, enhancing statistical power while reducing the required patient sample size. However, composite outcomes do no...

Full description

Saved in:
Bibliographic Details
Main Authors: Wilson Fandino, Matthew Dodd, Gudrun Kunst, Tim Clayton
Format: Article
Language:English
Published: BMC 2025-07-01
Series:Perioperative Medicine
Online Access:https://doi.org/10.1186/s13741-025-00550-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849238291827654656
author Wilson Fandino
Matthew Dodd
Gudrun Kunst
Tim Clayton
author_facet Wilson Fandino
Matthew Dodd
Gudrun Kunst
Tim Clayton
author_sort Wilson Fandino
collection DOAJ
description Abstract In randomized controlled clinical trials, composite outcomes are often used to study treatment effects. This approach is popular because it increases the number of observed events, enhancing statistical power while reducing the required patient sample size. However, composite outcomes do not provide insight into the effect of individual endpoints. This becomes particularly relevant when mortality is combined with less critical but clinically relevant endpoints or when the clinical importance of individual endpoints varies significantly. As a result, interpreting composite outcomes can be challenging. This narrative review introduces the win ratio (WR), a method for prioritizing individual endpoints within a composite outcome. The WR offers an alternative to composite outcomes by considering the clinical importance of each component and prioritizing the most critical endpoint, such as death, over less significant events. Despite the popularity of the WR among cardiovascular trialists, this approach has not been extensively used in other areas of clinical research. We contend, that perioperative and periprocedural researchers could consider the WR and related approaches when the outcomes of interest are not of similar clinical importance. To this end, understanding the benefits and limitations of the WR will be essential to exploit its benefits, while avoiding potential misuses of the technique.
format Article
id doaj-art-442e87beab294d2ab91ffecf00cae70c
institution Kabale University
issn 2047-0525
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series Perioperative Medicine
spelling doaj-art-442e87beab294d2ab91ffecf00cae70c2025-08-20T04:01:41ZengBMCPerioperative Medicine2047-05252025-07-0114111010.1186/s13741-025-00550-8Efficient statistical analysis of trial designs: win ratio and related approaches for composite outcomesWilson Fandino0Matthew Dodd1Gudrun Kunst2Tim Clayton3Anaesthetics Department, Guy’s and St Thomas’ Hospital NHS Foundation TrustClinical Trials Unit, London School of Hygiene and Tropical MedicineDepartment of Anaesthetics, King’s College Hospital NHS Foundation Trust and School of Cardiovascular and Metabolic Medicine & Sciences, King’s College LondonClinical Trials Unit, London School of Hygiene and Tropical MedicineAbstract In randomized controlled clinical trials, composite outcomes are often used to study treatment effects. This approach is popular because it increases the number of observed events, enhancing statistical power while reducing the required patient sample size. However, composite outcomes do not provide insight into the effect of individual endpoints. This becomes particularly relevant when mortality is combined with less critical but clinically relevant endpoints or when the clinical importance of individual endpoints varies significantly. As a result, interpreting composite outcomes can be challenging. This narrative review introduces the win ratio (WR), a method for prioritizing individual endpoints within a composite outcome. The WR offers an alternative to composite outcomes by considering the clinical importance of each component and prioritizing the most critical endpoint, such as death, over less significant events. Despite the popularity of the WR among cardiovascular trialists, this approach has not been extensively used in other areas of clinical research. We contend, that perioperative and periprocedural researchers could consider the WR and related approaches when the outcomes of interest are not of similar clinical importance. To this end, understanding the benefits and limitations of the WR will be essential to exploit its benefits, while avoiding potential misuses of the technique.https://doi.org/10.1186/s13741-025-00550-8
spellingShingle Wilson Fandino
Matthew Dodd
Gudrun Kunst
Tim Clayton
Efficient statistical analysis of trial designs: win ratio and related approaches for composite outcomes
Perioperative Medicine
title Efficient statistical analysis of trial designs: win ratio and related approaches for composite outcomes
title_full Efficient statistical analysis of trial designs: win ratio and related approaches for composite outcomes
title_fullStr Efficient statistical analysis of trial designs: win ratio and related approaches for composite outcomes
title_full_unstemmed Efficient statistical analysis of trial designs: win ratio and related approaches for composite outcomes
title_short Efficient statistical analysis of trial designs: win ratio and related approaches for composite outcomes
title_sort efficient statistical analysis of trial designs win ratio and related approaches for composite outcomes
url https://doi.org/10.1186/s13741-025-00550-8
work_keys_str_mv AT wilsonfandino efficientstatisticalanalysisoftrialdesignswinratioandrelatedapproachesforcompositeoutcomes
AT matthewdodd efficientstatisticalanalysisoftrialdesignswinratioandrelatedapproachesforcompositeoutcomes
AT gudrunkunst efficientstatisticalanalysisoftrialdesignswinratioandrelatedapproachesforcompositeoutcomes
AT timclayton efficientstatisticalanalysisoftrialdesignswinratioandrelatedapproachesforcompositeoutcomes