ORAHS2025
Abstract Submission

201. Condorcet ranking for developing co-designed outcome measures in clinical research under the “Tournament Methods” framework

Invited abstract in session MC-2: Implementation, stream Sessions.

Monday, 11:00-12:30
Room: NTNU, Realfagbygget R8

Authors (first author is the speaker)

1. HANNAH JOHNS
Melbourne Medical School, University of Melbourne
2. Leonid Churilov
Melbourne Medical School, The University of Melbourne

Abstract

It is increasingly mandated by funding bodies that clinical research is co-designed with people that have lived experience of the condition under investigation. “Tournament Methods” are a class of statistical methods that satisfy this need by defining for each possible pair of patients, who had the better outcome. Under this approach, “better outcome” can be co-designed with patients, carers and other stakeholders to reflect trade-offs between multiple, potentially conflicting criteria.

However, biostatistical literature has only just begun to developed methods for facilitating the co-design of definitions of “better outcome” under the Tournament Methods framework. Determining preferences among alternatives with multiple, conflicting criteria is a mainstay of multicriteria decision analysis methods.

In this presentation, we discuss the methods currently used in biostatistics to elicit preferences that can be used to define “better outcome” under the Tournament Methods framework. We then discuss how well-established methods from operations research may be leveraged to improve this practice.

We demonstrate the application of these methods in practice by using Condorcet ranking to co-design a definition of “better off” in stroke research with a group of stroke survivors. We apply this definition to analyse data from a secondary prevention trial in a manner that reflects the experiences and preferences of stoke survivors.

Keywords

Status: accepted

Back to the list of papers