Model‐Assisted Bayesian Estimators of Transparent Population Level Summary Measures for Ordinal Outcomes in Randomized Controlled Trials
Open Access
- 19 May 2026
- journal article
- Published by Wiley in Statistics in Medicine
- Vol. 45 (10-12) , e70603
Abstract
In randomized controlled trials, ordinal outcomes typically improve statistical efficiency over binary outcomes. The treatment effect on an ordinal outcome is usually described by the odds ratio from a proportional odds model, but this summary measure lacks transparency with respect to its emphasis on the components of the ordinal outcome when proportional odds is violated. We propose various summary measures for ordinal outcomes that are fully transparent in this regard, including ‘weighted geometric mean’ odds ratios and relative risks, and ‘weighted mean’ risk differences. We also develop and evaluate efficient model‐assisted Bayesian estimators for these population‐level summary measures based on nonproportional odds models that facilitate covariate adjustment with marginalization via the Bayesian bootstrap. We propose a weighting scheme that engenders appealing heuristic properties, including invariance to whether the ordinal outcome is ordered from best to worst versus worst to best. Using computer simulation, we show that comparative testing based on the proposed population‐level summary measures performs well relative to a proportional odds approach. We also report an analysis of the COVID‐OUT trial, which exhibits evidence of a nonproportional odds treatment effect.Keywords
All Related Versions
Funding Information
- Rainwater Charitable Foundation
- National Center for Advancing Translational Sciences (KL2TR002492)
- National Center for Advancing Translational Sciences (UL1TR002494)
- National Institute of Diabetes and Digestive and Kidney Diseases (K23 DK124654)
This publication has 19 references indexed in Scilit:
- Randomized Trial of Metformin, Ivermectin, and Fluvoxamine for Covid-19New England Journal of Medicine, 2022
- Model‐assisted analyses of longitudinal, ordinal outcomes with absorbing statesStatistics in Medicine, 2022
- The Net Benefit of a treatment should take the correlation between benefits and harms into accountJournal of Clinical Epidemiology, 2021
- Framework for Selecting Clinical Outcomes for Clinical Trials2019
- Choosing outcomes for clinical trials: a pragmatic perspectiveCurrent Opinion in Critical Care, 2018
- Enhanced Precision in the Analysis of Randomized Trials with Ordinal OutcomesBiometrics, 2015
- A Flexible Bayesian Approach to Monotone Missing Data in Longitudinal Studies With Nonignorable Missingness With Application to an Acute Schizophrenia Clinical TrialJournal of the American Statistical Association, 2015
- The proportional odds with partial proportionality constraints model for ordinal response variablesSocial Science Research, 2012
- Nonparametric causal effects based on marginal structural modelsJournal of Statistical Planning and Inference, 2007
- Some odds ratio statistics for the analysis of ordered categorical dataBiometrika, 1974