Bayesian network meta-regression for aggregate ordinal outcomes with imprecise categories

Comparing emerging treatment options is often challenging because of the sparseness of direct comparisons from head-to-head trials and inconsistencies in outcome measures among published placebo-controlled trials for each treatment. One potential solution is to aggregate the different outcome measures into a single ordinal response variable for consistent evaluation. The ordinal response variable will inevitably contain unknown response categories because they cannot be directly derived from published data in the literature. In this paper, we propose a statistical methodology to overcome such a common but unresolved issue in the context of network meta-regression for aggregate ordinal outcomes. Specifically, we introduce unobserved latent counts and model these counts within a Bayesian framework. The proposed approach includes several existing models as special cases and also allows us to conduct a proper statistical analysis in the presence of trials with certain missing categories. We then develop an efficient Markov chain Monte Carlo sampling algorithm to carry out Bayesian computation. Variations of the deviance information criterion and widely applicable information criterion are used for the assessment of goodness-of-fit under different distributions of the latent counts. A case study demonstrating the usefulness of the proposed methodology is conducted using aggregate ordinal outcome data from 18 clinical trials in treating Crohn's Disease.