A full Bayesian model to handle structural ones and missingness in economic evaluations from individual‐level data

Abstract

Funding informationThe Foundation BLANCEFLOR Boncompagni Ludovisi, née Bildt ; Mapi Group Economic evaluations from individual-level data are an important component of the process of technology appraisal, with a view to informing resource allocation decisions. A critical problem in these analyses is that both effectiveness and cost data typically present some complexity (eg, nonnormality, spikes, and missingness) that should be addressed using appropriate methods. However, in routine analyses, standardised approaches are typically used, possibly leading to biassed inferences. We present a general Bayesian framework that can handle the complexity. We show the benefits of using our approach with a motivating example, the MenSS trial, for which there are spikes at one in the effectiveness and missingness in both outcomes. We contrast a set of increasingly complex models and perform sensitivity analysis to assess the robustness of the conclusions to a range of plausible missingness assumptions. We demonstrate the flexibility of our approach with a second example, the PBS trial, and extend the framework to accommodate the characteristics of the data in this study. This paper highlights the importance of adopting a comprehensive modelling approach to economic evaluations and the strategic advantages of building these complex models within a Bayesian framework.

The Statistical Approach in Trial-based Economic Evaluations Matters: Get your Statistics Together!

“…Additionally, missing data was assumed to be MAR, however, this assumption might not always hold and data could be MNAR. Recently, an increasing number of guidelines and studies emphasize the importance of checking for possible departure from the MAR assumption [28,36,[75][76][77]. It is recommended to perform sensitivity analyses, using other methods such as selection and/or patternmixture models [76].…”

Comparing Methods for Handling Missing Cost and Outcome Data in Clinical Trial-based Cost-effectiveness Analysis

“…In principle BPA allows to fit the model of interest while simultaneously handling all the different issues related to the missing data as well as to correctly propagate and quantify uncertainty. Each unobserved quantity in the model is handled as if it were a parameter (17,(42)(43)(44). Costs and QALYS were estimated through a joint distribution based on Markov Chain Monte Carlo (MCMC) using the R function selection, within the missingHE package (17).…”

Pns311 Predictive-Adjusted Indirect Comparison (Paic): A Novel Method for Population-Adjusted Indirect Comparison

“…The generation of synthetic datasets is a fundamentally Bayesian operation, as is MI. 27,48,49 Both can be regarded as fully Bayesian approaches; with a few posterior samples making up the imputed datasets. In PAIC, we can use efficient sampling methods such as Markov Chain Monte Carlo (MCMC) in this stage.…”

“…In standard multiple imputation, the imputation and analysis stages may be performed simultaneously in a joint model. 49 However, this is problematic in PAIC as the dependent variable * of the analysis is completely synthesized. Consider the DAG in Figure 1.…”