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

Abstract

Background: Baseline imbalances, skewed costs, the correlation between costs and effects, and missing data are statistical challenges that are often not adequately accounted for in the analysis of cost-effectiveness data. This study aims to illustrate the impact of accounting for these statistical challenges in trial-based economic evaluations. Methods: Data from two trial-based economic evaluations, the REALISE and HypoAware studies, were used. In total, 14 full cost-effectiveness analyses were performed per study, in which the four statistical challenges in trial-based economic evaluations were taken into account step-by-step. Statistical approaches were compared in terms of the resulting cost and effect differences, ICERs, and probabilities of cost-effectiveness. Results: In the REALISE study and HypoAware study, the ICER ranged from 636,744€/QALY and 90,989€/QALY when ignoring all statistical challenges to -7,502€/QALY and 46,592€/QALY when accounting for all statistical challenges, respectively. The probabilities of the intervention being cost-effective at 0€/ QALY gained were 0.67 and 0.59 when ignoring all statistical challenges, and 0.54 and 0.27 when all of the statistical challenges were taken into account for the REALISE study and HypoAware study, respectively.Conclusions: Not accounting for baseline imbalances, skewed costs, correlated costs and effects, and missing data in trial-based economic evaluations may notably impact results. Therefore, when conducting trial-based economic evaluations, it is important to align the statistical approach with the identified statistical challenges in cost-effectiveness data. To facilitate researchers in handling statistical challenges in trial-based economic evaluations, software code is provided.

A scoping review of statistical methods for trial‐based economic evaluations: The current state of play

“…The large number of identified statistical methods also shows that researchers have a large arsenal of methods to choose from when conducting a trial‐based economic evaluation. As different statistical methods can result in different cost‐effectiveness outcomes and/or conclusions, it is important that researchers define, substantiate, and preferably publish their economic evaluation analyses in a Statistical Analysis Plan and/or a Health Economics Analysis Plan before looking at the acquired data (Mutubuki et al., 2021; Thorn et al., 2021).…”

“…Simply excluding participants with missing data results in lower precision and is likely to result in biased estimates. Hence, more advanced statistical methods are required to validly handle missing data in trial‐based economic evaluations (Burton, Billingham, & Bryan, 2007; Leurent, Gomes, & Carpenter, 2018; Mutubuki et al., 2021). Censoring When health effects are measured as time‐to‐event outcomes (e.g., survival time, QALYs), censoring might occur.…”

“…However, as RCTs do not have infinite samples, patient characteristics might still be imbalanced between groups (i.e., baseline imbalances) (de Boer et al, 2015). If these characteristics are associated with both the intervention and the outcome, confounding introduced by baseline imbalances may result in imprecise and biased estimates if they are not accounted for in the analysis of a trial‐based economic evaluation (Manca, Hawkins, & Sculpher, 2005; Mutubuki et al., 2021; Thompson & Nixon, 2005). Skewed costs and/or effects Cost data are typically right‐skewed, as there usually are few participants with (very) high costs and costs are bounded at zero.…”