Incorporating equity concerns in cost-effectiveness analyses: A systematic literature review

Abstract

Introduction The aim of this study is to review analytical methods that enable the incorporation of equity concerns within economic evaluation. Methods A systematic search of PubMed, Embase and EconLit was undertaken from database inception to February 2021. The search was designed to identify methodological approaches that are currently employed to evaluate health-related equity impacts in economic evaluation studies of health care interventions. Studies were eligible if they described or elaborated on a formal quantitative method used to integrate equity concerns within economic evaluation studies. Cost-utility, cost-effectiveness, cost-benefit, cost-minimisation and, cost-consequence analyses, as well as health technology appraisal and budget impact analysis, alongside any relevant literature reviews, were included. For each of the identified methods, summaries of the scope of equity considerations covered, the methods employed and their key attributes, data requirements, outcomes, and strengths and weaknesses were provided. A traffic light assessment of the practical suitability of each method was undertaken, alongside a worked example, applying the different methods to evaluate the same decision problem. Finally, the review summarises the typical trade-offs arising in cost-effectiveness analyses and discusses the extent to which the evaluation methods are able to capture these. Results In total, 68 studies were included in the review and methods could broadly be grouped into equity-based weighting (EBW) methods, extended cost-effectiveness analysis (ECEA), distributional cost-effectiveness analysis (DCEA), multi-criteria decision analysis (MCDA), and mathematical programming (MP). EBW and MP methods enable equity consideration through adjustment to incremental cost-effectiveness ratios, whilst equity considerations are represented through financial risk protection (FRP) outcomes in ECEA, social welfare functions (SWFs) in DCEA, and scoring/ranking systems in MCDA. The review identified potential concerns for EBW methods and MCDA with respect to data availability, and EBW methods and MP with respect to explicitly measuring changes in inequality. The only potential concern for ECEA relates to the use of FRP metrics which may not be relevant for all healthcare systems. In contrast, DCEA observed no significant concerns but relies on the use of SWFs which may be unfamiliar to some audiences and requires societal preference elicitation. Consideration of typical cost-effectiveness and equity-related trade-offs highlighted the flexibility of most methods with respect to their ability to capture such trade-offs. Notable exceptions were trade-offs between quality of life and length of life, for which we find DCEA and ECEA unsuitable, and the assessment of lost opportunity costs, for which we find only DCEA and MP to be suitable. The worked example demonstrated that each method is designed with fundamentally different analytical objectives in mind. Conclusions The review emphasises that, not only are some approaches better suited to particular decision problems than others, but also that methods are subject to different practical requirements and that significantly different conclusions can be observed depending on the choice of method and the assumptions made. Further, to fully operationalise these frameworks, there remains a need to develop consensus over the motivation for equity assessment, which should necessarily be informed with stakeholder involvement. Future research of this topic should be a priority, particularly within the context of equity evaluation in health care policy decisions.