Challenges of Systematic Reviews of Economic Evaluations: A Review of Recent Reviews and an Obesity Case Study

Decision makers increasingly require cost-effectiveness evidence to inform resource allocation and the need for systematic reviews of economic evaluations (SREEs) has grown accordingly. The objective of this article is to describe current practice and identify unique challenges in conducting and reporting SREEs. Current guideline documents for SREEs were consulted and summarised. A rapid review of English-language SREEs, using MEDLINE and EMBASE, published in 2017/2018, containing at least 20 studies was undertaken to describe current practice. Information on data extraction methods, quality assessment (QA) tools and reporting methods were narratively summarised. Lessons learned from a recently conducted SREE of weight loss interventions for severely obese adults were discussed. Sixty-three publications were included in the rapid review. Substantial heterogeneity in review methods, reporting standards and QA approaches was evident. Our recently conducted SREE on weight loss interventions identified scope to improve process efficiency, opportunity for more transparent and succinct reporting, and potential to improve consistency of QA. Practical solutions may include (1) using pre-piloted data extraction forms linked explicitly to results tables; (2) consistently reporting on key assumptions and sensitivity analyses that drive results; and (3) using checklists that include topic-specific items where relevant and allow reviewers to distinguish between reporting, justification and QA. The lack of a mutually agreed, standardised set of best practice guidelines has led to substantial heterogeneity in the conduct and reporting of SREEs. Future work is required to standardise the approach to conducting SREEs so that they can generate efficient, timely and relevant evidence to support decision-making. We suggest only data extracting information that will be reported, focusing discussion around the key drivers of cost-effectiveness, and improving consistency in QA by distinguishing between what is reported, justified by authors and deemed appropriate by the reviewer.