CHEERS Reporting Guidelines for Economic Evaluations

In 2009, the International Society for Pharmacoeconomics and Outcomes Research (ISPOR)¹ formed a 10-person taskforce to promote more standardized and rigorous reporting of economic evaluations. They conducted a 2-round modified Delphi panel including taskforce members plus an international, multidisciplinary team of 37 experts. Forty-four candidate items derived from 10 preexisting checklists and guidance documents were trimmed to 24. After comments from ISPOR members were solicited and reviewed, the ISPOR taskforce published CHEERS (Consolidated Health Economic Evaluation Reporting Standards) simultaneously across 10 English-language journals in 2013. It can be downloaded for free and has been translated into Portuguese and Spanish. Although many journals recommend adherence to CHEERS, it is unknown what proportion of published economic evaluations base their reporting on the CHEERS checklist. A systematic review² that includes original taskforce authors is being conducted to assess levels of transparency and reproducibility in contemporary economic evaluations of health care interventions and to investigate whether transparency and reproducibility have improved since the 2013 publication of CHEERS.

Use of the Reporting Guideline

The CHEERS guideline was developed for 2 primary audiences: (1) investigators conducting economic evaluations and (2) peer reviewers and editors at the journals to which these evaluations are submitted. Notably, the purpose of CHEERS is to assess the quality with which economic evaluations are reported rather than the quality of the research itself. But as observed with CONSORT (Consolidated Standards of Reporting Trials) and the subsequent quality of clinical trials,³ adoption of CHEERS by many journals as a requirement for submission may have contributed to an improvement in the methodologic rigor of published economic evaluations. The CHEERS guideline also addresses 2 potential problems. First, stakeholders can use CHEERS to determine the applicability of published evaluations to their own environments, thereby avoiding misapplication of findings and associated, potentially exorbitant opportunity costs. Second, in contrast to clinical trials, there is no centralized public registry of economic evaluations to allow for the iterative, longitudinal reevaluation now expected in an era of open science, so CHEERS provides a mechanism for accountability.¹

Health economic evaluations compare the relative costs of alternative interventions for a given condition, but they also typically compare the consequences or outcomes associated with these different interventions (Box). The nature of these consequences and the ways in which they are estimated are what characterize different types of economic evaluations, and CHEERS can be used to guide and assess their reporting (Table).

Box. Summary.

CHEERS (Consolidated Health Economic Evaluation Reporting Standards)

• What is it used for?

  • Cost-benefit analyses.
  • Cost-consequences analyses.
  • Cost-effectiveness analyses (including cost-utility analyses).
  • Cost-minimization analyses.
• How is it used?

  • 24-item checklist.
• Why is it used?

  • Ensures minimum level of quality for reporting economic analyses to facilitate.

    • Transparent reporting of data sources and methods.
    • Peer and editorial review.
    • Interpretability and generalizability of findings.

Table.

Types of Economic Evaluations^a

Type of economic evaluation	How consequences (ie, outcomes) are measured
Cost-benefit analysis	Consequences (ie, benefits) are measured using monetary units in the same way that costs are, thus a cost-benefit ratio is reported
Cost-consequences analysis	Consequences are reported as part of a nonordinal range and costs are also disaggregated, leaving consumers to determine the optimal solution for their own environment and circumstances
Cost-effectiveness analysis	Consequences are measured using “natural” units of health (eg, life-years)
Cost-utility analysisb	Consequences are measured using preference-based units of health (eg, QALYs); primary outcome measure is the incremental cost-effectiveness ratio, often expressed as cost per QALY
Cost-minimization analysis	Consequences of the comparators must, by definition, be equivalent, therefore only the costs of the alternative interventions are compared

Abbreviation: QALYs, quality-adjusted life-years.

^aAll economic evaluations measure and compare the costs of comparative interventions but they differ with regards to how and whether the outcomes of these interventions are measured, analyzed, and reported.

^bCost-utility analysis is a type of cost-effectiveness analysis.

Required Items

The 24 items in the CHEERS checklist are accompanied by recommendations as to the minimum amount of information that should be included for each aspect of the publication.

Title

This should indicate the type of economic evaluation being performed and which interventions are being compared.

Structured Abstract

This section should be explicit with regards to the units of measure for the costs and consequences, any employed discount rates, and any sensitivity analyses conducted to address uncertainty. Most importantly, it must have sufficient clarity and detail to serve as an accurate record of the evaluation that was conducted. Notably, abstracts for economic evaluations have been shown to frequently omit critical information needed to ensure appropriate interpretation of study methods and findings.⁴

Introduction

The section should briefly describe the study question (including study population, interventions being compared, and health care setting for the study), its implications for practice and/or policy, and the historical context in which the study should be situated. It should also identify whether the study is being conducted on behalf of or from the perspective of a particular stakeholder (eg, payer, device company, or society).

Methods

The study population, setting, perspective, and comparators should be detailed here along with the outcome measures (with an explanation as to why they were selected), estimates of effectiveness, and sources of cost data (eg, a clinical trial or an existing database) as well as outcome data (eg, patient-level for single-study analyses or model estimates for synthesis-based evaluations). For model-based evaluations, the type of model used should be described as well as reasons for its selection; any assumptions underpinning the model (eg, population- or disease-specific characteristics) as well as methods used to address data heterogeneity, missingness, skewing, and uncertainty should also be described.

Results

This section should present in tabular form all the clinical (eg, length of stay) and economic (eg, unit costs) parameters needed to calculate overall costs and consequences. For each intervention being compared, the appropriate measures of central tendency (eg, mean values) and precision (eg, standard deviation) for associated costs and consequences should be reported as well as, where applicable, the mean or median differences between comparators’ costs and consequences. Attention should be paid to reporting the results of any sensitivity analyses characterizing heterogeneity and uncertainty.

Discussion

This concluding section should summarize the key study findings as well as its limitations, generalizability, and relative contribution to and contextualization within the existing literature.

Potential Conflicts of Interest

Sources of monetary support and any conflicts of interest should be reported as fully as possible for any associations that fall within 5 years of publication, given the potentially significant application of many economic evaluations.

Limitations of the Reporting Guideline

As previously described, CHEERS is intended to be a guide for reporting, not conducting, economic evaluations. Guidelines to assist with performing and reporting them include the Consensus on Health Economic Criteria (CHEC) List, the Pediatrics Quality Appraisal Questionnaire (PQAQ), and the Quality of Health Economic Studies (QHES) List.⁵ Some have argued that while CHEERS is sufficiently robust for cost-effectiveness and cost-utility analyses, it performs less well for cost-benefit analyses. Finally, CHEERS is only meant to represent a bare minimum of publication requirements, and some journals may still require additional evidence of study quality. Nevertheless, adequate reporting based on CHEERS will still necessitate the use of appendices and online-only supplements given the wordcount limits of most journals.

Checklist and Flow Diagrams

The original ISPOR taskforce had intended to reevaluate CHEERS approximately 5 years after publication.⁶ A new task force is in the midst of creating CHEERS II, an update intended to address methodologic advances, the continued desire for transparency in reporting, the effect reporting standards have on analyses, and the increased use of CHEERS.⁷ The CHEERS II guideline as well as the aforementioned systematic review² on transparency and reproducibility in economic evaluations represent timely and important reassessments of the original CHEERS reporting guideline as well as its past and future effects on the quality of health economic evaluations.