Review of quality-assessment tools in systematic reviews of cost-effectiveness analyses. Are checklists fit for purpose?




Poster session 3 Friday: Evidence Tools / Evidence synthesis - creation, publication and updating in the digital age


Friday 15 September 2017 - 12:30 to 14:00


All authors in correct order:

Wei C1, Armstrong N1, Ryder S1, Kleijnen J1
1 Kleijnen Systematic Reviews Ltd, United Kingdom
Presenting author and contact person

Presenting author:

Ching-Yun Wei

Contact person:

Abstract text
Background: Cost-effectiveness analysis (CEA) is increasingly employed around the world to make reimbursement decisions. However, before constructing a de novo economic model it is important to review all published analyses. Quality assessment tools should guide the critical appraisal of CEAs and so inform model development. The Drummond checklist has been the most commonly used. However, it was developed about 20 years ago and others, more appropriate for modelling have been developed since. Nevertheless, the extent of use and usefulness remain uncertain.

Objectives: The study objective was to identify and compare quality assessment tools and assess how they are used in critically appraising CEAs in recent systematic reviews.

Methods: Eligible studies were retrieved from a set of about 15 000 records from comprehensive searches used to populate a database of all systematic reviews in healthcare. Data on quality assessment tools and methods of assessing methodological quality of CEAs, such as use of systematic reviews to parameterise models, were extracted.

Results: 30% of reviews used Drummond, 12% used CHEERS, 15% used QHES, 7% used the Philips checklist, 6% used CHEC, 7% used more than one checklist and 14% did not use any quality assessment tool. Most reviews used tools to assess transparency rather than methodological quality.

Conclusions: The critical appraisal of CEAs is not straightforward and can be subjective. Worryingly, the most commonly used tool is 20 years old and not designed for economic modelling. Other tools, designed specifically for models, such as the Philips checklist, could reasonably be regarded as more appropriate. Variation in approach to critical appraisal suggests a new instrument to assess CEA quality might be needed.