An investigation into methods for demonstrating when a generic, preference-based measure of health-related quality of life misrepresents health gain
NICE uses cost effectiveness analysis to inform guidance, with Quality Adjusted Life Years (QALYs) as the measure of benefit. For the sake of consistency, NICE prefers that EQ-5D-3L is used for the estimation of health benefits, unless shown to be inappropriate.
Whether EQ-5D, or some other preference based measure (PBM) is used, it is often claimed that the chosen instrument misses out some aspects of the disease that are important to patients, either because the relevant health domain is entirely absent from the ED-5D descriptive system, or that the response categories are too insensitive to small changes in health. This raises issues as to how such claims can be substantiated and, if they are, the extent to which any misrepresentation of health gain might impact on estimates of cost effectiveness.
The current NICE Methods Guide implicitly characterises the issue of “inappropriateness” as a binary concept when in most disease areas this may be more accurately considered a matter of degree. It is recommended that empirical evidence of standard psychometric based concepts like responsiveness are undertaken (see Wailoo et al 2010 for more information). However, appraisal committees are often left without evidence that EQ-5D is entirely inappropriate. In these situations they must take into account, in a qualitative manner, any potential underestimation of health benefit when formulating guidance.
It would be helpful to committees, and promote consistency of decision making, if methods were available to demonstrate both the validity of any claims that EQ-5D (or any other preference based instrument) underestimates health gain, based on the premise that this is likely to be a matter of degree in all disease areas. Where such claims are found to have merit, this project will seek to develop methods which quantify the degree (or range) of underestimation that could be considered feasible.
DSU Publication
M Hernández Alava, SE Pudney, AJ Wailoo (2023). Does EQ-5D tell the whole story? Statistical methods for comparing the thematic coverage of clinical and generic outcome measures, with application to breast cancer. Value in Health. DOI: https://doi.org/10.1016/j.jval.2023.05.016