A UK EQ-5D-5L value set is urgently required to enable the latest version of EQ-5D to inform policy, including evidence submitted to the National Institute for Health and Care Excellence as part of health technology appraisals. This paper presents the EQ-5D-5L UK value set generated from preference data elicited using the time trade-off (TTO) technique with a representative sample of the UK public.
In-person and videoconference interviews were undertaken using the composite TTO method for 102 health states with a representative sample of UK adults (age 18 years and older) across England, Wales, Scotland, and Northern Ireland. Data quality was rigorously and independently assessed throughout the study. TTO data was modeled using a range of models, with the value set being selected as the preferred model using predefined criteria.
Data quality standards were achieved. A total of 1200 interviews were conducted, 1102 (91.8%) by means of videoconference and 98 (8.2%) in person. The sample was representative of age, sex, ethnicity, and socioeconomic groups and proportionally representative for the 4 nations: England, Scotland, Wales, and Northern Ireland. TTO values highlighted by participants as ones they would reconsider were excluded. The value set was generated using a random effects Tobit model. The dimensions of anxiety/depression and usual activities have a greater relative impact on utilities than in the UK EQ-5D-3L value set, though the worst state value is comparable.
The study collected good-quality data from a representative sample of the UK public, modeled the data appropriately and transparently, to generate a UK EQ-5D-5L value set suitable for informing policy.
What is it about?
The EQ-5D-5L is a questionnaire measuring health-related quality of life across 5 dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each dimension has 5 response options, ie, levels of severity, and the EQ-5D-5L was developed to improve on EQ-5D-3L, a previous version with 3 response options per dimension. This study developed a value set for the EQ-5D-5L in the United Kingdom (UK), that assigns numerical values to health states that can calculate quality-adjusted life-years, which can inform evaluations of the cost-effectiveness of treatments. This study collected preference data from the UK public using time trade-off interviews, where participants indicated how many years of life they would trade to avoid impaired health. The research provides a robust, transparent scoring system reflecting current public values, which can be used to inform healthcare decision making in the United Kingdom.
How was the research conducted?
Researchers recruited 1200 representative UK adults from all 4 nations. Participants completed interviews between January and September 2023, with 1102 by videoconference using Zoom and 98 in-person to enhance inclusivity. Each participant valued 10 health states using time trade-off. This established method systematically collects preference data suitable for statistical modeling and has been successfully used to value EQ-5D-5L in over 30 countries.
What were the results?
The study generated a UK value set using a random effects Tobit model. Anxiety/depression and usual activities showed greater impact on health values than in the UK value set of EQ-5D-3L generated in the 1990s. The worst health state value is -0.567, which is comparable to the UK EQ-5D-3L worse state value of -0.594..
Why are the results important?
This value set allows utilities to be generated directly for the EQ-5D-5L using UK preferences. This is advantageous since the EQ-5D-5L has greater sensitivity in capturing health-related quality of life in patient populations than the EQ-5D-3L. These results can be used to inform policy, including health technology appraisals.
What are the strengths and weaknesses of this study?
The main strength is rigorous quality assurance including independent data quality assessment at 5 time points, comprehensive interviewer training, diverse stakeholder input, and offering a choice of interview via videoconference or in-person for representative sampling. One limitation is that the statistical model cannot simultaneously account for all data characteristics of the time trade-off data.
Note: This content was created with assistance from artificial intelligence (AI) and has been reviewed and edited by ISPOR staff. For more information or for inquiries on ISPOR’s AI policy, click here or contact us at info@ispor.org.