RESUMO
BACKGROUND: In trials incorporating a health economic evaluation component, reliable validated measures for health-related quality of life (HRQOL) are essential. The EQ-5D is the preferred measure for cost-effectiveness analysis in UK trials. This paper presents a qualitative evaluation of the use of the EQ-5D-3L in a feasibility randomised control trial with participants who had a mild- to moderate learning disability and type 2 diabetes. METHODS: Researchers administered the EQ-5D-3L to 82 participants at baseline and 77 at follow-up. After each interview, researchers rated the ease of administering the EQ-5D-3L and made free-text entries on the administration experience. For a subset of 16 interviews, researchers audio-recorded more detailed journal entries. Ease of administration data were analysed using descriptive statistics. Free-text responses were subject to a basic content analysis. The EQ-5D-3L-related journal entries were transcribed, coded and analysed thematically. RESULTS: Over half of participants were perceived to experience difficulty answering some or all of the items in the EQ-5D-3L (60% at baseline; 54% at follow-up). Analysis of the free-text entries and audio journals identified four themes that question the use of the EQ-5D-3L in this population. The first theme is related to observations of participant intellectual ability and difficulties, for example, in understanding the wording of the measure. Theme 2 is related to the normalisation of adjustments for impairments, which rendered the measure less sensitive in this population. Theme 3 is related to researcher adaptation and non-standard administration. An overarching fourth theme was identified in that people with learning disabilities were viewed as 'unreliable witnesses' by both researchers and supporters. CONCLUSIONS: It is recommended that the EQ-5D-3L should not be used in isolation to assess health-related quality of life outcomes in trials research in adults with a learning disability. Further research is required to develop and evaluate a version of the EQ-5D appropriate for this population in trials research. It is unrealistic to expect that adjustments to the wording alone will deliver an appropriate measure: supporter or researcher involvement will almost always be required. This requirement needs to be factored into the development and administration guidelines of any new version of the EQ-5D for adults with a learning disability. TRIAL REGISTRATION: Current Controlled Trials ISRCTN41897033 [registered 21 January 2013].
