Measuring health-related quality of life in multiple sclerosis: comparing the acceptability, validity and responsiveness of the EQ-5D-3L and MSIS-8D

Concerns have been raised about the sensitivity and responsiveness of the EQ-5D, one of the most commonly used preference-based health-related quality of life measures, in the context of multiple sclerosis (MS). In response to these concerns, a condition-specific preference-based measure, the Multiple Sclerosis Impact Scale Eight Dimensions (MSIS-8D), was developed. This research aimed to assess the psychometric and distributional properties of the MSIS-8D compared to the EQ-5D-3L, in people with MS. Analyses were undertaken using data from the UK MS Register. Both measures were compared in terms of acceptability (missing data), distributional properties (health state frequencies, health state density curves and indices), construct validity in relation to disability, mobility, fatigue, anxiety and depression (discriminative and convergent validity, using ANOVA, independent t-tests and Spearman correlations), and responsiveness to symptom onset and relapse (mean change scores, standardised response means, standardised effect sizes, paired t-tests). The MSIS-8D exhibited superior distributional properties, while the EQ-5D-3L showed greater acceptability. Both measures demonstrated excellent construct validity. Neither measure appeared responsive to symptom onset, and only the MSIS-8D met all criteria for responsiveness when people moved from a non-relapse to a relapse state. Although the MSIS-8D appears to offer superior distributional properties and responsiveness compared to the EQ-5D-3L, the responsiveness of both measures in this analysis was limited. This adds weight to existing concerns about the ability of utility measures used in healthcare decision-making to fully capture treatment effects in MS. In the UK, healthcare treatments must be shown to be cost-effective in order to be provided on the NHS. When assessing cost-effectiveness, researchers usually measure the effects of treatments on people’s health using a questionnaire called the EQ-5D. However, concerns have been raised about the ability of the EQ-5D to capture the effects of treatments for multiple sclerosis. This could affect decisions about which new treatments are made available to people with multiple sclerosis. Another questionnaire, called the MSIS-8D, has been developed specifically to measure the effects of treatments on the health of people with multiple sclerosis. Evidence is needed to assess whether the MSIS-8D is better able to capture these treatment effects in people with multiple sclerosis, compared to the EQ-5D. In this study, we used data collected by the UK Multiple Sclerosis Register to compare the EQ-5D and the MSIS-8D. We compared their relevance to people with multiple sclerosis, their ability to distinguish between groups of people with more or less severe symptoms, and their ability to capture changes in people’s health over time. The results indicate that the MSIS-8D is more relevant to people with multiple sclerosis and better able to capture changes in people’s health over time, compared to the EQ-5D. However, both the EQ-5D and the MSIS-8D showed limited ability to capture changes over time. This suggests that more research is required to understand how the effects of treatments for multiple sclerosis can be better measured, and fully represented in assessments of cost-effectiveness.