Statistical Analysis of EQ-5D Profiles: Does the Use of Value Sets Bias Inference?

Abstract

SummaryHealth state profile data, such as those provided by the EQ-5D, are widely collected in clinical trials, population surveys and a growing range of other important health sector applications. However, these profile data are difficult to summarise to give an overall view of the health of a given population that can be analysed for differences between groups or within groups over time. A common way of short-cutting this problem is to transform profiles into a single number, or index, using sets of weights, often elicited from the general public in the form of values. Are there any problems with this procedure? In this paper we demonstrate the underlying effects of the use of value sets as a means of weighting profile data. We show that any set of weights introduces an exogenous source of variance to health profile data. These can distort findings about the significance of changes in health between groups or over time. No set of weights is neutral its effect. If a summary of patient reported outcomes is required, it may be better to use an instrument that yields this directly -such as the EQ VAS -along with the descriptive instrument. If this is not possible, researchers should have a clear rationale for their choice of weights; and be aware that those weighs may exert a non-trivial effect on their analysis. This paper focuses on the EQ-5D, but the arguments and their implications for statistical analysis are relevant to all health state descriptive systems.

Variations in outcome and costs among NHS providers for common surgical procedures: econometric analyses of routinely collected data

“…patient choice, accountability, identification of best practice) and who decides the relative weighting of each component and objective. 12,66 Our study does not intend to resolve this debate. Rather, we present a means of making inferences about hospital quality and presenting results when health outcomes are assessed through the EQ-5D instrument.…”

“…12 In some circumstances, one may be willing to accept the value judgement implied or explicitly expressed by the aggregation function. For example, the convention of using the preferences of the general public to aggregate EQ-5D profiles has a clearly articulated rationale in the context of cost-effectiveness analysis and decisions concerning the allocation of taxpayer funding.…”

Patient‐reported outcome measures in the NHS: new methods for analysing and reporting EQ‐5D data

“…More importantly, this process also introduces an exogenous source of variance into the data, unrelated to the underlying levels of and changes in health reported by patients. Specifically, the statistical significance of any changes in health might be affected by the particular set of values chosen (Parkin et al, 2010). There is therefore a risk of concluding that a significant change has occurred in a given direction -an improvement or deterioration in health -and with a given magnitude, when that result is contingent upon the properties of the particular value set used.…”

Estimation and Comparison of EQ-5D Health States' Utility Weights for Pneumoccocal and Human Papillomavirus Diseases in Argentina, Chile, and the United Kingdom

“…There are studies that reported that the use of anchoring vignettes successfully improved the comparability of self reported measures [28]. Newer studies also underline the problems of using value sets to weight profile data as EQ-5D derived health states and that caution should be taken when choosing a summary measure [29,30].…”