The impact of losses in income due to ill health: Does the EQ-5D reflect lost earnings?

C Tilling, M Krol, A Tsuchiya, J Brazier, J Exel van, W Brouwer


In order to allocate scarce health care resources, the National Institute for Health and Clinical Excellence (NICE) promotes the use of economic evaluation to decide which health care technologies should be recommended for use in the NHS. The ‘reference case’ in the current NICE Guide to the Methods of Technology Appraisal (2008) does not include any indirect or productivity costs in the evaluation, since the perspective used for costs is that of the NHS and Personal Social Services, (the same applies to the previous Guide; NICE, 2004). At the same time, the Guide aims to capture “all health effects on individuals” on the outcomes side. Thus, if people take into account the impact of lost earnings in the health valuation exercise, then indirect costs would be included in the analyses, albeit implicitly. On the other hand, the House of Commons Health Select Committee has recently recommended that “wider benefits and costs […] be more fully incorporated into NICE’s assessment” (Health Select Committee, 2007) so we could see future legislative changes that require productivity costs to be
incorporated more explicitly in future economic evaluations for NICE. Either way, two key questions in the context of UK health policy are: do the published preference indices for EQ-5D reflect the impact of lost earnings? Are we currently implicitly including indirect costs in our analyses? It is crucial to investigate whether
or not individuals take into account any possible impact of lost income in health state valuation exercises. If respondents do consider income effects, and these considerations change valuations, then these effects would need to be excluded both under the current NICE reference case, or where productivity costs are included in the numerator to avoid double counting. This study adapts the study design used to generate population value sets for EQ-5D, as first used in the Measurement and Valuation of Health (MVH) Study (Dolan, 1997), and carries out valuations of hypothetical EQ-5D states using Time Trade Off (TTO) exercises through an online
survey administered in the Netherlands. Furthermore, this study uses a number of different TTO questions to explore the impact of losses in income on the valuation of hypothetical health states, and to determine the relationship between income and health. To understand the effect that income considerations may have in health state valuation exercises it is necessary to understand the relative importance of health and income when valued both simultaneously and independently. For example, would the same loss of income be valued as worse when it is associated with worse health states? Specifically our objectives are to (a) examine whether EQ-5D health state values, obtained through online TTO, reflect losses in income due to ill health; (b) examine
the impact of including specific ex-post instructions to consider, or not to consider, income changes when hypothetical EQ-5D states are valued, on the health state values; (c) examine how the above impact is distributed across the different dimensions of EQ-5D, and (d) explore the possible interactions between health and income in health state valuation.