Does the QALY reflect lost earnings due to ill health? Evidence from a Dutch online survey using the EQ-5D

Presenter: Carl Tilling, University of Sheffield

Abstract

BACKGROUND: Cost-Effectiveness Analysis is now the most common method of economic evaluation in healthcare, and this method uses the Quality Adjusted Life Year (QALY) as the benefit measure. Productivity costs are the costs to society of sick individuals being out of work. Initially, in countries where they were deemed relevant for inclusion, productivity costs were calculated in monetary terms and included in the numerator of the cost-effectiveness ratio. However, in 1996 The US Panel on Cost Effectiveness in Health and Medicine (known as the Washington Panel) recommended against this practice, arguing that productivity costs are included in the QALY measure. In other words, the Panel made the assumption that respondents in health state valuation exercises take income losses into account, and these income losses act as a proxy for productivity costs. While many questioned this assumption it could not be countered since there was no evidence regarding what people consider in these exercises. If respondents in health state valuation exercises do consider income losses, and if this changes health state valuations, then this will mean that all past economic evaluations that have included productivity costs in the numerator may have double counted these costs. Alternatively, if respondents do not consider income losses then all past economic evaluations that have not included productivity costs in the numerator (the most common approach) have failed to account for sizeable societal costs. Either way, determining what respondents consider in health state valuation exercises is vital for the development of an appropriate standardised framework for economic evaluations, so they produce consistent and comparable results.

AIMS: To determine what people consider when valuing EQ-5D health states in Time Trade Off (TTO) exercises of hypothetical health states; and to develop a greater understanding of the relative importance of health and income effects in these exercises, and the nature of any interactions between these two effects.

METHODS: This is based on an online survey in the Netherlands, presented in Dutch, to 300 members of the general public. Respondents are asked to value four hypothetical EQ-5D states through numerous different TTO exercises. TTO 1 is the standard TTO question and is succeeded by a follow-up question asking whether or not they considered income effects. TTO 2 repeats TTO 1 but asks respondents to include/exclude income effects depending upon their response to the follow up question to TTO 1. TTO 3 explicitly tells respondents how much income they will lose in the given health state. TTO’s 4 and 5 are income gain/loss questions respectively, in which there is no mention of health at all. These TTO questions in different combinations will enable us to test a number of hypotheses: whether people take income considerations into account when there is no mention of income, whether valuations of those that do and do not spoantaneoulsy include income effects differ, and whether those that do not spontaneously include income effects alter their valuations when asked to include these effects.

RESULTS AND CONCLUSION: Data are currently being collected, and are due in early November.