Weighting and valuing quality-adjusted life-years using stated preference methods: preliminary results from the Social Value of a QALY Project.

Objectives: To identify characteristics of beneficiaries of health care over which relative weights should be derived and to estimate relative weights to be attached to health gains according to characteristics of recipients of these gains (relativities study); and to assess the feasibility of estimating a willingness-to-pay (WTP)-based value of a quality-adjusted life-year (QALY) (valuation study).

Design: Two interview-based surveys were administered - one (for the relativities study) to a nationally representative sample of the population in England and the other (for the valuation study) to a smaller convenience sample.

Setting: The two surveys were administered by the National Centre for Social Research (NatCen) in respondents' homes.

Participants: 587 members of the public were interviewed for the relativities study and 409 for the valuation study.

Methods: In the relativities study, in-depth qualitative work and considerations of policy relevance resulted in the identification of age and severity of illness as relevant characteristics. Scenarios reflecting these, along with additional components reflecting gains in QALYs, were presented to respondents in a series of pairwise choices using two types of question: discrete choice and matching. These questions were part of a longer questionnaire (including attitudinal and sociodemographic questions), which was administered face to face using a computer-assisted personal interview. In the valuation study, respondents were asked about their WTP to avoid/prevent different durations of headache or stomach illness and to value these states on a scale (death = 0; full health = 1) using standard gamble (SG) questions.

Results: Discrete choice results showed that age and severity variables did not have a strong impact on respondents' choices over and above the health (QALY) gains presented. In contrast, matching showed age and severity impacts to be strong: depending on method of aggregation, gains to some groups were weighted three to four times more highly than gains to others. Results from the WTP and SG questions were combined in different ways to arrive at values of a QALY. These vary from values which are in the vicinity of the current National Institute for Health and Clinical Excellence (NICE) threshold to extremely high values.

Conclusions: With respect to relative weights, more research is required to explore methodological differences with respect to age and severity weighting. On valuation, there are particular issues concerning the extent to which 'noise' and 'error' in people's responses might generate extreme and unreliable figures. Methods of aggregation and measures of central tendency were issues in both weighting and valuation procedures and require further exploration.