Equity weights for economic evaluation: An Australian Discrete Choice Experiment, CHERE Working Paper 2008/5

Background: Economic evaluation of healthcare interventions and technologies using the quality-adjusted life year (or life year) usually values outcomes independently of who they accrue to. This is a simplifying assumption relating to the more complex societal preferences. While the premise of equal value has been criticised as being unreflective of societal views, no alternative has gained significant traction. Aims: To identify the trade-offs made by an Australian population between total gain in life expectancy, initial life expectancy, gender, income and smoking status, and then to generate equity weights for economic evaluation from these results. Method: A discrete choice experiment was used in an online panel. 241 respondents answered twelve binary choices, and the results were analysed using logistic regression. Equity weights were then generated using Hicksian compensating variation. Results: A typical individual was willing to discriminate based on smoking and income, but not on gender or initial life expectancy (although the last of these is considered within a narrow range of 55-75 years). However, there was considerable heterogeneity in respondents. Equity weights ranged from 0.673 for smokers with an above average income to 1.207 for non-smokers with a below average income. This result was sensitive to the point at which the marginal utility of time was estimated. Conclusion: Healthcare decision making, using an orthodox QALY model, does not capture the views of society, particularly with regard to smoking or income. We have presented an alternative approach, weighting outcomes dependent on the personal characteristics of the individual receiving them. The feasibility of including this finding in economic evaluation is as yet uncertain and has to be investigated further.