Influence of construct-irrelevant factors and effects of methodological choices on EQ-5D health state valuation
Appears in the following Collection
AbstractEfficient allocation of health-care resources requires weighing of costs and benefits. Benefits are typically measured using quality-adjusted life years (QALYs), which combine health-related quality of life (HRQoL) and length of life in a single metric. To measure HRQoL in large groups of patients, the standard approach is to use specialized questionnaires asking patients to describe their health along a specified set of dimensions. QALY-values are typically assigned to health states by asking members of the general public to state their preferences for hypothetical health states. The overarching aim of this thesis was to investigate the influence of theoretically irrelevant factors and methodological choices on values assigned to health states described by the EQ-5D, the most widely used generic, preference-based HRQoL-questionnaire.
Methods used to determine preference-based values for health states, particularly the time trade-off (TTO) method, were found to be susceptible to influence from anchoring from the starting point of the task. Furthermore, respondents assign more extreme values as they gain experience with the valuation task. Theoretically equivalent variants of the valuation methods result in different elicited preference values, and different methods result in different estimates of the relative importance of specific dimensions of health. Attitudes toward death seem to influence values assigned to health states, and how death is implemented in valuation methods influences the distribution of derived values. The thesis concludes that methods used to assign values to health states described by the EQ-5D are influenced by several factors of disputable relevance, suggesting that the actual uncertainty around such values is under-communicated. Decision makers should use appropriate caution when interpreting results of studies involving the use of instruments like the EQ-5D.
LIST OF PAPERS
1: Kim Rand-Hendriksen, Liv Ariane Augestad, Fredrik Andreas Dahl, Ivar Sønbø Kristiansen, and Knut Stavem. A Shortcut to Mean-Based Time Tradeoff Tariffs for the EQ-5D? Medical Decision Making 32, no. 4 (2012): 569–577. DOI: 10.1177/0272989X11431607
2: L. A. Augestad, K. Rand-Hendriksen, I. S. Kristiansen, and K. Stavem. Impact of Transformation of Negative Values and Regression Models on Differences Between the UK and US EQ-5D Time Trade-Off Value Sets. PharmacoEconomics (2012). DOI: 10.2165/11595420-000000000-00000
3: Liv Ariane Augestad, Kim Rand-Hendriksen, Ivar Sønbø Kristiansen, and Knut Stavem. Learning Effects in Time Trade-Off Based Valuation of EQ-5D Health States. Value in Health: The Journal of the International Society for Pharmacoeconomics and Outcomes Research 15, no. 2 March 2012): 340–345. DOI: 10.1016/j.jval.2011.10.010
4: Kim Rand-Hendriksen and Liv Ariane Augestad. Time Trade-Off and Ranking Exercises Are Sensitive to Different Dimensions of EQ-5D Health States. Value in Health 15, no. 5 (2012): 777–782. DOI: 10.1016/j.jval.2012.04.002
5: Kim Rand-Hendriksen, Liv Ariane Augestad, Ivar Sønbø Kristiansen, and Knut Stavem. Comparison of Hypothetical and Experienced EQ-5D Valuations: Relative Weights of the Five Dimensions. Quality of Life Research 21, no. 6 (2012): 1005–1012. DOI: 10.1007/s11136-011-0016-3
6: Liv Ariane Augestad, Kim Rand-Hendriksen, K. Stavem, and I. S. Kristiansen. Time Trade-off and Attitudes Toward Euthanasia: Implications of Using ‘death’ as an Anchor in Health State Valuation.” Quality of Life Research. In Press (2012). Published online: 08 June 2012 DOI: 10.1007/s11136-012-0192-9
7: Samuelsen CH, Augestad LA, Stavem K, Kristiansen IS, Rand-Hendriksen K. Anchoring bias in the Lead-Time Time Time Trade-Off. Heuristic Influences on the Elicitation of Preferences for Hypothetical Health States. Submitted.