Searching for a QALY threshold range: Some research based policy lessons

Transcription

1 Symposium: The value of health: What is the threshold Erasmus University, Netherlands, Searching for a QALY threshold range: Some research based policy lessons Jan Abel Olsen University of Tromsø, Norway and Monash University, Australia

2 Is there a Norwegian threshold? No Yes, well, sort of NOK 500,000 = 66,666 (GDP/capita NOK 547,000)

3 Job s suggestions your perspective on the importance of research into the value of health for health care decision making; More important to focus on convincing documentation of outcomes measured in health terms, than to value health outcomes in money terms some highlights of research in this area that you have done in the past or are currently involved in; Framing: absolute WTP-values can easily be manipulated Context: relative values differ due to true preferences some of the problems you have encountered in determining theoretically valid and practically meaningful values of health; There is no such thing as one valid value of health What is practically meaningful depends on what is politically acceptable your ideas about the research agenda in this area. The fallacy of the ratio: We have non-linear preferences for the numerator as well as the denominator Identify which programme characteristics make us willing to accept lower vs higher cost/qaly

4 The logic of one treshold value Policy objectives Single objective: health maximization only No concerns for equity Preferences Individual Linear over gains, risk neutral Societal: Neutral to contextual differences Other health streams Recipient characteristics

5 Some lessons from own (and others ) research 1: Framing WTP sensitive to framing WTP is sensitive to theoretically irrelevant factors and insensitive to theoretically relevant factors Sensitive to the opening bid, or whichever numerical anchor Insensitive to the size of the good WTP/unit of health gain depends on the question format Don t trust the absolute value Unfortunately, framing effects can be seen in all sorts of preference elicitation methodologies, not only in WTP

6 Behavioural economics Any number can be an anchor for WTP Unrelated numbers an experiment: - Write down the last 2 digits [##] in your social security number - Would you have been willing to pay ## for good X? Yes / No - What is your max WTP for good X? - Result: WTP depends on [##]!!!

7 Some lessons from own (and others ) research 2: Context Other streams of health matter Prospective health; end-of-life treatments 10 9 = 2 1, but V(10-9) < V(2-1) Past health = 30 29, but V(80-79) < V(30-29) Recipient characteristics matter Causes Individual responsibility Social avoidability Consequences Others wellbeing Productivity gains A QALY is not a QALY is not a QALY

8 The value of a health gain depends on the sizes of other health streams Age No-treatment profile Gain Peter Harry

9 Some lessons from own (and others ) research 3: Perspective Different perspectives in preference elicitation surveys Individual utility of health gains Social value of health gains Consumer vs citizen Preferences stated qua individuals are not neccessarily transferable to a societal decision making context Perspective Link preferences to the real policy context How would you like to see health care decision makers choose? Rather than: How much are you willing to pay?

10 i.e. In our roles as (private) individuals and (social) citizens, we value a health gain unit differently depending on a range of context specific characteristics Which of these characteristics would we wish decision makers to take into account, and in which ways? Estimate precise weights? For the purpose of an equity weighted Super QALY One threshold value: cost/weighted QALY Discretionary judgments? Unweighted QALYs + describe equity impacts Threshold range: cost range/unweighted QALY

11 The fallacy of the ratio Cost/QALY: 50,000/1 = 500/0.01 = 500,000/10 The size of the numerator determines the insurance motive Low cost treatments have no insurance motive The size of the denominator: Non-linear utility S-shaped Increasing marginal utility of small increases in outcomes Diminishing MU of increases on top of large outcomes

12 For which reason should which characteristics drive the range? Non-linear preferences for the numerator and the denominator The size of the QALY gain matter Reduce inequalities in lifetime health Patients age matter End-of-life concerns The no-treatment profile matter Avoidable inequalities Social deprivation should be compensated Equal health opportunities Responsibility for own health

13 Suggested criteria for a cost/qaly threshold range Characteristics Low threshold High threshold The size of the QALY gain < 0.03 > 0.25 Total lifetime health (age) > Life expectancy < 60 Remaining QALYs without treatment > 15 < 2 Social inequalities (avoidability) Deprived Equal health opportunities (responsibility) Well informed unhealthy behaviour

14 Conclusion Framing Don t trust the absolute WTP/QALY value Context Trust that there are relative differences in WTP/QALY depending on many contextual characteristics Which contextual characteristic should be taken into account depends on: The strength of the preference for each characteristic The health policy acceptability of applying weights The alternative to measuring characteristics weights for a super-qaly is to describe them and apply a wide cost/qaly threshold range

15 Own research in this field Olsen JA, Røgeberg OJ, Stavem K: What explains willingness-to-pay for smoking-cessation treatments: Addiction level, quit rate effectiveness or the opening bid? Applied Health Economics and Health Policy, 2012 Olsen JA, Richardson J: Preferences for the normative basis of health care priority setting: Some evidence from two countries. Health Economics (in press) 2012 Desser AS, Gyrd-Hansen D, Olsen JA, Grepperud S, Kristiansen IS: Societal views on orphan drugs: crosssectional survey of Norwegians aged 40 to 67. British Medical Journal 2010: 341 Kvamme MK, Gyrd-Hansen D, Olsen JA, Kristiansen IS: Increasing marginal utility of small increases in lifeexpectancy? Results from a population survey Journal of Health Economics, 2010; 29: Olsen JA, Donaldson C, Shackley P. Implicit vs explicit ranking: On inferring ordinal preferences for health care programmes based on differences in willingness-to-pay. Journal of Health Economics, 2005; 24 (5): Olsen JA, Kidholm K, Donaldson C, Shackley P. Willingness to pay for public health care: a comparison of two approaches. Health Policy 2004, 70: Olsen JA, Donaldson C, Pereira J. The insensitivity of willingness-to-pay to the size of the good: new evidence for health care, Journal of Economic Psychology, 2004; 25: Olsen JA, Richardson J, Dolan P, Menzel P. The moral relevance of personal characteristics in setting health care priorities, Social Science and Medicine, 2003; 57: Dolan P, Olsen JA, Richardson J, Menzel P. An inquiry into the perspectives that can be used when eliciting preferences in health, Health Economics, 2003; 12: Dolan P, Olsen JA. Equity in health: the importance of different health streams, Journal of Health Economics, 2001; 20: Olsen JA, Smith R. Theory versus practice: A review of willingness-to-pay in health and health care, Health Economics, 2001; 10: Olsen JA. A note on eliciting distributive preferences for health, Journal of Health Economics, 2000; 19: Olsen JA, Donaldson C. Helicopters, hearts and hips: Using willingness to pay to set priori-ties for public sector health care programmes, Social Science and Medicine, 1998; 46: Olsen JA: Aiding priority setting in health care: Is there a role for the Contingent Valuation method? Health Economics, 1997; 6: Olsen JA: Theories of justice and their implications for priority setting in health care, Journal of Health Economics, 1997; 16: Olsen JA: Persons vs years: Two ways of eliciting implicit weights, Health Economics, 1994; 3: