Outcomes: part II Emma Frew Introduction to health economics, MSc HEHP, October 2012

Obtaining QoL values for QALYs Value judgement Search literature for published values Measure values Direct valuation – by patients Visual analogue Standard Gamble Time trade-off Indirect valuation – by patients, public, others Using standard tariffs for QoL instruments Using direct valuation methods with scenarios

Visual analogue scale Many variants ‘Thermometer’ scale is the one mainly used. 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 There are many variants for the rating scale, the scale can have numbers (e.g. 0-100), categories (e.g. 0-10) or just consist of a 10cm line on a page. The most commonly used is the ‘visual analogue scale’ and this consists of a line on a page, usually 10cm in length, with clearly defined endpoints with or without marks along the line. Here the clearly defined endpts are the best imaginable health state coded as 1 and the worst coded as 0 (this doesn’t always refer to death). Important to stress that the distance between the states is as important as the order AND should correspond with relative differences in preference. 1 Worst imaginable health state Best imaginable health state

Standard Gamble probability p HEALTHY Choice 1 probability 1-p DEAD The classical method for measuring cardinal prefs is the SG technique, this is known as the ‘gold standard’ as it is based on axioms of U theory. Subjects are asked to make trades or to consider aspects of uncertainty that is consistent with what happens in real-life. Subjects are asked what risk of immediate death they would be WTA to escape from a specified state of poor health. Obviously the bigger acceptable risk, the worse the poor health state. The subject is offered 2 choices. Choice 1: treatment with 2 outcomes. Outcome 1 - patient returns to normal health and lives t years with a prob = p. Outcome 2 - patient dies immediately with a prob = 1-p. Choice 2: certain outcome to stay in that chronic state i for t years. The probability p is varied until respondent is indifferent btw choice 1 and 2. Choice 2 STATE i

Standard Gamble Probability p = QoL measure Advantages: Based on axioms of utility theory Disadvantages: Not many chronic diseases that approximate gamble Subjects may find concept of probability difficult to understand The prob p is = QOL measure, e.g. 60% chance of wellness and a 40% chance of death, p=0.6 advs: based on axioms of U theory. Disadvs: treatment of most chronic diseases do not approximate a gamble, eg. no product that will make the patient better nor is there one that is likely to lead to immediate death. Subjects find the concept of prob difficult to understand.

Time Trade-Off VALUE Alternative 1 Healthy 1.0 Alternative 2 hi The alternative measure to elicit QALYs is the TTO technique. This is meant to be simpler for subjects to understand. It introduces trade off with time. The idea is that subjects are offered 2 alts: I) chronic condition for time t followed by immediate death. Ii) OR healthy for time x which is less than t followed by death. It is assumed that the subject will choose the state of wellness over the state of illness. Time x is varied until the resp is indiff between the 2 alts. The QOL measure is calculated as time x/time t. e.g. a subject may rate being in a wheelchair for 2 yrs as equiv to perfect wellness for 1yr. If this is so, 1/2 = 0.5 (no of years in full health/no of yrs in chronic state) dead TIME x t

Measuring outcomes: exercises 2 & 3

Challenges with QALYs: QoL measures Validity – does the instrument accurately measure what it is supposed to measure? Reliability – do you consistently obtains the same results using the instrument? Sensitivity to change – can the instrument measure (clinically important?) change? Feasibility of use – can the instrument be easily used with the population of interest?

The relationship between validity and reliability

Validity of quality of life questionnaires No gold standard measure of health to compare EQ-5D to. Accumulate evidence over a range of aspects of validity: Content validity: sufficient items and coverage? Construct validity: anticipated relationships with other variables (e.g. disability, age, long standing illness are as anticipated)? Convergent validity: correlates with other measures of same phenomenon?

Challenges with QALYs: QoL measures Validity – does the instrument accurately measure what it is supposed to measure? Reliability – do you consistently obtains the same results using the instrument? Sensitivity to change – can the instrument measure (clinically important?) change? Feasibility of use – can the instrument be easily used with the population of interest?

Potential trade-off between sensitivity and feasibility Hydrodensitometry Skin fold measure

Sensitivity and feasibility of quality of life questionnaires “The EQ-5D …[is] more responsive than any of the other measures, except pain and doctor-assessed disease activity” [Hurst et al. (1997) Brit. J. of Rheum.] “The weighted TTO-score of EuroQoL-5D, … did however not correspond with these [reduced psychotic symptoms] changes, which indicates that it is less sensitive to changes in social and psychological well-being.” [van de Willige et al. (2005) Qual. Life Res]. Feasibility: Patient burden (EQ-5D has 5 questions each with 3 possible responses). Valuation burden (EQ-5D has 243 possible health state permutations/ SF-6D has 18,000 possible permutations).

Challenges with QALYs: theory Assumes health status can be measured on a cardinal scale Assumes it is possible to equate x years in less than full health with y years in full health, where y<x Assumes can compare utility scores across individuals Possible to equate same state if one person deteriorating and the other improving (independence assumption) MU of health is constant, i.e. 2 QALYs to 1 person is equivalent to 1 QALY each to 2 people

Challenges with QALYs: methodology Possible to equate to death when death is unknown Different methods lead to different values Description of alternatives leads to different values Values creep towards 1 as health deteriorates with age Values differ depending upon the duration of the state Framing effects

Challenges with QALYs: ethical Life saving should always be a priority? Ageist? Attributes greater importance to maximising health than how that health is distributed Potential for discrimination? “Double jeopardy”?

Measuring outcomes: discussion

Selected reading TEXTS Drummond M, Sculpher M, Torrance G, O'Brien B, Stoddart G. Methods for the Economic Evaluation of Health Care Programmes. 3rd ed. Oxford: Oxford University Press; 2005. Chapter 6. Morris S, Devlin N, Parkin D. Economic analysis in health care. Chichester, UK: John Wiley & Sons, Ltd; 2007. Chapter 10. Brazier J, Ratcliffe J, Salomon J, Tsuchiya A. Measuring and valuing health benefits for economic evaluation. Oxford: Oxford University Press; 2007. EARLY REFERENCES TO QALY METHODOLOGY Williams A. Economics of coronary artery bypass grafting. British Medical Journal 1985; 291:326-329. Klarman H, Francis J, Rosenthal G. Cost-effectiveness analysis applied to the treatment of chronic renal disease. Medical Care 1966; 6(1):48-54. Torrance G. Measurement of health state utilities for economic appraisal. Journal of Health Economics 1986; 5:1-30.

Selected reading II OUTCOME MEASURES AND VALUATION OF HEALTH STATES Brazier J, Roberts J, Deverill M. The estimation of a preference-based measure of health from the SF-36. Journal of Health Economics 2002; 21:271-292. Brooks R. EuroQol: the current state of play. Health Policy 1996; 37:53-72. Richardson J. Cost Utility Analysis: What Should Be Measured? Social Science and Medicine 1994; 39(1):7-21. Robinson A, Dolan P, Williams A. Valuing health status using VAS and TTO: what lies behind the numbers? Social Science and Medicine 1997; 45(8):1289-1297. Dolan P, Gudex C, Kind P, Williams A. The time trade-off method: results from a general population study. Health Economics 1996; 5(2):141-154. VALIDITY AND ETHICS OF QALY METHODLOGY Brazier J, Deverill M. A checklist for judging preference-based measures of health related quality of life: learning from psychometrics. Health Economics 1999; 8:41-51. Loomes G, McKenzie L. The use of QALYs in health care decision making. Social Science and Medicine 1989; 28(4):299-308. Harris J. QALYfying the value of life. Journal of Medical Ethics 1987; 13:117-123.