National Sun Yat-sen University Institutional Repository:Item 987654321/33488

行政院國家科學委員會專題研究計畫 成果報告

成本效能分析中應用 QALYs 與 DAYLs 所造成的差異比較─以

急性腦中風治療為例

中 華 民 國 95 年 10 月 31 日

成本效能分析中應用 QALYs 與 DAYLs 所造成的差異比較─以急性腦中風治療為例

中文摘要

有限的醫療資源往往迫使我們在面對醫療資源分配的決策時，不得不考量相關醫療處置 之成本與其預期之效能。因此我們不難理解，為有效率地運用醫療資源，經濟評估與醫 療技術評估的需求日益殷切。現階段主要歐美國家都以所謂「健康調整後之生命年數 (health-adjusted life years) 」來進行醫療介入(interventions)的成本效能分析，或用以估算

疾病負擔。「健康調整後之生命年數」本質上是個集合名詞，在其眾多指標中，以品質

調整後之生命年數(quality-adjust life years, 簡稱 QALYs)與失能調整後之生命年數 (disability-adjust life years, 簡稱 DALYs)最常見。本研究藉由探討腦中風的成本效能分 析，比較使用 DALYs 與 QALYs 所可能造成的評估差異，填補腦中風醫療管理與相關 經濟評估研究之不足。鑑於 QALYs 的效用衡量或 DALYs 的失能衡量缺乏共識，成本 效能研究中宜清楚說明這些指標的衡量的方法，同時避免不同基礎的不當比較。 關鍵詞：成本效能研究、品質調整後之生命年數、失能調整後之生命年數、腦中風

Abstract

Limits on health care resources mandate that resource-allocation decisions be guided by consideration of cost in relation to the associated expected benefits. It has been leading to an increased demand for economic evaluation and medical technology assessment.

Nowadays, cost-effectiveness analysis (CEA) has emerged as a favored analytic technique for economic evaluation in health care. Currently, the leading standard in Western countries is the use of health-adjusted life years (HAYLs) in CEAs of medical care and health

interventions and for estimating burden of a disease. Being an umbrella term of a family of measures, HALYs includes quality-adjusted life year (QALYs) and disability-adjusted life years (DALYs). In this study we give a brief review on QALYs and DALYs, and examine variations in QALYs and DALYs estimates currently used in stroke studies. Recognizing the lack of reliable and valid measurement methods in eliciting utility values in QALYs or disability values in DALYs, CEA needs to be addressed in a clear and transparent way to ensure its proper use in practice.

Keywords: cost-effectiveness analysis, quality-adjusted life years (QALYs), disability-adjusted life years (DALYs), ischemic stroke

Introduction

Limits on health care resources naturally mandate that resource-allocation decisions be guided by considerations of cost in relation to expected benefits. It therefore leads to an increased demand for economic evaluation and medical technology assessment. There are some major kinds of economic evaluation in health and medicine [1,2]. Among them, cost-effectiveness analysis (CEA) has emerged as a favored analytic technique. While the National Health Insurance (NHI) system in Taiwan is facing a severe financial crisis, it is not surprising to find that CEA would attract more and more attention in health care

decision-making.

In CEA, the resources (costs) used and the health outcomes (effects) associated with each specific intervention (such as drugs, devices, procedures, and programs) are estimated and compared to show the tradeoffs involved in choosing among interventions. Health outcomes have long been evaluated by mortality-based indicators such as life expectancy, all-cause and disease-specific mortality. However, these mortality-based rates are

insufficient for evaluating disease burden, health outcomes, or the comparative impact of an intervention because the burden of chronic disease, injury, and disability among survivors are unrecorded. Moreover, to assess relative worth in different health interventions across diverse diseases, a standard measurement is required.

Nowadays, the leading standard in Western countries is the use of health-adjusted life years (HALYs) in CEA of medical care and health interventions and for estimating burden of disease. Theoretically, HALYs are summary measures of population health that merge estimates of life expectancy with measures of the quality of life with a disease. Being an umbrella term of a family of measures, HALYs include quality-adjusted life years (QALYs) and disability-adjusted life years (DALYs). Of note is that the particular type of CEA that uses QALYs or DALYs is sometimes referred to as cost-utility analysis (CUA) [1].

Following some authors [2], we make no distinction between CEA and its variant, CUA, throughout our discussion. We simply describe these analyses as CEA with QALYs or DALYs when they are used as the measure of effectiveness. Meanwhile, we use the terms ‘utility’,‘value’,and ‘preference’interchangeably,although in fact there are different [1].

The purpose of this article is to give a brief review on QALYs and DALYs, and to examine variations in QALYs and DALYs estimates currently used in stroke studies.

The Concept of QALYs

As a common metric for health, QALYs combine information regarding both length and quality of life. Every QALY is equivalent to one year of life in full health. For example, if major stroke is assumed to have a quality of life 0.4, then for a patient suffering a major stroke, each year spent in that state of health is considered to be 0.4 QALY, and 8 years lived after a major stroke would be counted as 3.2 (8×0.4) QALYs. Thus, the number of QALYs gained is the arithmetic product of the additional life expectancy gained owing to the health care intervention and the quality of these years.

In CEA, information regarding additional life expectancy gained after a disease, usually measured from clinical trials or epidemiological data, is relatively easy to obtain. Assessing the quality of that additional life, however, is more difficult. The preference or utility values for different health states, also called quality-of-life (QoL) weights, have to be assessed with appropriate measurement instruments. Such instruments include, for example, standard gamble (SG), time trade-off (TTO), and direct rating (DR) methods (including category rating and visual analogue scales) [1,2]. These are the three most widely used methods to measure directly the preferences of individuals for health outcomes. An alternative that is also widely used is to use any of the multi-attribute health status classification systems such as Health Utilities Index (HUI), Quality of Well-Being (QWB), or the European Quality of

Life Questionnaire (EuroQoL).

In general, utility values range from 0 for death to 1 for perfect health. Negative utility values are possible for some health states. For example, some people may consider being in a permanent vegetative state less preferable than death, and thus the health state would be given a negative utility value.

QALYs were developed in the late 1960s, primarily for use in CEA. A CEA of chronic renal disease is among the earliest [3]. After much deliberation, the US Public Health Service Panel on Cost-Effectiveness in Health and Medicine, as well as other consensus groups, concludes that a QALY measure is required for generalizable cost-effectiveness calculations [2]. However, different elicitation methods (and perspectives of the evaluators) tend to yield different preference or utility values from the same respondents for the same health states, and there is little consensus on which is to be preferred. For a more detailed background on QALY, see for example [1,2,4,5].

The Concept of DALYs

DALYs were first introduced in the World Development Report (World Bank 1993) as a method for estimating the global burden of disease, and as an outcome measure for use in CEA [6,7]. DALYs are the sum of the present value of years of lifetime lost through premature mortality and the present value of years of lifetime adjusted for the average severity of disability (mental or physical) caused by a disease or injury. That is to say, DALYs are the sum of years of life lost (YLLs) and years of life lived with disability (YLDs). It is noted that DALYs are a measure of something‘lost’ratherthan ‘gained’.

With the DALY approach, each state of health is assigned a disability weighting factor on a scale from 0 (perfect health) to 1 (death) by an expert panel. Future burdens are discounted (usually at a rate of 3% per year), and years of life in childhood and old age are

counted less. Specifically, the general equation for YLLs is as follows: YLL = ₂[ (r )(L a)[ ( )( ) 1] (r )a[ ( ) 1]] 1 (1 rL) ra e r K a r e a L r e r KCe    _{    }  _{   }  _     

where a is the age at death, L the remaining standard life expectancy at death, r the discount rate, βthe parameter for the age-weighting factor, K the age-weighting factor, and C a constant (0.1658).

The formula for YLD is basically the same except for that it is multiplied with a numeric disability weight (D), and that a is the age of onset of disability and L is the duration of disability: YLD =                         ) 1 ( 1 ]] 1 ) ( [ ] 1 ) )( ( [ [ ( )( ) ( ) 2 rL a r a L r ra e r K a r e a L r e r KCe D     

Just like the increasing awareness that valuations for QALYs may differ when the QALYs accrue to different patients or are evaluated at different times, there are concerns about the ethics and validity of DALYs. Particularly, the DALY framework not only is

health-weighted, but also contains a weighting of life years that accentuates the dependency of the very young and the very old. This unique property of DALY has naturally caused some controversy. On the other hand, however, the assumption that all QALYs are of equal value no matter who gains them or when they occur during the life span is just equally controversial [2, 8].

Quality of Life after Stroke

not surprised to find different studies report different utility values after stroke. In their comprehensive review of all QoL estimates for stroke appearing in the peer-reviewed

literature between 1985 and 2000, Tengs et al [9] found that QoL estimates range from -0.02 (indicating that major stroke is worse than death) to 0.71 for major stroke, from 0.12 to 0.81 for moderate stroke, from 0.45 to 0.92 for minor stroke, and from 0.29 to 0.903 for general stroke. Possible reasons for this variation include the aforementioned different utility assessment methods, different types of respondents (patients, general population, or

clinicians), and the defined bounds of the scale (1 corresponds to perfect health or merely the absence of stroke).

Post et al [10], a similar review published closely after aforementioned review, found that patients at risk for stroke assigned utilities of 0.26 and 0.55 to major (Rankin Scale 4 to 5) and minor (Rankin Scale 2 to 3) stroke, respectively. Stroke survivors assigned relatively higher utilities, 0.32 for major stroke and 0.71 for minor stroke.

Meta-analyses of QoL studies are rare in general. A meta-analysis of QoL estimated for stroke [11] reported utility values of 0.87 for minor stroke, 0.68 for moderate stroke, and 0.52 for major stroke if the TTO method is used and respondents are community members when the scale bounds range from death to perfect health. Furthermore, the same study found that the utility values are sensitive to severity of stroke and the bounds of the scale, but not to the elicitation methods or who the respondents are.

Since the publication of the results of NINDS rtPA Stroke Trial in December 1995, a number of comprehensive CEA of rtPA therapy for acute ischemic stroke have been

conducted [12-14]. The utility values used in theses studies vary greatly [15]. In a recent CEA of rtPA [16], a survey using the EuroQoL with a random sample of 100 first-ever stroke patients alive 1 year after discharge found the mean utility values were 0.22 for disabled (Barthel Index <95) patients and 0.77 for autonomous (BarthelIndex ≥95). Among the

morerecentstudiesreporting utility valuesafterstroke,O’Brien and Gage[17] use 0.75, 0.39 and 0.12, respectively, as QoL estimates for mild residua, moderate to severe residua, and residua from recurrent stroke. Their utility estimates were based on a study [18] published almost 10 years ago. In the Duke Stroke Policy Model (DSPM) [19], utility values, which were based on a large survey of patients at risk for major stroke, were 0.9 for TIA, and 0.8, 0.65, 0.5, 0.35 and 0.2 for Rankin Score 1 to 5, respectively. In a study on the

cost-effectiveness of clopidogrel versus aspirin in secondary prevention after vascular events [20], the utility value for peripheral arterial disease was 0.8, mild stroke, 0.76, moderate stroke, 0.39, and for severe stroke, 0.11. In a study of CEA of screening for obstructive sleep apnea (OSA) in stroke patients [21], utility of stroke without OSA was 0.6 and stroke with untreated and treated OSA were 0.4 and 0.6, respectively, in their reference case.

None of the above mentioned CEA of rtPA therapy applied the DALY approach. However, DALYs have been used in CEA across various conditions [22,23], including one that compared aspirin therapy with thrombolytic therapy in stroke care [24]. Different disability weights for stroke have been used in different studies. In the 1990 Global Burden of Disease (GBD) study, disability weights for first-ever stroke with long-term disability is 0.224 for treated patients and 0.262 for untreated patients [25], so is in the GBD 2000 study. The main reason not to estimate the disability weights according to level of severity in the GBD 1990 study is the lack of disability data from the vast majority of the world regions. A Dutch study, however, provides some information: disability weights for first-ever stroke with mild, moderate and severe level of long-term disability are 0.36, 0.63 and 0.92, respectively [26].

Discussion

about the underlying “true”QoL weights of QALYs that reflecting the desirability of living in some health state or set of outcomes. There are continuing debates over which method (to obtain the required preference weights) and whose preferences to use in QALY

calculations [1,2]. According to a review focusing on utility values elicited with the TTO method and from thepatient’sperspectiveregarding own currenthealth [24], CEA based on currently published TTO QoL weights should not be trusted as a tool for setting priorities among diagnostic groups. Worse still, studies have found that individual preferences may be inconsistent with the strict assumptions of expected utility theory that underline the QALY concept [1,2,27]. For all the criticism of QALYs, no acceptable alternative has arisen yet [1,27].

Recently, scholars at the Harvard School of Public Health and Tufts-New England Medical Center developed a comprehensive registry of CEA

(http://www.tufts-nemc.org/cearegistry/) and updated the utility values catalog to 2001 [28]. Given the lack of consensus in estimating QoL after stroke, any of the published QoL weights may be used in a CEA of stroke management. This naturally raises concerns about the meaningfulness in calculating the cost per QALY when considering public reimbursement of interventions or decision-making in health care resource allocation.

To our knowledge, very little is known about QoL after stroke patients Taiwan. Data reported in European countries and the United States are not fully applicable to Taiwan because of culture differences. Furthermore, a notable degree of heterogeneity in QoL was reported in a recent US study [29], even among patients with similar levels of functional status after stroke. Overall, nowadays there are competing views on economic evaluation (welfristview versusdecision maker’sapproach)[30].

Health services researchers have long been interested in the question of how people value health. Studies of CEA in general, and the CEA of stroke treatment in particular, are

still limited in Taiwan. As cost analysis and CEA become more essential to health care policy, it is increasingly important that we understand the strengths and limitations of such work. One of the most useful aspects of health outcome measures, such as QALYs or DALYs,isthatthey allow the‘valueofmoney’provided by differentinterventionsto be measured in a common unit. However, given the lack of reliable and valid measurement methods in eliciting utility values in QALYs or disability values in DALYs, CEA needs to be addressed in a clear and transparent way to ensure its proper use in practice.

References

1. Drummond MF, O'Brien B, Stoddart GL, Torrance G.W. Methods for the Economic

Evaluation of Health Care Programmes, 3rd ed. NY: Oxford University Press, 2005.

2. Gold MR, Siegel JE, Russel LB, Weinstein MC (eds.). Cost-Effectiveness in Health and

Medicine. NY: Oxford University Press, 1996.

3. Klarman HE, Francis JO, Rosenthal GD. Cost-effectiveness analysis applied to the treatment of chronic renal disease. Med. Care. 1968;6:48-54

4. Gold MR, Stevenson D, Fryback DG. HALYS and QALYS and DALYS, Oh My:

similarities and differences in summary measures of population health. Annu Rev Public

Health. 2002;23:115-134.

5. Neumann PJ, Goldie SJ, Weinstein MC. Preference-based measures in economic evaluation in health care. Annu. Rev. Public Health. 2000;21:587-611.

6. Murray CJL, Acharya AK. Understanding DALYs. J Health Econ. 1997;16:703-730. 7. Fox-Rushby J, Hanson K. Calculating and presenting disability adjusted life years

(DALYs) in cost-effectiveness analysis. Health Policy and Planning. 2001;16:326-331. 8. Tsuchiya A. QALYs and ageism: philosophical theories and age weighting. Health Econ.

2000;9:57-68.

9. Tengs TO, Yu M, Luistro E. Health-related quality of life after stroke a comprehensive review. Stroke. 2001;32:964-972.

10. Post PN, Stiggelbout AM, Wakker PP. The utility of health states after stroke: a systematic review of the literature. Stroke. 2001;32:1425-1429.

11. Tengs TO, Lin TH. A meta-analysis of quality-of-life estimates for stroke.

Pharmacoeconomics. 2003;21:191-200.

12. Sandercock P, Berge E, Dennis M, Forbes J, Hand P, Kwan J, Lewis S, Lindley R, Neilson A, Wardlaw J. Cost-effectiveness of thrombolysis with recombinant tissue

plasminogen activator for acute ischemic stroke assessed by a model based on UK NHS costs. Stroke. 2004;35:1490-1497.

13. Sinclair SE, Frighetto L, Loewen PS, Sunderji R, Teal P, Fagan SC, Marra CA. Cost-Utility analysis of tissue plasminogen activator therapy for acute ischaemic stroke: a Canadian healthcare perspective. Pharmacoeconomics. 2001;19:927-936. 14. Fagan SC, Morgenstern LB, Petitta A, Ward RE, Tilley BC, Marler JR, Levine SR,

Broderick JP, Kwiatkowski TG, Frankel M, Brott TG, Walker MD. Cost-effectiveness of tissue plasminogen activator for acute ischemic stroke. NINDS rt-PA Stroke Study Group. Neurology. 1998;50:883-890.

15. Tseng MC, Chang KC. Cost-effectiveness analysis of tissue plasminogen activator for acute ischemic stroke: a comparative review. Acta Neurol Taiwan. 2004;13:149-155. 16. Mar J, Begiristain JM, Arrazola A. Cost-effectiveness analysis of thrombolytic treatment

for stroke. Cerebrovasc Dis. 2005;20:193-200.

17. O'Brien CL, Gage BF. Costs and effectiveness of ximelagatran for stroke prophylaxis in chronic atrial fibrillation. JAMA. 2005;293:699-706.

18. Gage BF, Cardinalli AB, Owens DK. The effect of stroke and stroke prophylaxis with aspirin or warfarin on quality of life. Arch Intern Med. 1996;156:1829-1836.

19. Matchar DB, Samsa GP, Liu S. Cost-effectiveness of antiplatelet agents in secondary stroke prevention: the limits of certainty. Value Health. 2005;8:572-580.

20. Schleinitz MD, Weiss JP, Owens DK. Clopidogrel versus aspirin for secondary prophylaxis of vascular events: a cost-effectiveness analysis. Am J Med. 2004;116:797-806.

21. Brown DL, Chervin RD, Hickenbottom SL, Langa KM, Morgenstern LB. Screening for obstructive sleep apnea in stroke patients: a cost-effectiveness analysis. Stroke. 2005;36:1291-1293.

22. Creese A, Floyd K, Alban A, Guinness L. Cost-effectiveness of HIV/AIDS interventions in Africa: a systematic review of the evidence. Lancet. 2002;359:1635-1643.

23. Muennig P, Pallin D, Sell RL, Chan MS. The cost effectiveness of strategies for the treatment of intestinal parasites in immigrants. N Engl J Med. 1999;340(10):773-779. 24. Moodie ML, Carter R, Mihalopoulos C, Thrift AG, Chambers BR, Donnan GA,

Dewey HM. Trial application of a Model of Resource Utilization, Costs, and Outcomes for Stroke (MORUCOS) to assist priority setting in stroke. Stroke. 2004;35:1041-1046.

25. Truelsen T, Begg S, Mathers C, Satoh T. The global burden of cerebrovascular disease

in the year 2000. GBD 2000 Working Paper. WHO, 2002.

26. Arnesen T, Trommald M. Roughly right or precisely wrong? A systematic review of quality of life weights elicited with the time trade-off. Journal of Health Services

Research and Policy. 2004;9:43-50.

27. Brauer CA, Neumann PJ. Quality-adjusted life years: how useful in medico economic studies. Fundam Clin Pharmacol. 2005;19:603-607.

28. Brauer CA, Rosen AB, Greenberg D, Neumann PJ. Trends in the measurement of health utilities in published cost-utility analyses. Value Health. 2006;9:213-218.

29. Samsa GP, Matchar DB. How strong is the relationship between functional status and quality of life among persons with stroke? J Rehabil Res Dev. 2004;41:279-282. 30. Brouwer WB, Koopmanschap MA. On the economic foundations of CEA. Ladies and