Economics of healthcare systems and health FINANCING: Cost-utility analysis and cost-benefit analysis

Order Details;

Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 1/25 Health Economics & Finance Weekly Notes Week 6: Cost-utility analysis and cost-benefit analysis Introduction Last week you were introduced to the main types of economic evaluation. You learnt how to interpret cost-effectiveness ratios and apply simple forms of discounting. We considered how uncertainty plays a role in economic evaluation and looked at how sensitivity analysis is used to check whether or not the results of an economic study can be considered to be robust. The aim of Week 6 is to discuss how healthcare benefits are identified, measured, and valued in cost-utility analysis and cost-benefit analysis. This session will explore, in depth, the use of quality adjusted life years (QALYs) and monetary values in measuring healthcare output. Different methods for eliciting preferences and monetary values will be considered. The core reading for this week is:  Morris, S., Devlin, N. & Parkin, D. (2012) ‘Chapter 11: Measuring and valuing health care output. In: Economic analysis in health care. 2nd. Chichester, England: John Wiley & Sons, pp. 253-285. Cost-utility analysis and quality adjusted life years (QALYs) Cost-utility analysis (CUA) is a form of economic evaluation that focusses particular attention on the quality of the health outcomes caused or avoided by healthcare programmes, services, and/or technologies. CUA is able to incorporate both the increase in the quantity of life (reduced mortality) and the increase in quality of life (reduced morbidity). In CUA, health improvements are typically measured in QALYs. The results of CUAs can be Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 2/25 used to compare treatments both within and across different specialties. CUA is useful when the programmes that you want to compare have a wide range of outcomes and you want a common unit of output for comparison. CUAs and cost-effectiveness analyses (CEAs) have some similarities, so a lot of the points discussed in CEA are relevant to the discussion of CUA. It is inappropriate to use CUA when: (i) only intermediate effectiveness data can be collected; (ii) effectiveness data shows that the alternatives have similar outcomes; (iii) quality of life is important but cannot be adequately assessed; (iv) the extra cost of obtaining and using utility values is judged to be in itself not to be cost-effective. Calculating and interpreting cost-per-QALY ratios Calculating QALYs requires the collection of (i) disease-specific survival data with and without treatment and (ii) quality-of-life weights for health states with and without treatment. Economists prefer to measure quality of life in terms of utilities. Utilities are values which are assigned to health states on a scale where zero (0) represents the state of death and one (1) represents the state of perfect health. Utility values simply represent the strength of preferences of individuals for health states; higher utility values reflect preferred health states. Von Neumann & Morgenstern (1944) developed the expected utility theory for rational decision making under uncertainty. The model developed by them is normative—it presents how individuals should make decisions in the face of uncertainty. It is not a behavioural model, as it does not describe actual behaviour of individuals. The von NeumannMorgenstern (vNM) theory defines cardinal utility under uncertainty, which is different from ordinal utility under certainty defined in contemporary microeconomics. Since the demand for healthcare is uncertain, vNM utilities (preferences under uncertainty) should be measured in order to estimate quality-of-life weights. Imagine that a patient is expected to live in health state A for 5 years without Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 3/25 treatment and that the utility value which corresponds to the patient’s quality of life in that health state is 0.5. By weighting survival by the utility value, the number of QALYs associated with being in health state A for 5 years can be estimated. In this example the number of QALYs is 2.5 [i.e., 5 years (survival) x 0.5 (quality of life) = 2.5 QALYs]. However, if the patient receives treatment, he is now expected to live for 10 years in health state B, which has a utility value of 0.7. By weighting the estimated survival by the utility value, the number of QALYs associated with being in health state B for 7 years can be estimated. In this example the number of QALYs is 7 [i.e., 10 years (survival) x 0.7 (quality of life) = 7 QALYs]. Therefore the incremental QALYs gained from treatment the patient is 7 QALYs  2.5 QALYs = 4.5 QALYs. What are the main sources of utility values for health states? There are three main sources of utility values: (i) health professionals’ judgement; (ii) values from published sources [e.g., the EQ-5D, the Health Utilities Index 2 (HUI2), the Short Form 36 (SF-36)]; and (iii) values from measurements on a sample of subjects. Often health professionals are asked to make judgements on behalf of patients; however, patient and health professional measurements of quality of life may not be the same. Where estimates described in the published literature exist, it may be possible to use the quality-of-life data in an economic evaluation if the patients and the settings are appropriate to the health economics question under study. Both direct and indirect methods can be used to generate utility values from measurements on a sample of subjects (see below). Direct estimation of utility values from a sample of subjects There are three main techniques that can be used to directly estimate the utility values associated with different health states: (i) visual analogue scales, (ii) time trade-off, and (iii) standard gamble. Each of the methods has advantages and disadvantages associated with use. For a discussion of the issues specifically in relation to CUA, see the paper by Torrance (2006).  Visual analogue scales—Individuals are asked to indicate, by putting a Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 4/25 cross on a line, how they would rate a pre-defined health state. The scale generally runs from 0 to 100. The best health state is at one end of the line and the worst is at the opposite end of the line.  Time trade-off—Individuals are asked to exchange longevity for better health. For example, respondents may be given a scenario where they have 10 years to live. They can either choose to live these 10 years in a given health state (for instance, suffering from severe anxiety) or can trade these 10 years in a given health state for a lower number of years of perfect health. So if on average, respondents chose to trade 10 years of severe anxiety for 6 years of perfect health, the state of severe anxiety would be given a utility value of 6 divided by 10 = 0.6. Thus, 2 years of living with severe anxiety would be measured as (2 × 0.6 =) 1.2 QALYs. An intervention that would restore someone from severe anxiety to perfect health for 2 years would have a QALY gain of (2  1.2 =) 0.8 QALYs.  Standard gamble—Individuals are asked to exchange their uncertain survival for better health. For example, the respondent may be asked to make a trade-off between the certainty of having ill health for a specified amount of time and a gamble that has two possible alternatives: staying in good health for the same period or death. For instance, respondents are given a scenario where they have chronic lung disease and are given information about how this would affect their lives. They are given a choice between staying in that health state, i.e., having chronic lung disease, or taking a gamble where one possible outcome is perfect health and the other possible outcome is death. Respondents are asked what probability of the better outcome would make them indifferent between remaining in the described health state (chronic lung disease) for certain or taking the gamble. Hence, if they are indifferent between the chronic lung disease state and a gamble with a 0.8 probability of the better outcome (but 0.2 probability of the death outcome), 0.8 represents the utility of the Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 5/25 chronic lung disease health state. Since standard gamble measures choice under uncertainty, it is consistent with vNM theory. It also has similarities to some real-life choicesfor example, choosing whether or not to have risky heart surgery. An issue with standard gamble is that people are
generally risk averse, so for some mild illnesses people would rarely gamble at all. Standard gamble and time trade-off do not usually produce the same utility values, so it is worth considering what approach has been used when you are critically appraising an economic evaluation that has used direct measurement. With direct measurement, it is important to question the validity of the utility scores generated (i.e., were the respondents appropriate, did the health state descriptions adequately describe the health states, were all the questions framed in a balanced and neutral way, and is there evidence that the techniques used to elicit valuations were reliable and valid?). Whose values are appropriate for the direct measurement of utility values? There is some debate in the published literature regarding whose values should be used to estimate utility values. There are three different groups from whom we could directly elicit utility values: (i) patients, (ii) general public/community representatives, and (iii) clinical experts/healthcare professionals. Again, there are advantages and disadvantages to asking the different groups.  PatientsWould patients be inclined to exaggerate their condition in order to try and attract more investment into treatments to cure their disease? Or would they underestimate the impact of some health states because they are used to being in them?  General publicHow does a person judge whether or not s/he is sufficiently informed about a health state if s/he does not have the condition? The balance between brevity and burden of information is Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 6/25 difficult to assess. The values of the general public are arguably important, as the results of CUAs may be used in public-sector decision making.  Clinical expertsDo patients and clinical experts always agree about the important determinants of quality of life? Indirect estimation of utility values from a sample of respondents Estimating utility values from a sample of respondents can be time-consuming, expensive, and complicated. Another approach to the estimation of utility values is the use of published multi-attribute health–related quality-of-life instruments (e.g., EQ-5D, HUI2, SF-36, etc.). A range of diverse instruments can be used to measure health-related quality of life. These instruments fall into two very different categoriesgeneric measures and disease-specific measuresand both types of instruments have advantages and disadvantages. Whether a generic or a disease-specific measure is used in a study depends on a number of factors, e.g., the severity and nature of the disease, the expected/unexpected benefits and adverse effects of treatment, length of study, type of analysis planned, etc. Typically, both types of measures are used in clinical studies.  Generic measures have a broad application across many different healthcare treatments and conditions. The same generic instrument can be used to compare and measure the quality of life of patients with different diseases (e.g., patients with heart disease can be compared with patients with lung cancer). Examples of generic measures are the Sickness Impact Profile, Nottingham Health Profile, the HUI2 and HUI3, the SF-36, and the EQ-5D. The dimensions addressed in generic instruments tend to be comprehensive and broad based. One of the criticisms of generic instruments is that they tend to lack sensitivity; i.e., they are unable to adequately distinguish between patients in different health states or they might not be able pick up small but significant changes in patient quality of life over time. Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 7/25  Disease-specific measures are designed to focus on assessing the impact of specific disease states on patients. It is argued that they are more user-friendly as they reduce patient burden and increase acceptability by only focussing on relevant clinical dimensions; i.e., these measures are likely to be able to pick up small but subtle changes in health states over time. One of the disadvantages of using a disease-specific measure is that if only a few dimensions are specified, then the unexpected effects of treatment might not be captured. Examples of these kinds of measures include the Arthritis Impact Measurement Scales and the Back Pain Disability questionnaire. Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 8/25 Key questions that need to be asked of any quality-of-life instrument Measuring health-related quality of life is a complex and time-consuming task, and there is no gold standard against which to assess the properties of quality-of-life instruments. However, there are five key questions which need to be asked before deciding on a quality-of-life instrument for the measurement of health-related quality of life: 1. Is the instrument reliable? 2. Is the instrument valid? 3. Is the instrument sensitive to changes in health? 4. Is the instrument appropriate? 5. Is the instrument practical to administer? Interpreting cost/QALY ratios In any CUA, we end up with two (or more) different cost-per-QALY ratios. The calculation of the incremental cost-per-QALY rati
o reveals the cost per unit of benefit of moving from one healthcare option to an alternative option. Incremental cost-per-QALY ratios are calculated using the simple equation noted below. Incremental cost-per-QALY ratio = (Drug A cost  Drug B cost) = Difference in costs (A  B) (Drug A QALYs – Drug B QALYs) Difference in QALYs (A  B) Whether or not a cost-per-QALY ratio is acceptable in terms of costeffectiveness is very much dependent on the healthcare technology being evaluated and the aims and objectives of the research. The question that must be asked is whether or not the incremental ratio is higher or lower than the decision-maker is prepared to pay per unit of health benefit. It is often Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 9/25 suggested that in the UK, the National Institute for Health and Clinical Effectiveness (NICE) has a willingness to pay (estimate of the ceiling ratio/threshold) of £20,000 to £30,000 per QALY gained (Appleby, Devlin, & Parkin, 2007). This means that any healthcare technology which yields an incremental cost-per-QALY gained of more than £30,000 is unlikely to be recommended by NICE for use in the UK. In the US, acceptable ICERs are estimated to range from $50,000 to $100,000 (Ubel et al., 2003). Arguments against the use of QALYs include the following: 1. QALYs are ageist. Older people do not have as many years left to live as younger people, and so the elderly lose out via the application of weights to the number of life years gained. Following this argument, life years gained and disability days avoided are also ageist measures of effectiveness. Some proponents of QALYs argue ‘so what if QALYs are ageistthe elderly have had their fair innings’. Some even argue that QALYs are not ageist enough! 2. QALYs may not measure the full benefits that are received from healthcare (e.g., a mother may benefit from her child being treated as her quality of life might improve from knowing that the health of her son has improved). QALYs do not take this into account. 3. The derivation of utility values may be incorrect. The way in which utility values are elicited is always open to criticism as there is no right or wrong way to elicit valuationsthis is especially relevant when people without the disease are being consulted or hypothetical questions are being asked. Also in need of consideration is the fact that utility values for a particular health state might change with the expected duration of the condition (e.g., suffering a chronic condition for 2 years might be more than twice as bad as suffering the same condition for 1 year). 4. Opponents of QALYs argue that a QALY is not a QALY is not a QALY. Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 10/25 The theory underpinning the calculation of QALYs assumes that a QALY is a QALY is a QALY (e.g., a life year gained is valued the same for a 4- year-old as it is for an 80-year-old; e.g., 20 life years saved for one person is the same as one life year saved for 20 people). Arguments in favour of the use of QALYs include the following: 1. The use of QALYs has considerable advantages over most singledimensional measures of outcome (e.g., mortality or survival). 2. QALYs are final measures of outcome. Final measures are usually preferred to intermediate outcomes (e.g., cases treated or the extent to which cholesterol levels are lowered). 3. Others would argue that QALYs ’are better than nothing’, and at least by having them the relevant issues associated with measuring quality of life can be debated. 4. Is it better to calculate QALYs rather than attach monetary values to years of life saved and deaths avoided? There is value in the fact that cost-per-QALY ratios of different treatments both across and within specialties can be compared and that these ratios can help allocate scarce resources. QALYs and end of life In January 2009, NICE issued supplementary guidance that effectively increased the threshold for treatments that extend life to £50,000 per QALY gained. This was for treatments that extended life for more than 3 months, for individuals who were not expected to live more than 24 months. Some commentators believed this to be a response to adverse media attention where NICE had rejected some cancer drugs that extend life as not being cost-effective. NICE also had specific arrangements for assessing so-called Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 11/25 ultra-orphan drugs; these are drugs for serious rare conditions that affect less than 1 in 50,000 people. NICE recognises that using the traditional ICER of £20,000£30,000 may not be appropriate for rare diseases where companies may not be able to recoup their development costs without charging more for their drugs. The article below presents a discussion on measuring quality of life:  Carr, A.J. & Higginson, I.J. (2001) ‘Are quality of life measures patientcentred?’, British Medical Journal, 322 (7298), pp. 13571360 [Online]. Available from: http://search.ebscohost.com.ezproxy.liv.ac.uk/login.aspx?direct=true &db=mnh&AN=11387189&site=ehost-live&scope=site (Accessed: 18 August 2010). There is a useful “extra: additional information” section on the Web site:  Properties needed by measures used in clinical practice. British Medical Journal, 322 (7297) [Online]. Available from: http://www.bmj.com/cgi/content/full/322/7297/1297/DC1 (Accessed: 10 March 2010). Use of cost-per-QALY league tables Cost-per-QALY league tables are used to rank different healthcare interventions according to their cost per QALY gain. Decision-makers often need to assess the relative value for money of competing healthcare interventions. In theory, league tables can be used to help guide decision-makers as they attempt to allocate scarce resources. If decision-makers are guided by league tables, then they will invest more money in procedures with low cost-per-QALY ratios and try to avoid investing money in procedures with high cost-per-QALY ratios. League tables can be considered a formal approach to resource allocation in healthcare. See Table 1 below for an example of a cost-per-QALY league table. Table 1: Cost per QALY league table Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 12/25 Treatment Cost per QALY (€) Stop smoking advice 750 Hip replacement 1,600 Heart transplant 10,000 Haemodialysis in hospital 30,000 QALY league tables are often criticised as being counterintuitive to the clinician. To illustrate, in this example, the league table suggests that we should spend money on getting people to stop smoking before we do heart transplants. However, as long as we look at the league table from a population perspective we might be able to see the value in this. For an extra €20,000 we can generate an extra 27 QALYs if we invest in getting people to stop smoking, whereas if we invest our extra €20,000 in heart transplants we only generate 2 extra QALYs. An idea related to this is the ‘rule of rescue’, i.e., whether there is a preference to save the life of someone in imminent danger of dying instead of improving the quality of life of someone else whose life is not in immediate danger or saving future lives through diseaseprevention programmes. Refer to the article below for a discussion on utility measurement in healthcare:  Torrance, G.W. (2006) Utility measurement in healthcare: the things I never got to, PharmacoEconomics, 24 (11), pp. 10691078 [Online]. Available from: http://search.ebscohost.com.ezproxy.liv.ac.uk/login.aspx?direct=true &db=mnh&AN=17067192&site=ehost-live&scope=site (Accessed: 18 August 2010). Types of economic evaluation Table 2 summarises the main methods of economic evaluation, including the methods we covered in last week’s materials. Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 13/25 Table 2: Methods of economic evaluation Economic analysis Cost measure Type of consequences identified for all alternatives Benefit measure Result Type of efficiency Cost-benefit analysis (CBA) Money Single or multiple effects, not necessarily common to all alternatives Money Net cost:benefit ratio Allocative Cost-utility analysis (CUA) Money Single or multiple effects, not necessarily common to all alternatives QALYs (quality adjusted life years) Cost per QALY, ICER (Incremental costeffectiveness ratio) Allocative within health sector, otherwise technical Costeffectiveness analysis (CEA) Money One single clinical or health effect of interest to both alternatives Natural units (i.e., life years gained, change in blood pressure, etc.) Cost per unit of consequence, e.g., cost per unit drop in blood pressure, ICER Technical Costconsequence analysis (CCA) Money Clinical, health, and non-health impacts Listing of separate consequences (e.g., number of GP visits) with no comparable valuation Costs and consequences listed separately Technical Costminimisation analysis (CMA) Money Clinical or health effect needs to be identical between alternatives Benefits are equivalent Costs for each alternative Technical Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 14/25 Introduction to cost-benefit analysis Cost-benefit analysis (CBA) is the only economic study design where costs and benefits are both measured in monetary values. Unlike CEA, where outcomes are measured in natural units (e.g., life years gained or disability days avoided), CBA measures outcomes in pound notes, dollars, or euros. In theory, CBA is undertaken from a societal perspective and aims to identify, measure, and value all costs and benefits associated with the healthcare intervention under evaluation no matter who or which sectors of the economy incur costs and benefits. However, in practice, CBA becomes a very narrow form of economic evaluation because it is often restricted by measurement and valuation problems. It is the most theoretically solid form of economic evaluation, as it is based on complex welfare economic theory. It is the most widely used form of economic evaluation in the public sector and has a long history in transport and the environmental sectors. Although it was the first analytical study design to be introduced into health economics, it is not one of the forms of economic evaluation most frequently carried out. Interpreting a cost-benefit ratio The main strength of CBA is that it helps us to answer the question: is the healthcare
intervention, programme, or service worth it? CBA is the only method of economic evaluation that allows us to question the worthiness of an individual project or intervention (i.e., individual interventions can be assessedall other forms of economic evaluation require a comparison of at least two alternatives). CBA is used to determine whether the benefits of an intervention are greater than the costs of an intervention. If the benefits are greater than the costs, then the intervention is judged to be worthwhile. Imagine that a hip replacement costs the health service 5,000 euros and produces benefits of 7,000 euros. Subtracting costs from benefits reveals a Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 15/25 positive net benefit of 2,000 euros. The hip replacement is judged to be worthwhile, and the intervention should therefore be carried out. If the intervention costs more than the benefits producedi.e., there is a negative net benefitthen the intervention should not be carried out. The simple decision rule for CBA is that projects or programmes should be introduced or expanded if B – C > 0. A second strength of CBA is that it allows very different healthcare programmes to be compared (e.g., smoking cessation programme vs. mental health community care programme). When CBA is used to compare different programmes, the projects are ranked according to their cost-benefit ratios. CBA is not restricted to comparing programmes within the health sector; it can be used to inform resource allocation decisions between different sectors of the economy, e.g., healthcare programmes can be legitimately compared with non-healthcare programmes. Given that there are many determinants of health, including education policies, housing programmes, and environmental projects, this approach is the ideal economic evaluation method to compare interventions from different sectors which impact on health. For example, CBA is often advocated by those involved in health impact assessment. By valuing costs and benefits in monetary terms, there is a common denominator, and this allows the comparison of interventions both between and across sectors. CBA in healtha controversial approach? In the real world, those involved in healthcare decision making are often uncomfortable assigning monetary values to the outcomes of healthcare interventions. The major disadvantage of CBA is the requirement that human lives and quality of life be valued in monetary units. Many decision-makers find this difficult or unethical or do not trust analyses that depend upon such valuations. A second reason for the apparent lack of interest in the CBA of healthcare Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 16/25 interventions and programmes is that there are many practical measurement and valuation problems associated with assigning monetary values to health benefits. Eliciting the willingness to pay for health benefits is difficult for several reasons, firstly because in countries where individuals do not pay directly for healthcare (i.e., it is paid for through tax or insurance), people often underestimate what they pay for health because they have no reference points. Secondly, willingness to pay is closely related to ability to pay, so any monetary value of health interventions can be skewed towards the preferences of more affluent individuals. The benefits of healthcare technologies are not easily measured in monetary terms; in particular, estimating the value of a life saved in euros is often a very controversial issue. As a result, CBAs may ignore many intangible but very important benefits, e.g., relief of anxiety or freedom from pain. Salaries are often used to estimate the monetary benefit of increased health if an individual is able to return to work. CBA may therefore seem to discriminate against those for whom a return to productive employment is unlikely, e.g., the elderly or the unemployed. Some of these measurement and valuation problems are the focus of the next section. Valuing health benefits in monetary terms Monetary valuation of gains and losses plays a central role in CBA. When using CBA within the framework of healthcare, it is often difficult to attach meaningful values to such gains and losses, as they can include years of life saved, pain relief, and/or improved mobility. Firstly, even in markets in which healthcare goods and services (e.g., diagnostic tests or pharmaceuticals) are traded, market distortions often exist, including a lack of competition, tax subsidy measures and externalities, etc. By looking to the markets for these prices, it is assumed that the underlying distribution of purchasing power is acceptable and that individuals are the best judges of their own welfare. For non-traded goods and services, such markets do not exist. Incommensurate (e.g., satisfaction with GP service), and intangible (e.g., anxiety) costs and benefits are therefore even more difficult to measure in monetary terms. Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 17/25 Advances in (i) health status measurement and (ii) methods for the valuation of non-traded healthcare goods and services have led to improved techniques which have been incorporated into the framework of CBA in the field of healthcare. Where prices do not reflect real values or there are no prices even charged, three methods are currently used to value and compare the costs and benefits of healthcare. (i) Human capital approach The human capital approach attempts to value human life through the estimation of loss of productivity if an individual is absent from work (i.e., the value of a life lost or gained is calculated as being equal to the value of a person’s future flow of income). There are many criticisms of the human capital approach, including its disregard of the individual’s valuation of his/her own welfare to be important and the implicit superiority of the outcomes of the young over the old and of the employed over the unemployed. In addition, the method fails to recognise pain and suffering and the psychosocial consequences of illness. (ii) Revealed preference approach In the revealed preference approach, valuations of health benefits are estimated from actual choices and behaviours that are observed in the real world. By studying the actions of individuals, including decision-makers, values that are placed on goods and services can then be estimated; this includes scrutiny of injury compensation, court awards, and historical health programme funding. In practice, court awards for injury compensation to elicit the minimum value that society places on health outcomes are often criticised, as they are mainly based on the human capital calculation of lost earnings. Previous funding decisions have also been analysed to extract valuations of health benefits. For example, if 1 million dollars were spent on building and maintaining a road which will save 100 lives through a reduction in the number of road accidents, then there is an implicit valuation that a life is worth at least 10,000 dollars. Wage-risks studies have also been carried out Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any add
itional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 18/25 to examine the relationship between particular health risks associated with dangerous jobs and the wage rates that individuals require to accept those risky jobs. Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 19/25 (iii) Stated preference approach The stated preference approach uses surveys or experiments to estimate credible values for improvements or reductions in health status. The stated preference approach is also known as the willingness-to-pay approach or the willingness-tobenefit approach, of which contingent valuation is the most commonly used method. This approach involves the use of survey methods to present respondents with hypothetical scenarios about healthcare programmes or projects under investigation (e.g., asking people how much they would be prepared to accept increases in risk of death or injury or asking people how much they would be willing to pay to reduce their risk of death or injury). The stated preference approach has been used in the following contexts:  Preventative technologies: Respondents are asked how much they would be willing to pay for vehicle devices to lower the risk of injury in traffic accidents  Treatments and services: Respondents are asked how much they would be willing to pay for a reduction in pain after surgery  Health states: Respondents are asked how much they would be willing to pay to be rid of symptoms such as nausea and coughing, etc. Participants must have a good understanding of the scenarios described, make up a representative sample of respondents, and not feel that they have a burden of information to digest. The major criticisms of this type of approach can be outlined as follows: the behaviour of individuals may not be captured by hypothetical scenarios; decisions are often influenced by ability to pay; the tendency of respondents is to try to please the researchers rather than state how they feel; individuals are vulnerable, and the framing of questions asked and scenarios Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 20/25 presented may influence their answers. In addition, whose views should count in these surveys and experiments? Should patients or the general public’s views be elicited? This site offers an excellent overview of CBA:  Centers for Disease Control and Prevention (2010) Cost benefit analysis. “Cost Benefit Analysis”, copyright 2010 by Centers for Disease Control and Prevention, U.S. Department of Health & Human Services. [Accessed August 23, 2010] Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 21/25 The article below presents challenges associated with willingness-to-pay approaches:  Cookson, R. (2003) ‘Willingness to pay methods in health care: a sceptical view’, Health Economics, 12 (11), pp. 891894 [Online]. Available from: http://sfxhosted.exlibrisgroup.com.ezproxy.liv.ac.uk/lpu?title=Health +Economics&volume=12&issue=11&spage=891&date=2003&issn=&e issn= (Accessed: 18 August 2010). Critical appraisal of a published economic evaluation In recent years, there has been a large increase in the amount of published economic evaluations in both the health economics and medical literatures. In response to this, the number of guidelines and checklists aimed at facilitating the critical appraisal of economic studies has grown. The use of guidelines and checklists is intended to improve the quality with which economic evaluations are undertaken and to assist healthcare decisionmakers to assess the methodological quality and rigour of such studies. It is anticipated that use of such quality checks will help to improve and standardise the conduct of economic evaluations and make it easier for decision-makers to compare the results of different studies. Guidelines Guidelines for economic evaluation are typically referred to in the planning stages of the economic evaluation. Guidelines can be categorised as formal (definite requirement by government before reimbursement) or informal (recommendation prior to reimbursement) or as guidelines for health economics methods (to improve the methodological content of health economics studies). Indeed, many countries have issued guidelines for the conduct of CEAs, including Australia, Canada, the Netherlands, and the UK. In summary, good guidelines serve to instruct the analyst on how to adequately conduct and report economic studies. Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 22/25 Checklists In contrast, checklists are most often used after an economic study has been conducted in order to verify the extent to which clinical and economic issues have been addressed by the analysts. It is important to remember that the authors of published papers may find it impossible to meet all of the important points highlighted by checklists. Equally, in interpreting the quality of a study, it is often difficult to draw quality inference from a list of ticks or crosses against statements addressed by the checklist. Narrative discussion of the points considered within a checklist is vital to improving and placing in context the critical appraisal of the economic evaluation being assessed. Most of the guidelines and checklists that are used to critically appraise economic evaluations assume that the clinical evidence is derived from a single trial or from a systematic review of the published literature. However, as the number of published economic evaluations based on the results obtained from economic models rises, there is a need for assessment tools to extend their range to incorporate the quality assessment of the economic models used. A recent review to assess guidelines for good practice in decision-analytic modelling in health technology assessment reveals that, although guidelines do exist, further research is required in this rapidly evolving field. Using a critical appraisal checklist to quality-assess published economic evaluations As economic evaluations are very much reliant upon input from experts in a variety of established disciplines, it is very difficult to expect one single guideline or checklist to adequately cover all design and analysis issues; in theory, a minimum of three different tools (clinical, economic, and statistical) could be required to critically appraise a single economic evaluation. Clearly, this would be very unwieldy for the analyst and high in opportunity costs. According to Drummond et al. (2005), there are three important questions Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 23/25 that the reader must ask when reading an economic evaluation: 1. Is the study methodology employed in the study appropriate? 2. Are the study results valid? 3. If the study results are valid, will they apply to my set
ting? Drummond et al. (2005) provide the following 10-step checklist as a reporting framework for economic studies. Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 24/25 Drummond 10-point checklist: 1. Was a well-defined question posed in answerable form? 2. Was a comprehensive description of the competing alternatives given? 3. Was the effectiveness of the programme or service established? 4. Were all of the relevant costs and consequences of each alternative identified? 5. Were costs and consequences measured accurately in appropriate physical units? 6. Were costs and consequences valued credibly? 7. Were costs and consequences adjusted for differential timing? 8. Was an incremental analysis of costs and consequences of alternatives performed? 9. Was allowance made for uncertainty in the estimates of costs and consequences? 10. Did the presentation and discussion of study results include issues of concern to users? The CASP (Critical Appraisal Skills Programme) checklist is also a good framework for critically appraising economic evaluations. For more information, see the following resource:  Public Health Resource Unit, England (2006) Critical appraisal skills programme (CASP): 10 questions to help you make sense of economic evaluations [Online]. Available from: http://www.phru.nhs.uk/Doc_Links/Economic%20Evaluations%2010%2 0Questions.pdf (Accessed: 18 August 2010). Grading systems and criteria lists Health Economics & Finance Copyright—Laureate Online Education © All rights reserved, 2000–2010, The Module, in all its parts: Syllabus, guidelines, weekly notes, Discussion questions, technical notes, images and any additional material is copyrighted by Laureate Online Education B.V. Last update: 17 August 2012 25/25 More recently there has also been interest in the potential value of grading systems and criteria lists. Grading systems tend to go one step further than guidelines and checklists by providing a scoring system which allows simple comparisons to be made between studies. The main criticism of grading systems for this purpose is that their use may discourage further reading of studies which are labelled as poor quality but which may still contain information that may be of potential value to decision-makers. Criteria lists are a core set of items that can be used to assess the quality of studies in systematic reviews of the literature. Specifically, in systematic reviews of economic evaluations, criteria lists can be used to assess the methodological quality of economic evaluations. The following article presents a discussion on critical appraisal of CEA and CUA studies:  Soares, M. & Dumville, J.C. (2008) ‘Critical appraisal of costeffectiveness and cost-utility studies in health care’, Evidence-Based Nursing, 11 (4), pp. 99102 [Online]. Available from: http://sfxhosted.exlibrisgroup.com.ezproxy.liv.ac.uk/lpu?title=Evide nceBased+Nursing&volume=11&issue=4&spage=99&date=2008&issn= &eissn= (Accessed: 18 August 2010).