Health Economics (Health Econ)

The implementation of colorectal cancer mass screening is a high public health priority in France, as in most other industrialised countries. Despite evidences that screening using guaiac fecal occult blood test may reduce colorectal cancer mortality, no European country has organised widespread mass screening with this test. The low sensitivity of this test constitutes its main limitation. Immunological tests, which provide higher sensitivity than the guaiac test, may constitute a satisfactory alternative. This study was carried out to compare the costs and the effectiveness of 20 years of biennial colorectal cancer (CRC) screening with an automated reading immunological test (Magstream) with those obtained with a guaiac stool test (Haemoccult). The model used to estimate the costs and effectiveness of successive biennial CRC screening campaigns was a transitional probabilistic model. The parameters used in this model concerning costs and CRC epidemiological data were calculated from results obtained in the screening program run in Calvados or from published results of foreign studies because of the lack of French studies. The use of Magstream for 20 years of biennial screening costs 59 euros more than Haemoccult per target individual, and should lead to a mean increase in individual life expectancy of 0.0198 years (i.e. about one week), which corresponds to an incremental cost-effectiveness ratio of 2980 euros per years of life saved. Our results suggest that using an immunological test could increase the effectiveness of CRC screening at a reasonable cost for society. Copyright

Socio-economic status effects on total and cause-specific mortality are studied using data on all 15.8 million inhabitants of the Netherlands in 1999. Two problems are addressed that often hamper this kind of research: the lack of reliable social status information at the individual level and the intermingling of individual and neighbourhood status effects. The first problem is dealt with by using socio-economic status information of the very close environment of the detailed postcode areas (average 41 inhabitants) in which one is living and the second one by combining this information with such area information at the much larger level of neighbourhoods (1500 inhabitants) or boroughs (6600 inhabitants). Clear and independent effects of socio-economic status at all three levels of aggregation are found on total mortality and for a majority of causes of death. In almost all cases, the effects are to the disadvantage of people living in the lowest status areas. The effects are generally strongest at the detailed postcode level and weakest at the borough level, suggesting greater importance of factors at the nearby or individual level than at the farther away level(s).

An important subject of debate in cost-utility analysis of health care programmes is whether to include costs of unrelated medical care in life years gained. The inclusion of such costs is likely to be of consequence in the case of primary prevention. This paper presents different strategies regarding the inclusion not only of the costs, but also of the health effects of unrelated medical care in economic evaluations. Four different cost-utility ratios are presented and related to the criterion of internal consistency. In addition, the possibility to relate the ratios to a well-posed decision problem is analysed. An example computes the different ratios for smoking cessation interventions in different age groups. Including health care costs of unrelated medical care in life years gained increases cost utility ratios, but excluding unrelated medical costs favours smoking cessation interventions targeted at older smokers over those at younger smokers. We conclude that for primary prevention only a cost utility ratio that includes both the costs and effects of unrelated medical care meets the criterion of internal consistency and is related to a meaningful decision problem. Therefore, this type of cost-utility ratio should be preferred even if the data requirements may be substantial.

Because of aging trends in the US, the number of prevalent colorectal cancer patients is expected to increase. We projected economic burden to the Medicare program and its beneficiaries through the year 2020. Burden was estimated for the initial phase of care, the period following diagnosis, the last year of life, and the continuing phase. Projected burden was evaluated with varying assumptions about incidence, survival, and costs of care. Estimated costs of care in 2000 in the initial, continuing, and last year of life phases of care were approximately $3.18 billion, $1.68 billion, and $2.63 billion, respectively. By the year 2020 under the 'fixed' current incidence, survival, and cost scenario, projected costs for the initial, continuing, and last year of life phases were $4.75 billion, $2.63 billion, and $4.05 billion. Under the current trends scenario (decreasing incidence, improving survival, and increasing costs), costs were $5.19 billion, $3.57 billion, and $5.27 billion. By the year 2020, estimated costs of colorectal cancer care among individuals aged 65 and older increased by 53% in the fixed scenario and by 89% in the current trends scenario. The future economic burden of colorectal cancer to the Medicare program and its beneficiaries in the US will be substantial.

The 1918 Influenza Pandemic is used as a natural experiment to test the Fetal Origins Hypothesis. This hypothesis states that individual health as well as socioeconomic outcomes, such as educational attainment, employment status, and wages, are affected by the health of that individual while in utero. Repeated cross sections from the Pesquisa Mensal de Emprego (PME), a labor market survey from Brazil, are used to test this hypothesis. I find evidence to support the Fetal Origins Hypothesis. In particular, compared to individuals born in the few years surrounding the Influenza Pandemic, those who were in utero during the pandemic are less likely to be college educated, be employed, have formal employment, or know how to read and have fewer years of schooling and a lower hourly wage. These results underscore the importance of fetal health especially in developing countries.

Regression results show that nearly half of 1960-1993 growth in real per capita medical spending and almost two-thirds of its 1983-1993 growth were due to ever-increasing levels of insurance coverage (the spending portion paid by third parties). Growth in coverage may have played a minor part as well; we would not rule out the standard finding that it has had a positive but relatively small effect. Viewed from a different perspective, the results imply that about two-thirds of 1960-1993 spending growth came via cost-increasing advances in medical technology resulting from: (1) commercial research and development induced by coverage levels and (2) noncommercial medical research. The remaining one-third, was due to standard factors: age-sex mix changes, income growth and coverage growth (the latter playing a small but indeterminate part).

HESG was founded in 1972 as part of a conscious effort to establish health economics as an identifiable sub-discipline. It is debatable whether the growth of health economics was demand-led or supplier-driven, but in either case the existence of a HESG played a vital role. HESG was founded as a private club, in the tradition of English gentlemen's clubs, designed to provide a forum for debate and an invisible, supportive faculty for health economists dispersed between different organisations throughout the UK. It was given impetus by public economists at the University of York, who were effectively academic entrepreneurs, motivated in part by private gain, but by their actions overcoming the free-rider problem that might otherwise have retarded the development of health economics. Over the course of its first 25 years, HESG has changed and its membership has grown and altered in composition - over this period, HESG has evolved from a private club to a professional network. It has made a vital contribution to the existence and form of health economics as a subdiscipline in the United Kingdom, and has in turn itself been influenced by the subdiscipline. As a subdiscipline, UK health economics in the 1990s generally draws on a small body of economic theory and is practised by a distinct, identifiable group of economists. This paper was commissioned by HESG, as a history of the organisation. It also analyses the foundation and evolution of HESG as an institutional arrangement designed to overcome a collective action problem.

This paper contains a review of the impact of health economics on health policy in England during the past 25 years. Some health economists have expressed disappointment with the scale of the impact that health economics has had on policy but the record set out below suggests that there is modest cause for celebration. That is not to say that there is cause for complacency. There is still a long way to go before all important health policies are based on sound economics reasoning and evidence. The paper begins with some definitions and background; it covers nine areas of health policy, and health policy making, where past impacts of health economics have been postulated; it covers briefly the reciprocal impact of health policy on health economics; and it concludes with a discussion about the findings.

Analyzing cross-sectional data from the National Medical Expenditure Survey (NMES), we find that the predicted probability of private insurance coverage for low-income individuals as a group fell dramatically from 1977 to 1987. The results of a decompositional technique show that the relationship between full-time employment and private insurance has weakened over the period for low-income females, but has strengthened for males in this group. While it appears that low-income females benefit from part-time employment relative to their unemployed cohorts, no discernible difference is found in the likelihood of being covered by private insurance for part-time and unemployed males. Finally, evidence suggesting a weakening over time in the relationship between part-time employment and private insurance coverage is found among middle-income females and high-income males. From a policy perspective, passage of the Health Insurance Portability and Accountability Act of 1996 has taken an important first step in attempting to lower the number of uninsured, especially among full-time workers. Our findings, however, suggest that this legislation may be too limited in scope to effectively reach part-time workers presently uninsured.

In this paper it is shown that there have been significant structural changes in the composition of the Hospital and Community Health Services (HCHS) workforce over the 1980s. The number of doctors, nurses and other medical professionals has grown at the expense of support staff such as ancillaries and maintenance workers. The number of agency and contract staff has risen rapidly, partly offsetting the loss of directly-employed support staff. Changes in the workforce have been compared with changes in activity, as measured by the cost-weighted activity index. According to this measure labour productivity has grown by a compound rate of 1.9% annually. Adjusting the labour force index for the wage bill of each group reveals productivity growth of 1.5%. The effectiveness of treatment, as proxied by the decline in avoidable perinatal mortality, has grown by 3.4% annually. Unit labour costs have fallen over the period at an average annual rate of 0.3%. The trend conceals wide fluctuations, with labour costs falling slowly during the first half of the decade, and rising strongly during the second half. Medical professionals benefited disproportionately from wage increases in comparison with other HCHS groups during the mid to late 1980s.

This article aims to assess the development of the English National Health Service (NHS) over the period 1979--2005, against the original, and often repeated, core objectives of the system: that it be universal in offering coverage to all members of the population in times of health care need; that it be comprehensive in its provision of health care services; and that it be (largely) free at the point of use. Comprehensiveness is open to interpretation, and may depend upon the wealth of the nation. Universality and (largely) free care at the point of use, which lend themselves to the principle of equal access for equal need, are more concrete, and it is not difficult to ascertain if they have been substantially and/or systematically violated. The article details briefly the developments in resource allocation, provider payment mechanisms, incentives and accountability, and notes that much of the emphasis on health sector change since the mid 1980s has been placed upon improving supply side efficiency and reducing waiting lists/times. Improving NHS efficiency, and indeed related aspirations associated with choice and health outcomes, can be perceived as 'secondary' objectives, in that they should not serve to undermine the core objectives of the system, assuming that the security offered by having an accessible, universal health care system is considered worthy of protection. The overall conclusion is that the NHS has performed quite well against its core objectives to date, although it is possible that the current preoccupation with choice and health outcomes will lead us down a different policy path in the future.

The study aimed to estimate the value of the change in health in Sweden 1980/81 to 1996/97. Quality-adjusted life years (QALYs) for men and women at specific ages were estimated for 1980/81, 1988/89 and 1996/97, by combining survival rates and health state scores. Data from the Swedish Survey of Living Conditions (n = 39,966) were used to estimate age-specific health state scores. Responses to selected survey questions were mapped into the EQ-5D measure, using the UK EQ-5D index tariff to derive health state scores. The monetary value of a QALY was assumed to be 100,000 dollars. Life expectancy for infants increased by 3.68 years for males and 2.70 years for females between 1980/81 and 1996/97. Average health status decreased in younger age groups whereas it increased in older age groups. Expected QALYs for infants increased by 2.64 for males and 0.54 for females. With 3% discounting the gain was 0.11 QALYs (11,000 dollars) among males and a loss by 0.58 QALYs (58,000 dollars) among females. The corresponding gain in discounted QALYs for a 75-year-old was 1.15 (115,000 dollars) and 0.80 (80,000 dollars), respectively. It is concluded that older persons have experienced considerable health gains whereas the health gains have been small or non-existent for younger women.

As the Irish health system embarks upon its first major structural reorganisation in over 30 years, developments within this system over the past two decades are assessed. Real cuts in health expenditure achieved in the 1980s contrast sharply with the unprecedented increase in resources devoted to the health system in the 1990s. While successive statements of health strategy have prioritised the objectives of equity, efficiency and quality of care, questions arise regarding the return achieved with the increased investment. With higher levels of economic growth, more people have been buying private health insurance such that almost half the population are now privately insured. At the same time, the numbers with eligibility for health services without charge have decreased while those from lower socio-economic groups continue to have higher levels of utilisation. Equity issues arise, however, with regard to access to public hospitals as the rate of growth in admissions for private patients outstrips that for public patients. The establishment of a National Treatment Purchase Fund to purchase treatment in private facilities for public patients on waiting lists raises efficiency and equity questions as the treatment of private patients in public hospitals is heavily subsidised while the State pays full cost for the treatment of public patients in private facilities.

This paper estimates whether state-level implementation of community rating and guaranteed issue regulations in the non-group health insurance market during the 1990s affected the decision of taxpayers to be self-employed. Using a panel of tax returns that span 1987-2000, we find no statistically significant effect of the reforms on the propensity to be self-employed overall, although we find evidence of an increase in self-employment among older taxpayers and weaker evidence of decreases among younger cohorts. Copyright © 2013 John Wiley & Sons, Ltd.

The use of illicit drugs causes health and social problems which imply economic costs to society. This paper uses the cost-of-illness method, in particular, the human-capital approach to estimate the prevalence-based economic costs of illicit drug use in Ontario in 1992. This methodology is consistent with international guidelines formulated at the 1994 International Symposium on Economic and Social Costs of Substance Abuse. The economic cost of illicit drug use is estimated at $489.29 million (Canadian dollars) in 1992. Associated with these costs are health-related harms: 211 deaths, half of which occur before the age of 35; and 20 690 days stay in public hospitals.

A search was carried out for economic evaluations of hepatitis B (HBV) vaccination, published between 1994 and 2000. The results of these studies are discussed according to the level of HBV endemicity. The great majority of these evaluations were carried out for industrialized countries, for the most part situated in areas of low to very low HBV endemicity. In countries of very low endemicity economic evaluations have yielded contradictory results, depending on the type of epidemiological model they used. The cost-effectiveness of adding universal to selective vaccination strategies in these countries depends on the selective strategies' ability to sufficiently identify, reach and fully vaccinate persons in various risk groups. In areas of low, intermediate and high endemicity, universal vaccination seems justifiable on the basis of economic evaluation. In general, the accuracy of the models has improved over the years, but still the transparency, completeness and comparability of analyses could improve considerably. By noting this, the suitability of different methodologies for different areas of endemicity and vaccination strategies is discussed. It is recommended that specific guidelines for economic evaluation of the prevention of infectious diseases be developed to guarantee the relevance of and to improve the comparability between studies. Copyright

The General Health Questionnaire (GHQ) is frequently used as a measure of mental well-being with those people with values below a certain threshold regarded as suffering from mental stress. Comparison of mental stress levels across populations may then be sensitive to the chosen threshold. This paper uses stochastic dominance techniques to show that mental stress fell in Ireland over the 1994-2000 period regardless of the threshold chosen. Decomposition techniques suggest that changes in the proportion unemployed and in the protective effect of income, education and marital status upon mental health were the principal factors underlying this fall.

This paper examines the determinants of GP visiting in Ireland, using panel data from the Living in Ireland Survey from 1995-2001. While cross-sectional studies provide important information on GP visiting patterns at a certain point in time, with panel data we can also control for unobserved individual heterogeneity, as well as identify whether it is the same individuals who consistently visit their GP year on year, or whether there is more mobility in visiting. We therefore estimate dynamic models of GP utilisation, and attempt to decompose the observed variation in GP visiting into components attributable to observed individual characteristics, unobserved individual heterogeneity and state dependence.

The original article to which this Erratum refers was published in Health Economics 8(7) 1999, 579-598.

No Abstract

The human consequences of the recent global financial crisis for the developing world are presumed to be severe, but few studies have quantified them. This letter estimates the human cost of the 2008-2009 global financial crisis in one critical dimension-infant mortality-for countries in sub-Saharan Africa. The analysis pools birth-level data, as reported in female adult retrospective birth histories from all Demographic and Health Surveys collected in sub-Saharan Africa. This results in a data set of 639,000 births to 264,000 women in 30 countries. We use regression models with flexible controls for temporal trends to assess an infant's likelihood of death as a function of fluctuations in national income. We then calculate the expected number of excess deaths by combining these estimates with growth shortfalls as a result of the crisis. The results suggest 28,000-50,000 excess infant deaths in sub-Saharan Africa in the crisis-affected year of 2009. Notably, most of these additional deaths were concentrated among girls. Policies that protect the income of poor households and that maintain critical health services during times of economic contraction may reduce the expected increase in mortality. Interventions targeted at female infants and young girls can be particularly beneficial. Copyright © 2012 John Wiley & Sons, Ltd.

Optimising the design of discrete choice experiments (DCE) involves maximising not only the statistical efficiency, but also how the nature and complexity of the experiment itself affects model parameters and variance. The present paper contributes by investigating the impact of the number of DCE choice sets presented to each respondent on response rate, self-reported choice certainty, perceived choice difficulty, willingness-to-pay (WTP) estimates, and response variance. A sample of 1053 respondents was exposed to 5, 9 or 17 choice sets in a DCE eliciting preferences for dental services. Our results showed no differences in response rates and no systematic differences in the respondents' self-reported perception of the uncertainty of their DCE answers. There were some differences in WTP estimates suggesting that estimated preferences are to some extent context-dependent, but no differences in standard deviations for WTP estimates or goodness-of-fit statistics. Respondents exposed to 17 choice sets had somewhat higher response variance compared to those exposed to 5 choice sets, indicating that cognitive burden may increase with the number of choice sets beyond a certain threshold. Overall, our results suggest that respondents are capable of managing multiple choice sets - in this case 17 choice sets - without problems.

We studied the relationship between current cigarette smoking and price among 34145 respondents, aged 15-29 years, to the 1992-1993 Tobacco Use Supplements to the Current Population Survey. The price elasticity of current smoking varied inversely with age: -0.831 (S.E. 0.402) for ages 15-17; -0.524 (S.E. 0.256) for ages 18-20; -0.370 (S.E. 0.188) for ages 21-23; -0.202 (S.E. 0.175) for ages 24-26; and -0.095 (S.E. 0.157) for ages 27-29. In response to higher prices, older youth were more likely to reduce the number of cigarettes smoked per day than to quit entirely. Among 15-17-year-olds, smoking cigarettes 'some days' was more sensitive to price than smoking 'every day'. Cigarette smoking was inversely related to the prices of premium brands, but not discount brands.

A general approach is discussed to assess the uncertainty surrounding the cost effectiveness ratio (C/E-ratio) estimated on the basis of data from a randomised clinical trial. The approach includes the calculation of a 95% probability ellipse and introduces the concept of a so called C/E-acceptability curve. This last curve defines for each predefined C/E-ratio the probability that the C/E-ratio found in the study is acceptable. The approach is illustrated by estimates of costs per life saved and costs per patient discharged alive on the basis of data from a phase II trial addressing the value of anakinra in treating sepsis syndrome.

Recent research has derived preference scores from the SF-36. We compare the practicality and construct validity of SF-36 derived preference scores with directly elicited time trade off (TTO) and visual analogue scale (VAS) scores. In this observational study, low back pain (LBP), patients were asked to complete disease specific, generic (SF-36), and health state preference (VAS and TTO) instruments. Baseline SF-36 responses were converted to preference scores using six published algorithms. Response rates for the SF-36 derived and TTO preference values were 354 of 379 (93%) and 303 of 379 (80%), respectively. Thirty patients were excluded from the TTO exercise because of difficulties comprehending the scaling task. Choice based methods (standard gamble, TTO) yielded higher and more uniform estimates of preference (0.77-0.79) than non-choice based methods (VAS) (0.42-0.70). Directly elicited TTO values were variable and had less power to distinguish among patients with differing severity of LBP. All SF-36 derived preferences exhibited a minimum threshold implying a potential floor effect for severely ill patients. SF-36 derived preferences demonstrated good practicality and construct validity in this setting, however different methods will yield disparate estimates of marginal benefit. This emphasises the need for a standardised algorithm for deriving SF-36 preference scores.

The criteria used by the National Institute for Health and Clinical Excellence (NICE) for accepting higher incremental cost-effectiveness ratios for some medicines over others, and the recent introduction of the Cancer Drugs Fund (CDF) in England, are assumed to reflect societal preferences for National Health Service resource allocation. Robust empirical evidence to this effect is lacking. To explore societal preferences for these and other criteria, including those proposed for rewarding new medicines under the future value-based pricing (VBP) system, we conducted a choice-based experiment in 4118 UK adults via web-based surveys. Preferences were determined by asking respondents to allocate fixed funds between different patient and disease types reflecting nine specific prioritisation criteria. Respondents supported the criteria proposed under the VBP system (for severe diseases, address unmet needs, are innovative-provided they offered substantial health benefits, and have wider societal benefits) but did not support the end-of-life premium or the prioritisation of children or disadvantaged populations as specified by NICE, nor the special funding status for treatments of rare diseases, nor the CDF. Policies introduced on the basis of perceived-and not actual-societal values may lead to inappropriate resource allocation decisions with the potential for significant population health and economic consequences. Copyright © 2012 John Wiley & Sons, Ltd.

We sought to compare the performance of the EQ-5D and SF-6D with regard to the criteria of practicality, convergent validity, and construct validity, the level of agreement between the two measures was also assessed. Responses from 1865 individuals aged >or= 45 years in one general practice were analysed. Of these, 93.1% completed the EQ-5D, compared with 86.4% for the SF-6D, where individuals who were older, female, of a lower occupational skill level, from an area of lower deprivation, or used prescribed medication were significantly less likely to complete the SF-6D. The performance of both measures was comparable with regard to both convergent and construct validities, as both the EQ-5D and SF-6D scores were closely related to scores on the EuroQol visual analogue scale (VAS) (p<0.001) and able to discriminate between people who did and did not take: (i) analgesics and (ii) other prescribed medication. Despite EQ-5D and SF-6D scores being highly correlated (p<0.001), individuals who were healthier (according to the VAS) had higher mean scores on the EQ-5D (p<0.001), whereas less healthy individuals had higher mean scores on the SF-6D (individuals with knee pain, osteoarthritis, back pain, rheumatoid arthritis, and hip pain had significantly lower mean scores on the EQ-5D, p<0.001).

This paper examines the differences in health state evaluations given by patients when they are asked to value their own current states, and those given by members of the general population who were asked to value hypothetical health states. Patient data consist of 4137 observations on EQ-5D profiles and Visual Analogue Scale (VAS) obtained from 3376 patients, covering eight different conditions. General population data are taken from the EQ-5D valuation set. Two analyses were carried out. In the first, the patient self-rated VAS was compared with population VAS values for the same health states. In the second, the patient self-rated VAS values were modelled, and the regression coefficients were compared with the corresponding coefficients from the general population study. The first analysis resulted in a statistically significant mean difference of -0.012 (0.647 for patient VAS, 0.659 from the population value set). The second analysis found statistically significant differences between the coefficients for the EQ-5D health dimensions Pain/Discomfort, Mobility and Anxiety/Depression. Anxiety/Depression had the largest impact on the patient model compared with Pain/Discomfort in the general population model. A further regression analysis suggests that the magnitude of disagreement between patient self-rated VAS model and the population VAS model depends on the patients' condition.

In a landmark move, the UK Department of Health (DH) has introduced the routine collection of patient-reported outcome measures (PROMs) to measure the performance of health-care providers. From April 2009, generic (EQ-5D) and condition-specific PROMs are being collected from patients before and after four surgical procedures; eventually this will be extended to include a wide range of other NHS services. The aim of this article is to report analysis of the EQ-5D data generated from a pilot study commissioned by the DH and to consider the implications for the use of EQ-5D data in performance indicators and measures of patient benefit. We present two new methods that we have developed for analysing and displaying EQ-5D profile data: a Paretian Classification of Health Change and a health profile grid. We show that EQ-5D profile data can be readily analysed to generate insights into the nature of changes in patient-reported health that would be obscured by summarising these profiles by their index scores, or focusing just on the post operative outcomes. Our methods indicate differences between providers and between sub-groups of patients. Our results also show striking differences in changes in EQ-5D profiles between surgical procedures, which require further investigation.

Background: An important consideration for studies that derive utility scores using multi-attribute utility measures is the psychometric integrity of the measurement instrument. Of particular importance is the requirement to establish the empirical validity of multi-attribute utility measures; that is, whether they generate utility scores that, in practice, reflect people's preferences. We compared the empirical validity of EQ-5D versus SF-6D utility scores based on hypothetical preferences in a large, representative sample of the English population. Methods: Adult participants in the 1996 Health Survey for England (n=16 443) formed the basis of the investigation. The subjects were asked to complete the EQ-5D and SF-36 measures. Their responses were converted into utility scores using the York A1 tariff set and the SF-6D utility algorithm, respectively. One-way analysis of variance was used to test the hypothetically constructed preference rule that each set of utility scores differs significantly by self-reported health status (categorised as very good, good, fair, bad or very bad). The degree to which EQ-5D and SF-6D utility scores reflect alternative configurations of self-reported health status; illness, disability or infirmity, and medication use was tested using the relative efficiency statistic and receiver operating characteristic (ROC) curves. Results: The mean utility score for the EQ-5D was 0.845 (95% CI: 0.842, 0.849), whilst the mean utility score for the SF-6D was 0.799 (95% CI: 0.797, 0.802), representing a mean difference in utility score of 0.046 (95% CI: 0.044, 0.049; p<0.001). Bland-Altman plots displayed considerable lack of agreement between the two measures, particularly at the lower end of the utility scale. Both measures demonstrated statistically significant differences between subjects who described their health status as very good, good, fair, bad or very bad (p<0.001), as well as monotonically decreasing utility scores (test for linear trend: p<0.001). The SF-6D was between 30.9 and 100.4% more efficient than the EQ-5D at detecting differences in self-reported health status, and between 10.4 and 45.6% more efficient at detecting differences in illness, disability or infirmity and medication use. The area under the curve scores generated by the ROC curves were significantly higher for the SF-6D at the 0.1% significance level when self-reported health status was dichotomised as very good versus good, fair, bad or very bad. However, the AUC scores did not reveal any significant differences in the discriminatory powers of the measures when alternative configurations of illness, disability or infirmity and medication use were examined. Conclusions: This study provides evidence that the SF-6D is an empirically valid and efficient alternative multi-attribute utility measure to the EQ-5D, and is capable of discriminating between external indicators of health status. However, health economists should also consider other psychometric properties, such as practicality and reliability, when selecting either measure for evaluative purposes.

As more research is undertaken on the elderly, accurately assessing changes in their quality of life becomes increasingly important. Generic instruments are the most popular method to assess quality of life, and one of the most widely used is the EQ-5D. However, the range of dimensions, sensitivity of scales and completion rates have been raised as concerns when using this measure with the elderly. The AQoL is a newer instrument which offers greater richness in dimensions of health covered, and potentially offers greater sensitivity to changes in quality of life. This paper presents the results of a ‘head-to-head’ comparison of the EQ-5D and AQoL in terms of practicality, construct validity, agreement (of absolute scores and their change over time) and sensitivity to change, as part of a randomised controlled trial in the elderly. Poor agreement was found between both the absolute scores from each instrument and change in scores over time. Although the AQoL appeared to have more favourable construct validity, the EQ-5D was easier to administer, had a higher completion rate, and appeared more sensitive to change. We conclude that the AQoL is probably less well suited to measuring health status in a very elderly population than the EQ-5D. Copyright

Optimising the design of discrete choice experiments (DCE) involves maximising not only the statistical efficiency, but also how the nature and complexity of the experiment itself affects model parameters and variance. The present paper contributes by investigating the impact of the number of DCE choice sets presented to each respondent on response rate, self-reported choice certainty, perceived choice difficulty, willingness-to-pay (WTP) estimates, and response variance. A sample of 1053 respondents was exposed to 5, 9 or 17 choice sets in a DCE eliciting preferences for dental services. Our results showed no differences in response rates and no systematic differences in the respondents' self-reported perception of the uncertainty of their DCE answers. There were some differences in WTP estimates suggesting that estimated preferences are to some extent context-dependent, but no differences in standard deviations for WTP estimates or goodness-of-fit statistics. Respondents exposed to 17 choice sets had somewhat higher response variance compared to those exposed to 5 choice sets, indicating that cognitive burden may increase with the number of choice sets beyond a certain threshold. Overall, our results suggest that respondents are capable of managing multiple choice sets - in this case 17 choice sets - without problems.

We studied the relationship between current cigarette smoking and price among 34145 respondents, aged 15-29 years, to the 1992-1993 Tobacco Use Supplements to the Current Population Survey. The price elasticity of current smoking varied inversely with age: -0.831 (S.E. 0.402) for ages 15-17; -0.524 (S.E. 0.256) for ages 18-20; -0.370 (S.E. 0.188) for ages 21-23; -0.202 (S.E. 0.175) for ages 24-26; and -0.095 (S.E. 0.157) for ages 27-29. In response to higher prices, older youth were more likely to reduce the number of cigarettes smoked per day than to quit entirely. Among 15-17-year-olds, smoking cigarettes 'some days' was more sensitive to price than smoking 'every day'. Cigarette smoking was inversely related to the prices of premium brands, but not discount brands.

A general approach is discussed to assess the uncertainty surrounding the cost effectiveness ratio (C/E-ratio) estimated on the basis of data from a randomised clinical trial. The approach includes the calculation of a 95% probability ellipse and introduces the concept of a so called C/E-acceptability curve. This last curve defines for each predefined C/E-ratio the probability that the C/E-ratio found in the study is acceptable. The approach is illustrated by estimates of costs per life saved and costs per patient discharged alive on the basis of data from a phase II trial addressing the value of anakinra in treating sepsis syndrome.

Recent research has derived preference scores from the SF-36. We compare the practicality and construct validity of SF-36 derived preference scores with directly elicited time trade off (TTO) and visual analogue scale (VAS) scores. In this observational study, low back pain (LBP), patients were asked to complete disease specific, generic (SF-36), and health state preference (VAS and TTO) instruments. Baseline SF-36 responses were converted to preference scores using six published algorithms. Response rates for the SF-36 derived and TTO preference values were 354 of 379 (93%) and 303 of 379 (80%), respectively. Thirty patients were excluded from the TTO exercise because of difficulties comprehending the scaling task. Choice based methods (standard gamble, TTO) yielded higher and more uniform estimates of preference (0.77-0.79) than non-choice based methods (VAS) (0.42-0.70). Directly elicited TTO values were variable and had less power to distinguish among patients with differing severity of LBP. All SF-36 derived preferences exhibited a minimum threshold implying a potential floor effect for severely ill patients. SF-36 derived preferences demonstrated good practicality and construct validity in this setting, however different methods will yield disparate estimates of marginal benefit. This emphasises the need for a standardised algorithm for deriving SF-36 preference scores.

The criteria used by the National Institute for Health and Clinical Excellence (NICE) for accepting higher incremental cost-effectiveness ratios for some medicines over others, and the recent introduction of the Cancer Drugs Fund (CDF) in England, are assumed to reflect societal preferences for National Health Service resource allocation. Robust empirical evidence to this effect is lacking. To explore societal preferences for these and other criteria, including those proposed for rewarding new medicines under the future value-based pricing (VBP) system, we conducted a choice-based experiment in 4118 UK adults via web-based surveys. Preferences were determined by asking respondents to allocate fixed funds between different patient and disease types reflecting nine specific prioritisation criteria. Respondents supported the criteria proposed under the VBP system (for severe diseases, address unmet needs, are innovative-provided they offered substantial health benefits, and have wider societal benefits) but did not support the end-of-life premium or the prioritisation of children or disadvantaged populations as specified by NICE, nor the special funding status for treatments of rare diseases, nor the CDF. Policies introduced on the basis of perceived-and not actual-societal values may lead to inappropriate resource allocation decisions with the potential for significant population health and economic consequences. Copyright © 2012 John Wiley & Sons, Ltd.

We sought to compare the performance of the EQ-5D and SF-6D with regard to the criteria of practicality, convergent validity, and construct validity, the level of agreement between the two measures was also assessed. Responses from 1865 individuals aged >or= 45 years in one general practice were analysed. Of these, 93.1% completed the EQ-5D, compared with 86.4% for the SF-6D, where individuals who were older, female, of a lower occupational skill level, from an area of lower deprivation, or used prescribed medication were significantly less likely to complete the SF-6D. The performance of both measures was comparable with regard to both convergent and construct validities, as both the EQ-5D and SF-6D scores were closely related to scores on the EuroQol visual analogue scale (VAS) (p<0.001) and able to discriminate between people who did and did not take: (i) analgesics and (ii) other prescribed medication. Despite EQ-5D and SF-6D scores being highly correlated (p<0.001), individuals who were healthier (according to the VAS) had higher mean scores on the EQ-5D (p<0.001), whereas less healthy individuals had higher mean scores on the SF-6D (individuals with knee pain, osteoarthritis, back pain, rheumatoid arthritis, and hip pain had significantly lower mean scores on the EQ-5D, p<0.001).

Background: An important consideration for studies that derive utility scores using multi-attribute utility measures is the psychometric integrity of the measurement instrument. Of particular importance is the requirement to establish the empirical validity of multi-attribute utility measures; that is, whether they generate utility scores that, in practice, reflect people's preferences. We compared the empirical validity of EQ-5D versus SF-6D utility scores based on hypothetical preferences in a large, representative sample of the English population. Methods: Adult participants in the 1996 Health Survey for England (n=16 443) formed the basis of the investigation. The subjects were asked to complete the EQ-5D and SF-36 measures. Their responses were converted into utility scores using the York A1 tariff set and the SF-6D utility algorithm, respectively. One-way analysis of variance was used to test the hypothetically constructed preference rule that each set of utility scores differs significantly by self-reported health status (categorised as very good, good, fair, bad or very bad). The degree to which EQ-5D and SF-6D utility scores reflect alternative configurations of self-reported health status; illness, disability or infirmity, and medication use was tested using the relative efficiency statistic and receiver operating characteristic (ROC) curves. Results: The mean utility score for the EQ-5D was 0.845 (95% CI: 0.842, 0.849), whilst the mean utility score for the SF-6D was 0.799 (95% CI: 0.797, 0.802), representing a mean difference in utility score of 0.046 (95% CI: 0.044, 0.049; p<0.001). Bland-Altman plots displayed considerable lack of agreement between the two measures, particularly at the lower end of the utility scale. Both measures demonstrated statistically significant differences between subjects who described their health status as very good, good, fair, bad or very bad (p<0.001), as well as monotonically decreasing utility scores (test for linear trend: p<0.001). The SF-6D was between 30.9 and 100.4% more efficient than the EQ-5D at detecting differences in self-reported health status, and between 10.4 and 45.6% more efficient at detecting differences in illness, disability or infirmity and medication use. The area under the curve scores generated by the ROC curves were significantly higher for the SF-6D at the 0.1% significance level when self-reported health status was dichotomised as very good versus good, fair, bad or very bad. However, the AUC scores did not reveal any significant differences in the discriminatory powers of the measures when alternative configurations of illness, disability or infirmity and medication use were examined. Conclusions: This study provides evidence that the SF-6D is an empirically valid and efficient alternative multi-attribute utility measure to the EQ-5D, and is capable of discriminating between external indicators of health status. However, health economists should also consider other psychometric properties, such as practicality and reliability, when selecting either measure for evaluative purposes.