Comparison of valuation methods used to generate the EQ-5D and the SF-6D value sets

doi:10.1016/j.jhealeco.2005.09.003

Journal of Health Economics

Volume 25, Issue 2, March 2006, Pages 334-346

https://doi.org/10.1016/j.jhealeco.2005.09.003 Get rights and content

Abstract

An interview study with 101 members of public compared the protocols used in valuation studies for EQ-5D (using ranking, visual analogue scale, and time trade-off), and SF-6D (using ranking and standard gamble). Respondents were given one of the two protocols and asked to value four states each from EQ-5D and SF-6D. VAS scores suggest the narrower range of SF-6D values is partly attributable to the descriptive system; TTO values for milder states were higher than SG values; and the mean value for EQ-5D pits using TTO and SF-6D pits using SG were closer than across the two original valuation studies.

Section snippets

Background

The use of preference-based measures of health has grown considerably over the last decade with the increasing use of economic evaluation to inform health policy, for example through the establishment of bodies such as the National Institute for Clinical Excellence in England and Wales (NICE, 2001) and the Health Technology Board for Scotland (2002) and similar agencies in Australia (Commonwealth Department of Health, 1992) and Canada (Ministry of Health Ontario, 1994). Preference-based

The protocols and the variants

This study evaluated EQ-5D states using the “York protocol” (using TTO) and the “Sheffield protocol” (using SG), and evaluated SF-6D states also using these two protocols. There are several ways in which such studies could be carried out. The extreme would be to ask each respondent to do all four tasks (evaluate EQ-5D states with the two protocols, and then evaluate SF-6D states with the two protocols). More realistic designs are to split the sample so that one group evaluated EQ-5D states

Results

This section reports on the characteristics of the respondents (3.1), the results of the ranking exercise (3.2), the VAS results (3.3), the two questions regarding the “utility” scores (3.4), and the comparison between the current study and the two original studies (3.5). The main result concerning the effect of valuation protocols is reported in Section 3.4.2.

Discussion

This paper reports on a small-scale exploratory study comparing the effect of different methods for valuing health. The question we set off to investigate is what are the sources of the differences between EQ-5D indices and SF-6D indices. In an earlier paper (Brazier et al., 2004), we studied the descriptive system of the two instruments and concluded that the worst SF-6D states are less severe than the worst EQ-5D states so that EQ-5D indices for patients in severe states will be lower than

Acknowledgements

This study is funded by the EuroQol Foundation and the SF-6D Group. Survey and Statistical Research Centre, Sheffield Hallam University, carried out the sampling, interviewing, and data entry. We would like to thank the two anonymous referees, Richard Brooks, Paul Dolan, Chris McCabe, Stephen Walters, Tracey Young and all the respondents who took part. The usual disclaimer applies. An earlier version of this paper has been presented at the EuroQol Group Meeting, September 2002, York, UK.

References (18)

J. Brazier et al.
The estimation of a preference-based measure of health from the SF-36
Journal of Health Economics
(2002)
P. Dolan et al.
Valuing health states: a comparison of methods
Journal of Health Economics
(1996)
P. Dolan et al.
Mapping visual analogue scale health state valuations onto standard gamble and time trade-off values
Social Science and Medicine
(1997)
J. Brazier et al.
A comparison of the EQ-5D and SF-6D across seven patient groups
Health Economics
(2004)
J.E. Brazier et al.
A checklist for judging preference-based measures of health related quality of life: learning from psychometrics
Health Economics
(1999)
Commonwealth Department of Health, Housing and Community Service, 1992. Guidelines for the Pharmaceutical Industry on...
P. Dolan
Modeling valuations for EuroQol health states
Medical Care
(1997)
D.H. Feeny et al.
Multiattribute and single-attribute utility function the health utility index Mark 3 system
Medical Care
(2002)
M.R. Gold et al.
Cost-Effectiveness in Health and Medicine
(1996)

There are more references available in the full text version of this article.

Cited by (55)

The EQ-5D-5L Value Set for England: Response to the “Quality Assurance”
2020, Value in Health
Citation Excerpt :
The fact is that such a difference was predictable before our study: although the MVH value set has been recommended for use by NICE for more than 20 years, the unusual nature of this value set, especially the high proportion of negative values and the wide utility range, are widely known and reported. Studies undertaken in the United Kingdom using the same protocol in the years immediately after its publication were unable to replicate its properties.22 A new value set, whether for the EQ-5D-3L or EQ-5D-5L, is unlikely to recreate the characteristics of the MVH value set.
To respond to the ‘quality assurance’ of the EQ-5D-5L value set for England study.
We provide a point-by-point response to the issues raised by the authors of the quality assurance paper, drawing on theoretical arguments, empirical analyses and practical considerations.
We provide evidence to show that many of the points made by the authors of the quality assurance are misleading, suggest misunderstandings, or are irrelevant.
The modeling approaches which were used appropriately address the characteristics of the data and provide a reasonable representation of the average stated preferences of general public in England. We provide reflections on the conduct of stated preference studies, and suggestions for the way forward.
Measuring the impact of alcohol-related disorders on quality of life through general population preferences
2017, Gaceta Sanitaria
Citation Excerpt :
We used the probability lottery–equivalent method,27,28 a variant of the lottery-equivalent method,29 to derive utility weights. There is empirical evidence suggesting that this method mitigates the overvaluation of health states from the “standard gamble” approach.28,30 Another advantage of our approach is that the same procedure can be used to estimate utilities both of states better and worse than dead.
To estimate the intangible effects of alcohol misuse on the drinker's quality of life, based on general population preferences
The most important effects (dimensions) were identified by means of two focus groups conducted with patients and specialists. The levels of these dimensions were combined to yield different scenarios. A sample of 300 people taken from the general Spanish population evaluated a subset of these scenarios, selected by using a fractional factorial design. We used the probability lottery equivalent method to derive the utility score for the evaluated scenarios, and the random-effects regression model to estimate the relative importance of each dimension and to derive the utility score for the rest of scenarios not directly evaluated.
Four main dimensions were identified (family, physical health, psychological health and social) and divided into three levels of intensity. We found a wide variation in the utilities associated with the scenarios directly evaluated (ranging from 0.09 to 0.78). The dimensions with the greatest relative importance were physical health (36.4%) and family consequences (31.3%), followed by psychological (20.5%) and social consequences (11.8%).
Our findings confirm the benefits of adopting a heterogeneous approach to measure the effects of alcohol misuse. The estimated utilities could have both clinical and economic applications.
Estimar los efectos intangibles del consumo abusivo de alcohol en la calidad de vida del bebedor, según las preferencias sociales.
Los efectos más relevantes se identificaron mediante dos grupos focales realizados con pacientes y especialistas. Los niveles de estas dimensiones se combinaron para producir diferentes escenarios. Una muestra de 300 personas de la población general española evaluó un subconjunto de estos escenarios, seleccionados mediante un diseño factorial fraccional. Se utilizó el método de lotería equivalente para obtener la utilidad asociada a cada uno de los escenarios evaluados. Para estimar la importancia relativa de cada dimensión y obtener la utilidad para el resto de escenarios no evaluados se estimó una regresión con efectos aleatorios.
Se identificaron cuatro efectos intangibles relevantes (familia, salud física, salud psicológica y social) con tres niveles de intensidad. Las utilidades asociadas a cada uno de los escenarios evaluados presentan una amplia variación (entre 0,09 y 0,78). La dimensión con mayor importancia relativa son las consecuencias en la salud física (36,4%) y las consecuencias en la familia (31,3%), seguidas de las consecuencias psicológicas (20,5%) y las sociales (11,8%).
Nuestros resultados confirman la conveniencia de adoptar un enfoque heterogéneo para medir los efectos del abuso del alcohol. Las utilidades estimadas podrían tener aplicaciones tanto clínicas como económicas.
An exploratory study to test the impact on three "bolt-On" items to the EQ-5D
2015, Value in Health
Generic preference-based measures were criticized for being inappropriate in some conditions. One solution is to include “bolt-on” dimensions describing additional specific health problems.
This study aimed to develop bolt-on dimensions to the EuroQol five-dimensional questionnaire (EQ-5D) and assess their impact on health state values.
Bolt-on dimensions were developed for vision problems, hearing problems, and tiredness. Each bolt-on dimension had three severity levels to match the EQ-5D. Three “core” EQ-5D states across a range of severity were selected, and each level of a bolt-on item was added, resulting in nine states in each condition. Health states with and without the bolt-on dimensions were valued by 300 members of the UK general public using time trade-off in face-to-face interviews, and mean health state values were compared using t tests. Regression analysis examined the impact of the bolt-on variants and the level of the bolt-on items after controlling for sociodemographic characteristics.
Bolt-on dimensions had an impact on health state values of the EQ-5D; however, the size, direction, and significance of the impact depend on the severity of the core EQ-5D state and of the bolt-on dimension. Regression analysis demonstrated that after controlling for possible differences in sociodemographic characteristics between the groups, there were no significant differences in health state values between the three bolt-on dimensions but confirmed that the impact depended on the severity of the EQ-5D health state and the levels of bolt-on dimensions.
The impact of a bolt-on dimension on the EQ-5D depends on the core health state and the level of the bolt-on dimension. Further research in this area is encouraged.
Estimating the burden of disease in chronic pain with and without neuropathic characteristics: Does the choice between the EQ-5D and SF-6D matter?
2014, Pain
Citation Excerpt :
The EQ-5D [18] and Short Form (SF) 12/36 [53,54] questionnaires are widely used generic HRQoL measures. They are a common means of generating health state values using an algorithm to derive health utility scores (SF-6D from SF-12/36) [5] and are used by economists to calculate quality adjusted life years (QALYs) [35,49,57] and in economic evaluations to evaluate the cost effectiveness of health care interventions. The association between neuropathic pain conditions and health utilities has been described in a multicenter European cross-sectional survey [33] and a systematic review [16], both finding a significant relationship between increasing pain severity and reduced HRQoL.
The EQ-5D and Short Form (SF)12 are widely used generic health-related quality of life (HRQoL) questionnaires. They can be used to derive health utility index scores, on a scale where 0 is equivalent to death and 1 represents full health, with scores less than zero representing states “worse than death.” We compared EQ-5D or SF-6D health utility index scores in patients with no chronic pain, and chronic pain with and without neuropathic characteristics (NC), and to explore their discriminant ability for pain severity. Self-reported health and chronic pain status was collected as part of a UK general population survey (n = 4451). We found moderate agreement between individual dimensions of EQ-5D and SF-6D, with most highly correlated dimensions found for mental health and anxiety/depression, role limitations and usual activities, and pain and pain/discomfort. Overall 43% reported full health on the EQ-5D, compared with only 4.2% on the SF-6D. There were significant differences in mean utilities for chronic pain with NC (EQ-5D 0.47 vs SF-6D 0.62) and especially for severe pain (EQ-5D 0.33 vs SF-6D 0.58). On the EQ-5D, 17% of those with chronic pain with NC and 3% without NC scored “worse than death,” a state which is not possible using the SF-6D. Health utilities derived from EQ-5D and SF-12/36 can discriminate between group differences for chronic pain with and without NC and greater pain severity. However, the instruments generate widely differing HRQoL scores for the same patient groups. The choice between using the EQ-5D or SF-6D matters greatly when estimating the burden of disease.
Comparison of contemporaneous EQ-5D and SF-6D responses using scoring algorithms derived from similar valuation exercises
2014, Value in Health
Citation Excerpt :
What is unclear is whether differences in index scores are a result of variation in the respective descriptive systems or methods of valuation, or both. Previous attempts to look at this suggest that differences in the valuation technique may account for some of the differences [34,35]. Furthermore, a comparative evaluation of multiple generic instruments using item response theory demonstrated that a simple linear function will transform one index score into the other for poorer health states [36].
Poor agreement between preference-based health-related quality-of-life instruments has been widely reported across patient and community-based samples. This study compares index scores generated from contemporaneous EQ-5D (3-level version) and SF-6D (SF-36 version) responses using scoring algorithms derived from independently-conducted Australian population-representative discrete choice experiments (DCEs), providing the first comparative analysis of health state valuations using the same method of valuation across the full value sets.
EQ-5D and SF-6D responses from seven patient data sets were transformed into health state valuations using published DCE-derived scoring algorithms. The empirical comparative evaluation consisted of graphical illustration of the location and spread of index scores, reporting of basic descriptive statistics, exploration of between-measure differences in mean index scores, and analysis of agreement.
Compared with previously published findings regarding the comparability of “conventional” EQ-5D and SF-6D index scores, health state valuations from the DCE-derived scoring procedures showed that agreement between scores remained “fair” (intraclass correlation coefficient values across the seven data sets ranged from 0.375 to 0.615). Mean SF-6D scores were significantly lower than the respective mean EQ-5D score across all patient groups (mean difference for the whole sample = 0.253).
The magnitude of disagreement previously reported between EQ-5D and SF-6D index scores is not ameliorated through the application of DCE-derived value sets; sizeable discrepancies remain. These findings suggest that differences between EQ-5D and SF-6D index scores persist because of their respective descriptive systems. Further research is required to explore the implications of variations in the descriptive systems of preference-based instruments.
Sick but satisfied: The impact of life and health satisfaction on choice between health scenarios
2013, Journal of Health Economics
Citation Excerpt :
Many HTA agencies recommend the use of public values; that is, preferences over hypothetical health states. In principle, these methods place all health states on an interval scale between 0 (for dead) and 1 (for full health), thus allowing for the calculation of QALYs – see Dolan (1997) for an EQ-5D ‘tariff’ based on TTO valuations from the UK general public and Tsuchiya et al. (2006) for a comparison of methods used to elicit preferences. Typical preference elicitation studies present the respondents with hypothetical health states described, for instance by EQ-5D, with no further information.
Preference elicitation methods require respondents to predict the impact a change in health might have on their future selves. The focus on the change in health is at the possible expense of other experiences of life once in that health state. We analyse personal preferences to a pairwise choice task involving trade-offs between quality and length of life, where satisfaction levels with life or health are introduced in the description of the health states. We find that a health scenario including low levels of satisfaction increases the likelihood of preferring to die sooner in full health, whereas scenarios including high levels of satisfaction increase the likelihood of preferring to live for longer in poor health. The differences highlight the sensitivity of preferences to what is described in health states and therefore show the importance of on-going discussions about precisely what respondents should be asked to consider in preference elicitation studies.

View all citing articles on Scopus

View full text

Comparison of valuation methods used to generate the EQ-5D and the SF-6D value sets

Abstract

Section snippets

Background

The protocols and the variants

Results

Discussion

Acknowledgements

Journal of Health Economics

Journal of Health Economics

Social Science and Medicine

A comparison of the EQ-5D and SF-6D across seven patient groups

Health Economics

A checklist for judging preference-based measures of health related quality of life: learning from psychometrics

Health Economics

Modeling valuations for EuroQol health states

Medical Care

Multiattribute and single-attribute utility function the health utility index Mark 3 system

Medical Care

Cost-Effectiveness in Health and Medicine