Measuring and Valuing Health

This programme of research is composed of various related projects funded from a range of sources including MRC, NIHR, pharmaceutical companies and charities.

A stethoscope sitting on top of a data print out.
Off

On this page

See also

The last two decades have seen increasing use of economic evaluation to inform resource allocation in health care systems around the world (eg NICE). A core issue for economic evaluation is the way the benefits of health care are measured and valued. A widely used technique of economic evaluation in health care has been cost-effectiveness analysis using Quality Adjusted Life Years (QALYs) to assess effectiveness in units that are comparable across health care interventions. Commonly used methods for putting the 'Q' into the QALY are generic preference-based measures of health, such as the EQ-5D, SF-6D, HUI2 and the HUI3. These generic measures have been adopted by agencies such as NICE to populate their models of the cost-effectiveness of interventions.

The overall aim of the programme is to address the key questions of what should be valued (eg health or broader notions of wellbeing); how health and wellbeing should be described; how health and wellbeing should be valued; who should do the valuing and how to develop new measures.

Generic preference-based measures of health

Following on from the development of the SF-6D, research has continued with the replication of this work in countries around the world (China, Japan, Portugal, Brazil, Australia and Singapore). The research team has also been pursuing new methods for analyzing preference data, including the use of Bayesian methods. Several HEDS staff members are members of the EuroQoL Group. Members of HEDS are also involved in an MRC funded project, the Preparatory study for the Re-evaluation of the EQ-5D Tariff (PRET). Staff at ScHARR have also developed a new measure for children using mixed methods analysis.

Reviewing the appropriateness of generic measures

Health economists have often favoured generic measures in order to allow cross programme comparison. However this pre-supposes that they are valid and responsive in all groups of patients. We are undertaking a number of reviews of the validity of generic measures in vision, mental health, diabetes and cancer.

What to do when generic measures are not available

Many pivotal studies of new health care interventions do not use one of the generic preference-based measures due to concerns about patient burden or validity, but other non-preference based measures of health (eg condition-specific measures). One solution to this problem is to estimate functions that 'map' or 'cross-walk' condition-specific measures onto a generic preference-based measure. Research at ScHARR has critically reviewed these methods and examined the policy implications of errors in these models. A team at ScHARR also has just completed a research project for MRC examining a new method for mapping based on using preference data.

Development of condition-specific preference-based measures of health

Where generic measures are not regarded as appropriate, then the alternative is to develop a specific measure more relevant to the condition. This can involve either developing news measures from scratch or modifying an existing non-preference based measure for this purpose. Staff at ScHARR have developed measures from scratch for people with leg ulcers, and are currently working on a measure for children with amblyopia and a measure for pressure ulcers. A mixed-methods approach including qualitative research with patients and psychometric testing has been used to develop the health states.

To make the most of available evidence, a better strategy might be to develop a preference-based measure from an existing and widely accepted condition-specific measure. We are currently working on developing preference-based measures for dementia, cancer and common mental health problems and have completed measures for asthma and overactive bladder. This work has involved the application of a range of psychometric methods (ie Rasch analysis) to derive health state classifications and then surveys to obtain general population values for calculating QALYs. We have also just completed the COSMeQ project examining the methodological issues in developing condition-specific preference-based measures.

Valuation techniques

Conventional techniques for valuing health states have been Time Trade-Off (TTO), Standard Gamble (SG) and Visual Analogue Scaling (VAS). There are concerns that TTO and SG especially are too complex in vulnerable groups and that the values generated may be distorted by extraneous factors such as time preference and loss aversion that making it difficult to ascertain the value for health per se for calculating QALYs. For this reason, we have been looking at the use of ordinal methods, such as ranking and Discrete Choice Experiments (DCE), for valuing health states in this programme. The key contribution has been to develop methods for valuing health states on the full health-dead scale required for calculating QALYs. There have also been concerns with the way states worse than dead are handled in conventional TTO and we have been involved in the development and testing of a new 'lead time' TTO.

Whose values?

Our work has provided further evidence that the values put on health states by members of the general public and patients diverge. The more original contribution has been to show that people experiencing the state (eg patients) place a different weight on the dimensions of health to members of the general public trying to imagine what the state is like. One finding, for example, has been that people in the states place a higher relative value on mental health compared to physical health.

The divergence between patient and general public values has been explained in terms of adjustment and coping mechanisms, at least to physical health problems. A PhD has recently been completed exploring the impact on health state values of providing better information to general population respondents on these processes. We are also currently undertaking projects examining how values differ for patients and general population for epilepsy health states, and how values differ for patients, carers and general population for dementia health states.

Health or wellbeing?

An important question is whether the NHS should be primarily concerned with promoting health, or some broader notion of wellbeing. We are collaborating in the development of broader measures, including a preference-based measure of capability (ICECAP) and a measure for social care (OSCAR). We have also examined the relationship between health and broader measures of wellbeing in a micro-study of patients experiencing health change using mixed methods research. Future work will include the development of a wellbeing scale that can be used to calculate QALYs.

ScHARR is a major partner in CWiPP, the University Centre for Health and Wellbeing in Public Policy that focuses beyond conventional health and encourages collaboration between departments in the University in this exciting area of public policy research.

How can licenses for the Health Measures be obtained?

Available measures are copyrighted and are available on a license basis.

There are three forms of license:

  • A license is available free of charge for all non-commercial applications including work funded by research councils, government agencies and charities.
  • For commercial applications, there will be a per study license (eg clinical trial), though an open license for a fixed period is available.
  • The SF-6D measure is being used in software available from Quality Metric. (www.qualitymetric.com).

Income generated from the charges will go into a University-based research fund.

To obtain more information about obtaining licences please use the following link

Licenses are available for the following Measures:

  • SF-6D
  • SPVU-5D (evaluates the health-related quality of life impact of venous ulceration treatments)
  • CAT-QOL (Children’s Amblyopia Treatment Quality of Life Questionnaire)
  • CHU9D (Paediatric Quality of Life)
  • EORTC-8D (enables widely used cancer measure EORTC QLQ-C30 data to be used to directly calculate QALYs)

See more at https://licensing.sheffield.ac.uk/products/health-outcomes/list

Our work

E-QALY

Visit E-QALY website

Health Utilities Index (HUI2)

Overview

The HUI2 is the only preference-based multi-attribute health-related quality of life instrument specifically developed for use with children. It consists of seven dimensions (sensation, mobility, emotion, cognition, self-care, pain and fertility), each of which has between three and five levels. The levels range from "normal functioning for age" to "extreme disability".

Preference-based quality of life weights can be calculated for all health states in the descriptive system using a multiplicative multi-attribute utility function (MAUF) developed by Torrance and colleagues. This is based on interviews with parents of school-age children in Hamilton, Ontario, Canada. (Torrance GW et al. A multi-attribute utility function for a comprehensive health status classification system: Health Utilities Mark 2 Medical Care 1996;34(7):702-722)

As part of an MRC funded study on risk adjustment in paediatric intensive care (the UK PICOS project), we undertook a UK valuation of the HUI2.

Three health state valuation surveys were undertaken with 450 members of the UK general population. We estimated a multi-attribute utility function algorithm in the first survey, a statistical inference valuation algorithm in the second survey and compared the predictive performance of these algorithms in the third validation survey. We proposed alternative methods and models to the original Canadian algorithm and identified the new UK statistical inference valuation algorithm to be superior to both the Canadian and UK Multi-attribute utility valuation models in terms of predictive performance.

The results of the different UK models and associated work are published in the papers listed below.

Papers

  • Kharroubi, S. McCabe, C. Modelling HUI 2 health state preference data using a nonparametric Bayesian method. Medical Decision Making (forthcoming)
  • Stevens K, McCabe C J, Brazier J E, Roberts J. Multi-attribute utility function or statistical inference models: a comparison of health state valuation models using the HUI2 health state classification system. Journal of Health Economics. 2007; 26 (5); 992-1002.
  • Stevens K, McCabe C J, Brazier J E. Response to Shmueli. Health Economics Letters. 16: 759-761 (2007)
  • McCabe C, Stevens K, Roberts J, Brazier J E. Health State Values for the HUI2 descriptive system: results from a UK Survey. Health Economics 2005; 14 (3). [Link]

Please note: Since this paper was published, some errors in Table 1 and Table 3 have come to our attention. Most notably the health state descriptors in Table 1 were incorrect; and the coefficients reported in Table 3 for the Random Effects model, were the same as those for the OLS model. In order to ensure that these were in fact errors of presentation and transcription, we have redone the analyses reported in the original paper. Corrected versions of Table 1 and Table 3 can be downloaded above.

The discussion and conclusions of the original paper, including the preferred model for UK HUI2 health state values, are not affected by these errors.

  • McCabe C, Stevens K, Brazier J E. Utility values for the Health Utility Index Mark 2: An empirical assessment of alternative mapping functions. Medical Care, 2005, 43:627-635. [Medical Care Website]
  • McCabe C., Brazier JE. Gilks P. et al Estimating population cardinal health state valuation models from individual ordinal (rank) health state preference data. JHE, 2006, 25(3):418-431
  • Stevens K, McCabe C, Brazier J. Mapping between Visual Analogue Scale and Standard Gamble; Results from the UK Health Utilities Index 2 valuation survey. Health Economics, 2006, 15(5):527-533 [Link]
  • HEDS Discussion Paper 04/6 McCabe C, Stevens K. Visual Analogue Scales: do they have a role in the measurement of preferences for Health States? 2004.

Algorithm

Our preferred algorithm is the UK statistical inference valuation algorithm; however, should you wish to use a MAUF model, our preferred algorithm is the cubic MAUF. To obtain either algorithm, please use this link.

Further Information

If you would like further information about this work, please contact Katherine Stevens on K.Stevens@Sheffield.ac.uk.

See also

Download

CHU9D (Paediatric Quality of Life)

Obtain a license

The development of a paediatric health-related quality of life measure for use in economic evaluation: The Child Health Utility 9D (CHU9D)

Background

Use of economic evaluation to aid NHS decision making is widespread. Cost-utility analysis allows comparison of interventions both within and between disease areas by using outcome measures that combine length of life and quality of life into a single summary measure, conventionally the quality adjusted life year (QALY). Generic preference-based health-related quality of life measures have been developed for adults for this purpose. Research in this area for paediatric populations is more limited.

Research Project

In 2005 Katherine Stevens was awarded an MRC Special Training Fellowship to develop a paediatric health-related quality of life measure for use in economic evaluation, the Child Health Utility 9D (CHU9D). In 2009 she was awarded an ESRC/MRC/NIHR Early Career Post Doctoral Fellowship in the Economics of Health to further develop and apply this new measure.

Development of the CHU9D

The first stage of the research developed the descriptive system. Interviews were carried out with over 70 children during 2006, from two schools in Sheffield, in order to determine what dimensions of health-related quality of life were included. These dimensions were then used as the basis for developing the descriptive system.

This descriptive system was then piloted with 150 children in schools.

Schools involved in the research:

Further testing of the draft descriptive system was carried out on a clinical population at the Sheffield Children's Hospital, in collaboration with the newly established Clinical Research Facility. This work tested the psychometric performance of the instrument on 95 children, including children from the medical and surgical wards as well as daycare patients. The results of this work and the piloting in schools were used to refine the descriptive system.

(Children's Clinical Research Facility)

The final stage of the work was to obtain values for the health states described by the descriptive system from a sample of the UK general population using the standard gamble method.

Current Research

There are currently many research studies applying the CHU9D, including clinical trials, observational studies, cohort studies and others. These studies are taking place in many different countries worldwide.

There are now a large number of translations available including: Brazil, China, Canada, France, Germany, India, Mexico, Portugal, Russia, Spain and USA. See the link below on information for academics/researchers for further details.

There are also preference weights for Australia, China and the Netherlands that have been generated with collaborators, and for Australia and China these were generated using adolescent preferences.

Information for researchers and other users

What is the CHU9D?

The CHU9D is a paediatric generic preference-based measure of health-related quality of life. It consists of a descriptive system and a set of preference weights, giving utility values for each health state described by the descriptive system, allowing the calculation of quality adjusted life years (QALYs) for use in cost-utility analysis. The descriptive system and the preference weights are now available for use.

How was the CHU9D developed?

The CHU9D has been developed exclusively with children. The dimensions are based on interviews with over 70 children with a wide range of health problems. Further ranking work with children was undertaken to develop scales for the dimensions and a draft descriptive system was produced. This was tested on 150 school children and 98 children in hospital (including medical, surgical and day-case patients) and subsequently refined to produce the final descriptive system.

The Descriptive System (Questionnaire)

  • Application: generic
  • Number of dimensions (questions): 9
  • Number of levels (response options) per dimension: 5
  • Age range: 7-17 years
  • Mode of completion: self-completion (proxy completion also available for younger children)
  • Recall period: today/last night

You can view the descriptive system in the downloads box above, however, please note that this is to view only so that you can decide if you are interested in using the CHU9D.

Preference Weights (UK)

Preference weights for the CHU9D were obtained from a sample of the UK adult general population using the recognised valuation technique of standard gamble. Members of the general population were asked to value a selection of health states from which a model was estimated to predict all the health states described by the CHU9D.

Preference Weights (Australian)

A recent study has obtained Australian preference weights for the CHU9D using profile case best worst scaling methods. These are based on the preferences of Australian adolescents aged 11 to 17 years. If you would like to use these preference weights, please indicate this when completing your licence application. Alternatively, you can contact Professor Julie Ratcliffe, julie.ratcliffe@flinders.edu.au who led the work. Julie will also be able to provide you with the algorithms. Please note the descriptive system is only available through the University of Sheffield.

Applications

Whilst the measure was originally developed with children aged 7-11 years, since then it has been validated in an adolescent population (11-17 years). Details can be found on the publications page. Some studies are also underway trialling a proxy version with children age 5-7 years and a proxy version with guidance notes for children under 5 years. If you are interested in using it in the under 5 age group, please contact Katherine Stevens (k.stevens@sheffield.ac.uk) or Donna Rowen (d.rowen@sheffield.ac.uk) to discuss this. If you are interested in using the proxy version for children age 5-7 years, then please select this option on the licence request form.

Translations

There is the original UK English version and a large number of other translations are available, including: Argentina, Australia, Austria, Belguim (Dutch, French), Brazil, Canada (English, French), Chile, China, Croatia, Czech Republic, Denmark, Estonia (Estonian, Russian), France, Germany, Guatemala, India (Bengali, English, Hindi, Telugu), Israel (Arabic, Hebrew, Russian), Italy, Japan, Latvia (Latvian, Russian), Lithuania (Lithiuanian, Russian), Mexico, Netherlands, Poland, Portugal, Romania, Russia, Serbia, Slovakia, Spain, Sweden, Taiwan, Turkey, Ukraine (Ukrainian, Russian), USA (Spanish and English) and Wales. All translations have undergone a full linguistic validation process. If you require any of these, please indicate this on your licence application form.

How can the CHU9D be obtained?

The CHU9D is copyrighted and is available on a license basis.

There are two forms of licence:

  1. A license is available free of charge for all non-commercial applications including work funded by research councils, Government agencies and charities.
  2. For commercial applications, there will be a charge and the price will be negotiated upon application.

Even though it is free for non-commercial applications, we ask that you register your study with us so that we can keep a record of where the measure has been applied.

If you are a non-commercial user, please use the link to complete the non-commercial end-user licence registration form and we will contact you on receipt of this.

If you are a commercial user, please use the link to complete the commercial end-user licence registration form and we will contact you upon receipt of this.

Download final questionnaire

CHU9D Publications and Presentations

If you have any queries, please contact Katherine Stevens (k.stevens@sheffield.ac.uk) or Donna Rowen (d.rowen@sheffield.ac.uk).

SF-6D

Obtain a license

A brief overview

The SF-36 has become the most widely used measure of general health in clinical studies throughout the world. It currently generates eight dimension scores and two summary scores for physical and mental health. Whilst such scores provide an excellent means for judging the effectiveness of health care interventions, they have only a limited application in economic evaluation because they are not based on preferences.

The SF-6D provides a means for using the SF-36 and SF-12 in economic evaluation by estimating a preference-based single index measure for health from these data using general population values. The SF-6D allows the analyst to obtain quality adjusted life years (QALYs) from the SF-36 for use in cost-utility analysis.

What is the SF-6D?

The SF-6D is a classification for describing health derived from a selection of SF-36 items. It is composed of six multi-level dimensions. Any patient who completes the SF-36 or the SF-12 can be uniquely classified according to the SF-6D. The SF-6D describes 18,000 health states in all.

See this video for more information 

How does the SF-6D generate preference scores?

The SF-6D comes with a set of preference weights obtained from a sample of the general population using the recognised valuation technique of standard gamble. Members of the general population were asked to value a selection of health states from which a model has been estimated to predict all the health states described by the SF-6D.

How can the SF-6D be obtained?

A users manual and computer programmes can be obtained from here. The SF-6D is copyrighted and is available on a license basis.

There are three forms of licence:

  • A license is available free of charge for all non-commercial applications including work funded by research councils, Government agencies and charities.
  • For commercial applications, there will be a per study license (eg clinical trial), though an open license for a fixed period is available.
  • The SF-6D is being used in software available from Optum Insight. (https://www.optum.com)

Income generated from the charges will go into a University-based research fund.

For information about obtaining a copy of the SF-6D please use the University of Sheffield Licensing site.

Bayesian programme

A new excel programme is now available from the University of Sheffield to convert SF-36 data into the SF-6D utility score estimated using a set of non-parametric Bayesian preference weights. This approach has been shown to perform better in terms of predictive ability (mean absolute error of 0.089 compared to 0.104 on out of sample predictions) and overcomes the bias of the original regression models of under predicting the worst health states (e.g. it predicts a value of 0.203 for the worst SF-6D state compared to 0.301 using the original algorithm). The overall impact on mean health state values was between 0.01 and 0.04 across 4 data sets.

It is recommended that researchers use this algorithm in future work, but use the original algorithm for comparability.

How can the Bayesian programmes for the SF-6D be obtained?

To obtain the computer programs please complete the online user licence registration form using this link. We will then contact you about your request.

  • A licence is available free of charge for all non-commercial applications including work funded by research councils, government agencies and charities.
  • For commercial applications, there will be a per study licence (eg clinical trial), though an open license for a fixed period is available.

References

  • Kharroubi SA, Brazier JE, Roberts J, O´Hagan A. Modelling SF-6D health state preference data using a nonparametric Bayesian method. Journal of Health Economics 2007; 26:597-612
  • Kharroubi S, O'Hagan A, Brazier J. Estimating utilities from individual health preference data: a nonparametric Bayesian method. Applied Statistics 2005; 54:879-895

Valuation surveys in other countries

There is emerging evidence that SG health state values differ between countries for SG. There have been valuation surveys completed in Japan, Hong Kong, Portugal, Brazil and Spain using similar methods to those used in the UK. Details on these surveys can be found in the publications listed at the bottom of this note. Surveys have also been undertaken in Australia and Singapore, but the results have yet to be published.

Researchers interested in using the SF-6D in these countries should contact the names listed below:

Australia

Rosalie Viney: rosalie.viney@chere.uts.edu.au
Brendan Mulhern: Brendan.Mulhern@chere.uts.edu.au

Brazil

Luciane Cruz:lncruz@ig.com.br
Marcelo Fleck: mfleck.voy@zaz.com.br

Hong Kong

Cindy Lam: clklam@hku.hk
Sarah McGhee: smmcghee@hkucc.hku.hk

Japan

Shunichi Fukuhara: fukuhara@pbh.med.kyoto-u.ac.jp

Portugal

Lara Nobre Noronha Ferreira: lnferrei@ualg.pt
Pedro Lopes Ferreira: pedrof@fe.uc.pt

Singapore

Nan Luo: medln@nus.edu.sg

Spain

José María Abellán: dionisos@um.es

Publications on valuation surveys conducted in other countries

  • Lam CLK, Brazier J, McGhee SM. Valuation of the SF-6D health states is feasible, acceptable, reliable and valid in a Chinese population. Value in Health 2008;11:295-303.
  • Brazier JE, Fukuhara S, Roberts J, Kharoubi S et al. Estimating a preference-based index from the Japanese SF-36. Journal of Clinical Epidemiology; 62(12): 1323-1331.
  • Ferreira LN, Ferreira PL, Brazier J, Rowen D. A Portugese value set for the SF-6D. Value in Health 2010; 13(5): 624-630.
  • Abellan-Perpiñan  JM, Sanchez-Martinez FI, Martinez-Perez JE, Mendez I.  Lowering the 'floor' of the SF-6D scoring algorithm using a lottery equivalent method.  Health Economics 2012: 21; 1271-1285.
  • Méndez I, Abellán JM, Sanchez FI, Martinez JE.  Inverse probability weighted estimation of social tariffs: An illustration using SF-6D value sets.  Journal of Health Economics 2011: 30; 1280-1292.

Revised SF-6D scoring programmes

The SF-6D scoring programmes have recently been revised primarily in order to more accurately deal with missing SF-36/SF-12 item-level data. The table below summarises the issues raised by the revisions, the decisions implemented in the scoring programmes and the benefit to the user.

Issue Decision implemented in revised scoring programmes Benefit to user
Range checking for raw data An out of range value is generated for missing dimensions Clear indication of out of range values
Missing SF-36/SF-12 item values An SF-6D score can be computed with missing values for an item, if the score on that item would have no influence on the SF-6D score The SF-6D score can be computed with inessential missing values
Scoring inconsistency Values on the variable defining the worst state should take precedence over values defining better states Consistent set of weights used
Weighting of domain scores Weighting of domain scores from Brazier and Roberts (2004) Most recent published weights used
Different versions of the SF-36 and SF-12 Different scoring programmes are available for the different SF-36 and SF-12 versions with explicit explanation for which programme is relevant Ease of use of programmes for all versions of SF-36 and SF-12
Recoding of items in different versions of the SF-36 and SF-12 Each item response requiring random recoding is recoded independently Independent random assignment of SF-6D dimensions

The changes implemented in the revised SF-6D scoring programmes were agreed by all 3 previous providers of the programmes: John Brazier (Sheffield), Quality Metric and Dennis Fryback and Janel Hanmer (University of Wisconsin-Madison).

Refer to PRO newsletter No. 40 Fall Issue 2008 for details on the potential differences in the computed SF-6D score using the original and revised SF-6D scoring programmes for a dataset with seven patient groups.

New programmes available

A new excel programme is now available from the University of Sheffield to convert SF-36 data into the SF-6D utility score estimated using a set of non-parametric Bayesian preference weights. These nonparametric preference weights are an improvement on the parametric preference weights as the nonparametric model has many advantages over the conventional parametric random effects model which is reflected in improvements in the predictive ability of the model. For further details see Kharroubi et al. (2007).

Furthermore, a new excel programme is available to convert SF-36 data into the SF-6D utility score estimated using a set of preference weights obtained using an ordinal valuation technique for a sample of the general population. The estimates using ordinal data represent an alternative value set based on a different valuation technique which produces estimates that are comparable to estimates produced using standard gamble data. For further details see McCabe et al. (2006).

References

  • Brazier, JE, Roberts, JR,. The estimation of a preference-based index from the SF-12. Medical Care, 2004;42(9):851-859
  • Brazier, JE, Rowen, D, Hanmer, J,. Revised SF-6D scoring programmes: a summary of improvements. PRO newsletter, 2008;40:14-15
  • Kharroubi SA, Brazier JE, Roberts J, O´Hagan A. Modelling SF-6D health state preference data using a nonparametric Bayesian method Journal of Health Economics. Journal of Health Economics 2007; 26:597-612
  • Kharroubi S, O'Hagan A, Brazier J. Estimating utilities from individual health preference data: a nonparametric Bayesian method. Applied Statistics 2005; 54:879-895

Frequently asked questions

What is available from this website?

Programmes that convert your SF-36 or SF-12 data into an SF-6D health state and corresponding utility score for each observation in your dataset.

What do the programmes do?

The programme will generate for each row of your dataset the six dimension scores of the SF-6D, the six-digit health state and a utility value anchored at 1 for full health and 0 for dead.

What is the SF-6D?

The SF-6D is a generic preference-based single index measure of health that can be used to generate QALYs and hence which can be used in cost-utility analysis.

What types of programmes are available?

There are Excel, SPSS and SAS programmes available to convert SF-36 data into SF-6D data and there are SPSS and SAS programmes available to convert SF-12 data into SF-6D.

All programmes generate the SF-6D utility score estimated using a set of parametric preference weights obtained from a sample of the general population using the recognised valuation technique of standard gamble. In addition, a new excel programme is now available from the University of Sheffield to convert SF-36 data into the SF-6D utility score estimated using a set of non-parametric Bayesian preference weights. These nonparametric preference weights are an improvement on the parametric preference weights as the nonparametric model has many advantages over the conventional parametric random effects model which is reflected in improvements in the predictive ability of the model. For further details see Kharroubi et al. (2007).

Furthermore, a new excel programme is available to convert SF-36 data into the SF-6D utility score estimated using a set of preference weights obtained using an ordinal valuation technique for a sample of the general population. The estimates using ordinal data represent an alternative value set based on a different valuation technique which produces estimates that are comparable to estimates produced using standard gamble data. For further details see McCabe et al. (2006).

What do I need to include in my study to use these programmes?

The programmes which are available from our website derive the SF-6D from either the SF-12 or the SF-36 and therefore to use these algorithms you need to include the SF-12 or SF-36 in your study. Refer to Optum Insight for details on these. These consist of 12 or 36 questions about the individual´s health where the individual chooses an option from a list of responses.

Do you provide a licence for the SF-36 and SF-12?

Refer to Optum Insight for details on these.

Do I need to include a valuation in my study?

No. You need to include either the SF-12 or the SF-36 to use these programmes. These do not involve a valuation study or ranking. The programmes which are available from our website use already existing utility values from the valuation study described in Brazier and Roberts (2004).

Is it possible to use the SF-6D alone in a study, rather than using the SF-36 or SF-12 and converting this into the SF-6D?

It is theoretically possible but not generally recommended. The data will be of limited use if only the SF-6D is included in a study as there is no conversion from the SF-6D to the SF-36 or SF-12. Furthermore, the algorithms are based on the SF-6D derived from the SF-36 and we have no evidence whether this equals a directly administered SF-6D.

Can I use the SF-6D algorithm for a different study other than the study described in the initial application?

No. You need a licence agreement for each study for which you use the SF-6D algorithm.

How does the SF-6D generate preference scores?

The SF-6D comes with a set of preference weights obtained from a sample of the general population in the UK using the recognised valuation technique of standard gamble. Members of the general population in the UK were asked to value a selection of health states from which a model has been estimated to predict all the health states described by the SF-6D.

Is there a written document outlining the process for converting the SF-36 data to SF-6D scores?

Instructions are available for each of the programmes available but these do not outline how the process can be done using a different software package. We currently have algorithms available for use on SF-36 data in SAS, SPSS and excel, and for SF-12 data in SAS and SPSS.

Can I use these programmes for a study outside the UK?

The SF-6D utility score is generated using preference weights obtained from a sample of the general population in the UK. The UK population may have different preferences to non-UK populations.

Are there SF-6D preference scores available for other countries?

There is emerging evidence that standard gamble health state values differ between countries for SG, so it is often important to have local valuations. SF-6D preference scores are available for Australia, Brazil, Hong Kong, Japan, Portugal and Singapore. See below for details of who to contact for further information.

References

  • Brazier, JE, Roberts, JR,. The estimation of a preference-based index from the SF-12. Medical Care, 2004;42(9):851-859.
  • Kharroubi, S, Brazier, JE, Roberts, JR, et al. Modelling SF-6D health state preference data using a nonparametric Bayesian method. Journal of Health Economics, 2007;26(3):597-612.
  • McCabe, C, Brazier, JE, Gilks, P, et al. Using rank data to estimate health state utility models. Journal of Health Economics, 2006;25(3):418-431.

Published Applications of the SF-6D

Recovering Quality of Life (ReQoL)

Obtain a license

The development of the ReQoL was led by researchers at School of Health and Related Research (ScHARR), funded by the Department of Health. ReQoL is a condition-specific health-related quality of life instrument developed for use with   populations experiencing mental health difficulties aged 16 and over.  The measure contains seven themes (activity; belonging and relationships; choice, control and autonomy; hope; self-perception; well-being and; physical health). There are two versions of the ReQoL measures. ReQoL-10 contains 10 mental health items and ReQoL-20 contains ten items from ReQoL-10 plus ten extra items. Both versions contain one physical health question. They are suitable for use across all mental health populations including common mental health problems, severe, complex, and psychotic disorders.

The ReQoL measures are available for licence from the Clinical Outcomes team at Oxford University Innovation. The ReQoL has grown rapidly in its recognition as a quality of life measure for assessment with people experiencing mental health difficulties with over 200 licences issued. The international recognition of the ReQoL is demonstrated by the rapid growth of the library of translations, with currently 18 available and several in progress. The ReQoL-20 has been recommended as the new standard set of measures for Psychotic Disorders by the International Consortium for Health Outcome Measurement (ICHOM) for tracking three aspects of psychotic disorders - Quality of Life, Personal recovery, and Positive and negative symptoms.

The Recovering Quality of Life-Utility Index (ReQoL-UI) is the preference-based measure that has been constructed from a subset of the items of the items from ReQoL-10/ReQoL-20. Preference weights have been estimated from a sample of the general population using time-trade-off.

Papers

Main citation: Keetharuth, D., Brazier, J., Connell, J., Bjorner, J., Carlton, J., Taylor Buck, E., et al. (2018). Recovering Quality of Life (ReQoL): a new generic self-reported outcome measure for use with people experiencing mental health difficulties. The British Journal of Psychiatry: the journal of mental science, 212(1), 42-49. doi:10.1192/bjp.2017.10

Taylor Buck E, Smith C, Lane A, Keetharuth A, Young T, Cooke J. (2020). Use of a modified Word Café process to discuss and set priorities for a Community of Practice supporting implementation of ReQoL, a new mental health and quality of life patient reported outcome measure (PROM). (short report). Journal of Patient Reported Outcomes. 4, 38 (2020). https://doi.org/10.1186/s41687-020-00202-z

Grundy, A., Keetharuth, D., Barber, R., Carlton, J., Connell, J., Taylor Buck, E, Brazier, J. (2019). Public Involvement in health outcomes research: lessons learnt from the development of the Recovering Quality of Life (ReQoL) measures. Health and Quality of Life Outcomes. https://doi.org/10.1186/s12955-019-1123-z

Keetharuth, D., Bjorner, J. B., Barkham, M., Browne, J., Croudace, T., & Brazier, J. (2019). Exploring the item sets of the Recovering Quality of Life (ReQoL) measures using factor analysis. Quality of Life Research. doi:10.1007/s11136-018-2091-1

Keetharuth, D., Taylor Buck, E., Acquadro, C., Conway, K., Connell, J., Barkham, M., . . . Brazier, J. E. (2018). Integrating Qualitative and Quantitative Data in the Development of Outcome Measures: The Case of the Recovering Quality of Life (ReQoL) Measures in Mental Health Populations. International Journal of Environmental Research and Public Health, 15(7), 1342.  doi:10.3390/ijerph15071342

Connell, J., Carlton, J., Grundy, A., Taylor Buck, E., Keetharuth, A. D., Ricketts, T., . . . Brazier, J. (2018). The importance of content and face validity in instrument development: lessons learnt from service users when developing the Recovering Quality of Life measure (ReQoL). Quality of Life Research Vol. 27 (pp. 1893-1902). doi:10.1007/s11136-018-1847-y

Algorithm for preference weights

The manuscript for ReQoL-UI is under review. An algorithm to calculate utilities from the ReQoL measures will be available shortly and details will appear here.

Further information

If you would like further information about this work, please contact Anju Keetharuth on reqol@sheffield.ac.uk

Download Sample copies 

ReQoL-10

ReQoL-20

See also

ReQoL website

Oxford University Innovation (for licences)

EMHeP Health Professionals Survey

Background

Mental illness has a significant impact on individuals, society and the economy.

This research will analyse the efficiency, cost and quality of mental healthcare provision. The project involves researchers from the University of York and University of Birmingham as well as researchers from the University of Sheffield.

The project undertaken at the University of Sheffield assessed which aspects of mental healthcare quality were valued most by mental health service users, mental healthcare professionals and the general public.
We determined our aspects of mental healthcare quality using policy documents, focus groups with mental health service users and frontline mental healthcare professionals, data available across different NHS Trusts, and feedback from service users and carers to ensure appropriate wording.

What we did

We conducted a survey online using a technique called a discrete choice experiment, where participants were asked to choose which 1 of 2 different mental healthcare providers they thought was best where the two providers differed in aspects of quality. Statistical analyses were used to determine which aspects of quality were most important, by weighting them using a single measure of health outcome (called QALYs, quality adjusted life years).

What we found

We found that the most important aspect of quality was being treated with dignity and respect. Also important was whether the person or people you see organise the care and services you need well.

For mental healthcare service users and mental healthcare providers it was also important to be listened to carefully and given enough time to discuss your needs and treatment.

For members of the general population it was important whether the healthcare you receive is provided in your local area.

We found what was least important was waiting time, which we phrased as whether the time you wait to receive healthcare is appropriate for your needs.

What’s next?

Our analysis is still ongoing and is not yet published or publically available. When it is available we will post a link on this page. In the meantime, if you have any questions please contact emhep@sheffield.ac.uk.

EMHeP Poster for Health Professionals

EMHeP Service User Survey

Background

Mental illness has a significant impact on individuals, society and the economy.

This research will analyse the efficiency, cost and quality of mental healthcare provision. The project involves researchers from the University of York and University of Birmingham as well as researchers from the University of Sheffield.

The project undertaken at the University of Sheffield assessed which aspects of mental healthcare quality were valued most by mental health service users, mental healthcare professionals and the general public.
We determined our aspects of mental healthcare quality using policy documents, focus groups with mental health service users and frontline mental healthcare professionals, data available across different NHS Trusts, and feedback from service users and carers to ensure appropriate wording.

What we did

We conducted a survey online using a technique called a discrete choice experiment, where participants were asked to choose which 1 of 2 different mental healthcare providers they thought was best where the two providers differed in aspects of quality. Statistical analyses were used to determine which aspects of quality were most important, by weighting them using a single measure of health outcome (called QALYs, quality adjusted life years).

What we found

We found that the most important aspect of quality was being treated with dignity and respect. Also important was whether the person or people you see organise the care and services you need well.

For mental healthcare service users and mental healthcare providers it was also important to be listened to carefully and given enough time to discuss your needs and treatment.

For members of the general population it was important whether the healthcare you receive is provided in your local area.

We found what was least important was waiting time, which we phrased as whether the time you wait to receive healthcare is appropriate for your needs.

What’s next?

Our analysis is still ongoing and is not yet published or publically available. When it is available we will post a link on this page. In the meantime, if you have any questions please contact emhep@sheffield.ac.uk.

EMHeP Poster for Service Users

Related information

Centres of excellence

The University's cross-faculty research centres harness our interdisciplinary expertise to solve the world's most pressing challenges.

Centres of excellence