Guidelines on Methods for Measurement of Health Utilities

Going Global in Pharma: Understanding Preferred Health Technology Assessment (HTA) Utility Measures by Country

Health Utility Measurement

Are you up to date with HTA recommendations on health utility measurement for your cost utility model? Many HTA agencies make specific statements about their preferences for health utility measures, but their guidelines change over time. And it can be hard to find the right document. 

In our work, we must stay up-to-date on Health Technology Assessment authority guidelines for health state utility measurement methods across multiple countries. We've developed a summary table of these guidelines that we trust you'll find helpful.

Country Agency Guidance Source
Australia PBAC

No specific utility instrument is favored.
The generally preferred method of measuring QALYs is to use quality-of-life or utility data. Australian-based preference weights are preferred for use in the scoring algorithm to calculate the utility weights. When these weights are not available, outcomes may be valued using preferences that reflect the general population, with justification for doing so. Alternatively, scenario-based utility weights could also be used, along with utility weights published in the literature.

Pharmaceutical Benefits Advisory Committee (2016)
Brazil DECIT-CGATS SG, TTO, EQ-5D or SF-6D. Ministério da Saúde (2014)
Canada CADTH Health preferences should reflect the general Canadian population and should be obtained from an indirect method of measurement based on a generic classification system (e.g., EQ-5D, HUI, SF 6D). When an indirect method is not used, researchers must justify that decision. Selection of data sources should be based on their fitness for purpose, credibility, and consistency. Canadian Agency for Drugs and Technologies in Health (2017)
China No policy-generating agency Three direct valuation methods, such as SG, TTO, and VAS, are recommended.
Four indirect valuation methods, such as EQ-5D, SF-6D, HUI, and QWB, are recommended.
The utility weights should be estimated based on a value set for the Chinese population. If a Chinese value set is not available, the recommendation is to use a value set that is widely used and accepted (e.g., UK tariff) or a value set for a similar population (e.g., South Korean/Japanese tariff). 
Liu et al. (2015)
France HAS EQ-5D or HUI3
Use of a health status classification system for which validated preference-based scores are available in France is recommended. In the absence of French data, it is possible to use preference-based scores from foreign studies, subject to a critical analysis of their quality.
The use of any other method must be duly qualified, and its validity in France must be demonstrated.
Haute Autorité de Santé (2012)
Germany IQWiG/G-BA Note that no cost-utility analysis is required for G-BA submission.
Data should be collected in a clinical setting. No particular utility measure is preferred. For generic index instruments, a validated German version must be used to determine the utility value. A German tariff should be used. Justification for use of the generic instrument should be stated, and the measure should be comprehensible. Evidence of the instrument’s objectivity, reliability, validity, and responsiveness is required. In addition, disease-specific instruments should be included in clinical studies to determine quality of life. The mapping of disease-specific instruments to generic instruments is discouraged.
Institute for Quality and Efficiency in Health Care (2017)
Italy l’Ufficio Coordinamento OsMed e attività HTA EQ-5D
Use of the EQ-5D to measure the utility weights is recommended whenever possible. The utility weights should be estimated based on the value set for the Italian population.
If EQ-5D utility weights are not available or not suitable for the specific patient population, utility weights based on other instruments can be used. 
Associazione Italiana di Economia Sanitaria (2009)
Mexico CENETEC-SALUD  The EQ-5D is preferred. Centro Nacional de Excelencia Tecnológica en Salud (2017)
Netherlands ZiN The EQ-5D-5L should be used and valued using Dutch reference values.
Additional questionnaires and methods can be used to measure quality of life to identify the effect to its fullest; however, the EQ-5D-5L should still be administered.
Zorginstituut Nederland (2016)
Norway NoMA EQ-5D with UK tariff is recommended.
If the EQ-5D-5L questionnaire is used, utility weights should be calculated using the EQ-5D-3L crosswalk algorithm described by van Hout et al. (2012) and the English tariff published by Devlin et al. (2017) until further notice.
Use of the EQ-5D can be waived if no data from the EQ-5D for the disease in question are available, or if the EQ-5D has been judged to be not suitable for capturing relevant aspects of the patient population in question. For guidance on deciding whether or not the EQ-5D is suitable for the patient population in question, refer to NICE DSU Technical Support Document 8 (Brazier and Longworth, 2011).
Where patient-reported EQ-5D data are lacking, other generic preference-based instruments can be used (SF-6D, 15-D, HUI, AQoL, QWB). The preference-based values from alternative instruments must then be mapped to EQ-5D values, in accordance with validated methods. If no data are available from generic instruments, but data are available from disease-specific instruments, these must be mapped to predict EQ-5D values. Refer to NICE DSU Technical Support Document 10 (Longworth and Rowen, 2011) for methods for mapping quality-of-life data to the EQ-5D.
Norwegian Medicines Agency (2018)
South Korea HIRA

Direct measures (SG, TTO, Rating scale) or indirect measures (MAUI, EQ-5D, HUI 2/3, SF-6D, QWB) may be used.
Use of the Korean EQ-5D tariff is recommended. 

Health Insurance Review and Assessment Service (2011)
Spain Spanish HTA Network Indirect methods (Spanish recommendations and CATSALUT). Direct or indirect methods (OSTEBA).
European Network for Health Technology Assessment (2018)
Sweden TLV Calculations based on appraisals of persons with the health condition of interest are preferred over weights calculated from an average of a population estimating a condition depicted for the population, i.e., the social tariff from EQ-5D. Tandvårds-och läkemedelsförmånsverket (2003)
USA Second Panel on Cost-Effectiveness in Health and Medicine To enhance the comparability of results across studies, the use of a generic preference-based measure is recommended. The measure should be fit for purpose, with measurement properties that allow measurement of differences and changes in health across interventions. No specific, generic, preference-based measure is recommended. Sanders et al. (2016)
  ICER No specific measure of health utilities is recommended. For individuals with chronic, severe conditions, scenario analyses should be performed to examine the influence of lower utilities on cost-per-QALY findings. Institute for Clinical and Economic Review (2017)
UK NICE For adults, the EQ-5D-3L, using the UK general population TTO value set is recommended. The EQ-5D-5L descriptive system may be used; however, currently the 5L valuation set is not recommended for use, reference case analyses should calculate utility values by mapping the 5L descriptive system data onto the 3L valuation using the mapping function by van Hout et al. (2012). If analyses use data gathered using both the EQ-5D-3L and EQ-5D-5L descriptive systems, the 3L valuation set should be used to derive all utility values, with 5L mapped onto 3L where needed. When EQ-5D data are not available, these data can be estimated by mapping other health-related quality-of-life measures or health-related benefits observed in the relevant clinical trial(s) to the EQ 5D.
For children, an alternative health-related quality-of-life measure can be used. A version of the EQ-5D is available for children aged 7-12 years; however, no validated UK valuation set is available.
In some circumstances, the EQ-5D may not be the most appropriate measure. To make a case that the EQ-5D is inappropriate, qualitative empirical evidence on the lack of content validity for the EQ-5D should be provided, demonstrating that key dimensions of health are missing. This should be supported by evidence showing that the EQ-5D performs poorly on tests of construct validity and responsiveness in a particular patient population. This evidence should be derived from a synthesis of peer-reviewed literature.
National Institute for Health and Care Excellence (2013, 2018)
  SMC The EQ-5D is preferred in randomized controlled studies of medicines. The EQ-5D may not be appropriate in all circumstances; those submitting data should provide reasons for their choice of instrument.
Preference for health states to be measured in patients using a generic and validated classification system for which reliable and appropriate population preference values, elicited using a choice-based method such as the time trade-off or standard gamble (but not rating scale), are available. Utilities derived from observational studies of patients would be acceptable as long as it can be shown that the patients and health states adequately match those in the clinical studies used in the submitted economic evaluation.
Alternative, well-designed methods of utility measurement are allowed if generic utility data are not available (mapped from disease-specific quality-of-life measure; direct measurement in patients or vignettes valued using TTO or SG; or published data). The use of a non-QALY outcome measure is allowed if the measure is shown to be appropriate and the value of the medicine to National Health Service in Scotland can be demonstrated.
Scottish Medicines Consortium (2018)

If you need more, our integrated team of experts can help you with your utility measurement and HTA submission needs:
• Utility research design (trials, observational studies, surveys)
• Instrument/method selection
• Study implementation
• Utility data analysis
• Data incorporation into cost utility models