Value Sets: Breathing New Life into Old Data Using Choice Modeling - SurveyEngine GmbH

Value Sets: Breathing New Life into Old Data Using Choice Modeling

June 28, 2024

Are patients willing to sacrifice more pain for improved mobility?

A definitive answer to this question could have powerful policy implications, but until recently, an answer has been elusive. Even if you don’t work in health, read on to discover how preference models allow researchers to compare multidimensional concepts and help break the deadlock of poor policy decisions.

Data on patient outcomes has been collected for over a century using standardized questions. The term Patient Reported Outcome Measure (PROM) emerged in the 1960s, and around 300 PROMs covering both generic and condition‐specific measures are in use today. One example is the EORTC’s QLQ-C30, which assesses symptoms and functioning levels that make up quality of life for a cancer patient. Another is the EQ-5D from EuroQoL, designed to measure health-related quality of life for any kind of patient. These measures are relatively simple to deploy, and they are a useful tool for tracking improvement or decline in a particular dimension of health (such as fatigue).

But it isn’t clear how we should think about a patient improving in nausea but worsening in anxiety/depression, for example. In other words, we need a way to make holistic comparisons of multidimensional concepts.

Choice Modeling: The Power to Maximize Positive Outcomes

So how do people trade off between dimensions of health? Historically, researchers would use unvalidated algorithms such as comparing sums or averages of all dimensions. But pioneering investigators have moved the field beyond these weak methods to ones supported by economic theory. Now, we can use stated preference methods (like DCEs) to calculate meaningful average values for each holistic health state. And to reduce patient burden, we can recruit one large group to complete the choice exercise and use their average valuations for the entire population.

Specifically, we can calculate the average value or ‘utility’ people place on every possible health state of an instrument (e.g., scoring “1, 3, 2, 1, and 1” on the 5 dimensions of the EQ-5D). It’s typical to measure states’ utility relative to full health (valued at 1) and being dead (valued at 0). Values below 0 are possible for states considered worse than being dead. The set of all pairs of health states and their corresponding values is called a value set.

Once you have a value set, you can apply the values to survival data to measure the impact of an intervention on Quality-Adjusted Life Years (QALYs). With this standardized value, we can then compare across treatments and even across diseases. A healthcare decision maker faced with limited resources can then allocate those resources to achieve the most QALYs, i.e., the best population-level quality of life outcomes.

Country-Specific Value Sets: Allowing for Cultural Variation

Rather than assume that every culture in the world shares the same preferences, researchers are collecting value sets in different countries. These country-specific value sets allow decision makers to incorporate the values of their local population.

One example we can dive into is the QLU-C10D, a cancer-specific preference elicitation measure developed by the Multi-Attribute Utility in Cancer (MAUCa) Consortium (chaired by Dr. Madeleine King, Professor Emeritus at The University of Sydney) and the EORTC QOL Group. SurveyEngine managed the implementation of several valuation studies for this measure, resulting in over a dozen country-specific value sets. So what do we see?

The chart here (courtesy of Prof. King) is taken from a nine-country study and shows the relative importance of health domains for cancer patients. First, to answer our opening question: universally, across all nine countries, patients are willing to trade off increased pain for higher mobility. Beyond that, the rank order importance of some domains is similar across countries: mobility, pain, limited work/daily activities, and nausea. However there are a few notable exceptions. For example, Poland is alone in its strong emphasis on limited work/daily activities above even pain.

Other Special Populations

There is a new frontier of QOL measurement: children. Child-specific PROMs have been a new development in the last ten or so years, according to Dr. Elly Stolk, Professor of Measurement and Valuation of Health at Erasmus University. And moving from childrens’ PROMs to childrens’ value sets is even newer, such as Prof. Stolk’s work on youth versions of the EQ-5D. There are ongoing discussions in this area surrounding normative questions: should children value health, or should adults do it for them? If the answer is adults, how should they be asked to imagine a child’s health state? These decisions matter for the outcome.

The Impact of Outcome Valuation

According to Prof. Stolk, countries that incorporate health economic evaluations and QALYs in their policy decisions often achieve better health outcomes. The possibilities of future impact are endless, and not just within health. For example, we may have decades of data on a city’s transportation options’ comfort, speed, noise, environmental impact, etc. But how do commuters trade off between these dimensions? Combining the real-world data with a value set can allow policy makers to maximize commuters’ satisfaction.

Thinking about how value sets could benefit your field? Let’s talk!

References

Norman, R., Viney, R., Aaronson, N. K., Brazier, J. E., Cella, D., Costa, D. S., Fayers, P. M., Kemmler, G., Peacock, S., Pickard, A. S., Rowen, D., Street, D. J., Velikova, G., Young, T. A., & King, M. T. (2016). Using a discrete choice experiment to value the QLU-C10D: feasibility and sensitivity to presentation format. Quality of life research, 25(3), 637–649. https://doi.org/10.1007/s11136-015-1115-3

Gamper, E. M., Holzner, B., King, M. T., Norman, R., Viney, R., Nerich, V., & Kemmler, G. (2018). Test-Retest Reliability of Discrete Choice Experiment for Valuations of QLU-C10D Health States. Value in health, 21(8), 958–966. https://doi.org/10.1016/j.jval.2017.11.012

Gamper, E., Holzner, B., King, M. T., Norman, R., & Kemmler, G. (2016). Trading-Off Quality of Life and Survival Time–Feasibility of Web-Based Discrete Choice Experiments For QLU-C10D Utility Elicitation in Cancer Patients. Value in Health, 19(7), A746. https://doi.org/10.1016/j.jval.2016.09.2284

King, M. T., Viney, R., Simon Pickard, A., Rowen, D., Aaronson, N. K., Brazier, J. E., Cella, D., Costa, D. S. J., Fayers, P. M., Kemmler, G., McTaggart-Cowen, H., Mercieca-Bebber, R., Peacock, S., Street, D. J., Young, T. A., Norman, R., & MAUCa Consortium (2018). Australian Utility Weights for the EORTC QLU-C10D, a Multi-Attribute Utility Instrument Derived from the Cancer-Specific Quality of Life Questionnaire, EORTC QLQ-C30. PharmacoEconomics, 36(2), 225–238. https://doi.org/10.1007/s40273-017-0582-5

Norman, R., Mercieca-Bebber, R., Rowen, D., Brazier, J. E., Cella, D., Pickard, A. S., Street, D. J., Viney, R., Revicki, D., King, M. T., & European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Group and the MAUCa Consortium (2019). U.K. utility weights for the EORTC QLU-C10D. Health economics, 28(12), 1385–1401. https://doi.org/10.1002/hec.3950

McTaggart-Cowan, H., King, M. T., Norman, R., Costa, D. S. J., Pickard, A. S., Regier, D. A., Viney, R., & Peacock, S. J. (2019). The EORTC QLU-C10D: The Canadian Valuation Study and Algorithm to Derive Cancer-Specific Utilities From the EORTC QLQ-C30. MDM policy & practice, 4(1), 2381468319842532. https://doi.org/10.1177/2381468319842532

Kemmler, G., Gamper, E., Nerich, V., Norman, R., Viney, R., Holzner, B., King, M., & European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Group (2019). German value sets for the EORTC QLU-C10D, a cancer-specific utility instrument based on the EORTC QLQ-C30. Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation, 28(12), 3197–3211. https://doi.org/10.1007/s11136-019-02283-w

Nerich, V., Gamper, E. M., Norman, R., King, M., Holzner, B., Viney, R., & Kemmler, G. (2021). French Value-Set of the QLU-C10D, a Cancer-Specific Utility Measure Derived from the QLQ-C30. Applied health economics and health policy, 19(2), 191–202. https://doi.org/10.1007/s40258-020-00598-1

Gamper, E. M., King, M. T., Norman, R., Efficace, F., Cottone, F., Holzner, B., Kemmler, G., & European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Group (2020). EORTC QLU-C10D value sets for Austria, Italy, and Poland. Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation, 29(9), 2485–2495. https://doi.org/10.1007/s11136-020-02536-z

Jansen, F., Verdonck-de Leeuw, I. M., Gamper, E., Norman, R., Holzner, B., King, M., Kemmler, G., & European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Group (2021). Dutch utility weights for the EORTC cancer-specific utility instrument: the Dutch EORTC QLU-C10D. Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation, 30(7), 2009–2019. https://doi.org/10.1007/s11136-021-02767-8

Xu, R. H., Wong, E. L., Luo, N., Norman, R., Lehmann, J., Holzner, B., King, M. T., Kemmler, G., & EORTC QLG (2023). The EORTC QLU-C10D: the Hong Kong valuation study. The European journal of health economics : HEPAC : health economics in prevention and care, 10.1007/s10198-023-01632-4. Advance online publication. https://doi.org/10.1007/s10198-023-01632-4

Revicki, D. A., King, M. T., Viney, R., Pickard, A. S., Mercieca-Bebber, R., Shaw, J. W., Müller, F., & Norman, R. (2021). United States Utility Algorithm for the EORTC QLU-C10D, a Multiattribute Utility Instrument Based on a Cancer-Specific Quality-of-Life Instrument. Medical decision making : an international journal of the Society for Medical Decision Making, 41(4), 485–501. https://doi.org/10.1177/0272989X211003569

Finch, A. P., Gamper, E., Norman, R., Viney, R., Holzner, B., King, M., Kemmler, G., & EORTC Quality of Life Group (2021). Estimation of an EORTC QLU-C10 Value Set for Spain Using a Discrete Choice Experiment. PharmacoEconomics, 39(9), 1085–1098. https://doi.org/10.1007/s40273-021-01058-x

Lehmann, J., Rojas-Concha, L., Petersen, M. A., Holzner, B., Norman, R., King, M. T., Kemmler, G., & EORTC Quality of Life Group (2024). Danish value sets for the EORTC QLU-C10D utility instrument. Quality of life research, 33(3), 831–841. https://doi.org/10.1007/s11136-023-03569-w

About the Author

Avatar photo

carol

Carol Iskiwitch is a behavioral market research consultant at SurveyEngine. She has designed and conducted survey and experimental research for over 10 years.

Scroll to Top