Valuing the Health of Children: Understanding What Drives Differences in Health Preferences for Children Versus Adults

By Christiane Truelove

Ask the average person on the street about what is the value of a child’s life and the automatic reply might be, “Children are priceless.” But health economists and policy makers can’t think this way when the health system they administer has a finite amount of money and they need to provide universal care. With the advancement of expensive therapies for treating rare diseases in children—such as Spinraza (nusinersen) for spinal muscular atrophy, Lumizyme (alglucosidase alfa) for Pompe disease, and the gene therapy Zolgensma (onasemnogene abeparvovec-xioi), also for spinal muscular atrophy—there is a pressing need to figure out how to appropriately measure the quality of life in children and the true value of these therapies.

Through the Eyes of a Child? Not Quite.
According to Reckers-Droog, although the quality-of-life stages in the adult and the child versions of the EQ-5D instrument are similar, research shows these utilities differ between adult and child health states. Specifically, the first value sets that were developed in the child’s version of EQ-5D instrument showed that members of the public attribute a higher value to the health states for children than for adults with similar health states, and “there’s more noise in the data,” Reckers-Droog says. The study’s goal was to explain what drives these differences.

Researchers found that those who participated in the study were quite able to value health states for themselves, “but they really, really wanted to avoid thinking about children in bad health states with a short life expectancy,” Reckers-Droog said. “They said, ‘I just don’t want to think about this. Please get through these questions. I want to move on. This is so uncomfortable.’” This means some of the differences noted may derive from people randomly clicking on answers to get through the questions as quickly as possible.

Researchers also noted that the participants found it difficult to “give up” life years for children. If given a choice between a child living a shorter number of years but in perfect health, to a child living longer but with challenges, many opted for the longer life. “People just really didn’t want to give up any life years for improving quality of life in children,” Reckers-Droog said. “They said it’s important that the child completes primary school or secondary school. Or, ‘I feel like I’m a bad parent, but I want my child to stay with me. I know they’re in pain, but I promise I will take good care of them.’”

"The public attributes a higher value to the health states for children than for adults with similar health states.Researchers found that those who participated in the study were quite able to value health states for themselves, 'but they really, really wanted to avoid thinking about children in bad health states with a short life expectancy.'" —Vivian Reckers-Droog, PhD

This means in cost-effectiveness analysis, when estimating the incremental gains that children could gain from getting access to a new treatment in comparison to the standard treatment, that if all of the values of the health states are relatively similar and high, there is not much to gain for them, Reckers-Droog points out. Technologies that may be cost-effective for adults may be less so for children because of these high and yet small differences between health states. For children, the adults will overvalue small gains in quality of life and life expectancy.

Another thing the researchers noticed is that the adults tended to imagine different children when answering the questions. “Sometimes within the same question, they think about their own child or grandchild, the neighbor, or someone who was in their class,” Reckers-Droog said. A teacher would think about all the different children they have had in a class. And some of the participants would come up with hypothetical children, theorizing that while some would like to play outside, others would prefer to draw and don’t mind not being able to walk. This is quite different than the adult EQ-5D studies, where researchers can make sure that the study sample is relatively representative of the larger population.

There were also valuations that mattered more for adults than for children, Reckers-Droog said. When valuing the importance of self-care for an adult versus self-care for a child, respondents stated that it was more important for an adult be able to take care of themselves. “Study participants didn’t mind it so much if the children could not wash or dress themselves, because they had parents to take care of this.”

Researchers also found that not all children were equally considered when it came to mental health. Most notably, when asked about depression and anxiety in a 10-year-old versus a 15-year-old, participants expressed that “mental health was not considered to be such a bad problem [for a 15-year old], because they thought that some mental health issues are just part of puberty,” Reckers-Droog said. “An unhappy 10-year-old child was much more problematic for study participants than an unhappy 15-year-old.”

Perfecting the Perspective
According to Reckers-Droog, the study results showed that the current valuation methods for children’s health need to be improved. “In order to compare the cost-effectiveness, or the health gains from different health technologies in different populations, you need to be better, or you need to be more certain that you’re comparing the same thing,” she said.

The next iteration of this type of study may be trying to involve children in some way, but this will be difficult to do, said Koonal Shah, PhD, Science Policy and Research Programme, National Institute for Health and Care Excellence, London, England, United Kingdom. Shah is one of the authors of another paper in the July issue of Value in Health, “Exploration of the Reasons Why Health State Valuation Differs for Children Compared With Adults: A Mixed Methods Approach.” Like Reckers-Droog’s group, Shah and colleagues conducted interviews with adults and had them value health states from adult and child perspectives. These researchers found that fewer life years are traded against a higher quality of life for health states referring to children than for adults by adult respondents, resulting in higher utility values for children using standard time trade-off methodologies. “It was found that the quality-adjusted life year has a different interpretation for children compared with adults; therefore, society’s willingness to pay for additional childhood life years may also be different and a youth-specific cost-effectiveness threshold may be needed to fund access to healthcare for children,” these researchers said.

"The quality-adjusted life year has a different interpretation for children compared with adults; therefore, society’s willingness to pay for additional childhood life years may also be different."

However, as “cognitively difficult” as it is for adults to answer these questions, there are real ethical considerations about “asking children to consider hypothetical scenarios that relate to death and dying and severe ill health,” Shah says. “And you just may not get valid or reliable responses.” He does say it is important to understand younger people’s preferences, and “you can use alternative methods that are less cognitively challenging and don’t involve explicit consideration of death.”

A discrete choice experiment, for example, can be applied without a death or a health state duration attribute. Shah said he was involved in research where a discrete choice experiment that had been used for adults was tweaked and used for 11- to 17-year-olds. “We did find that we got reasonable responses that suggested that adolescents were capable of expressing their preferences about health states in that way.”

While the adolescents’ responses were not hugely different from the adults, there were some differences, Shah said, and these justified looking at the adolescent preferences.

Implications for Health Policy Makers
Measuring and valuing health-related quality of life for children is difficult, Shah admitted. At NICE, “we’ve got very well-established methods for measuring and valuing health-related quality of life in adults, and we’ve struggled to make recommendations about how to measure and value health in children and younger people because there are these methodological issues.”

On one hand, while it can be appropriate to base child health state values on the preferences of children because it is children who are affected by the health states, NICE and many other similar HTA bodies take a view that the preferences of the general population should be sought when generating utility values, and not necessarily those who are currently affected by a specific health state. “Many adult members of the public are taxpayers, and in a taxpayer-funded healthcare system, like we have in in the United Kingdom, it is appropriate that their views count towards how technologies are evaluated and funded,” Shah said. There is also the argument that everybody, not just current patients, are potential users of healthcare in the future.

"As “cognitively difficult” as it is for adults to answer these questions, there are real ethical considerations about asking children to consider hypothetical scenarios that relate to death and dying and severe ill health."

Another argument for eliciting children’s health preferences is that even though adults could be future patients, they can never be the users of a child’s health intervention. Shah points out that adults, however, make decisions on behalf of children all the time, including healthcare decisions for their own kids.

“The challenge for decision makers and policy makers like NICE is to find an appropriate balance between those arguments,” Shah said. “And we could do with research from ethicists and from pediatric specialists to help guide us through the various issues.”

And for the future? “While we’re already seeing lots of good empirical research done on the impact of different perspectives, and different preference elicitation methods, I’d like to see a focus on the normative arguments and practical considerations,” Shah said. “I think that would be really helpful for informing the policies of organizations like NICE going forward.”

Editor’s Note: In addition to the 2 articles mentioned in this story, the July 2022 issue of Value in Health contained a commentary by Nancy J. Devlin, PhD entitled, “Valuing Child Health Isn’t Child’s Play” and an article by Juan M. Ramos-Goñi et al entitled, “Does Changing the Age of a Child to Be Considered in 3-Level Version of EQ-5D-Y Discrete Choice Experiment-Based Valuation Studies Affect Health Preferences?” Follow this link to read this special collection of articles on valuing health of children in Value in Health.