Reconceptualizing Treatment Value: A Q&A with Professor Lou Garrison on Augmented Cost-Effectiveness Analyses

Professor Lou Garrison – Professor Emeritus of The Comparative Health Outcomes, Policy, and Economics (CHOICE) Institute at the School of Pharmacy at the University of Washington – has deep health care policy research experience, including on value frameworks. Analysis Group Managing Principal Noam Kirson sat down with Professor Garrison to discuss why stakeholders may want to consider non-traditional elements of value in their assessments of medical treatments, and how economists could help with such assessments. 

As the co-chair of ISPOR’s Special Task Force on US Value Assessment Frameworks, you suggested integrating novel elements of value into more traditional cost-effectiveness analyses of health technologies.¹ Why was this work needed?

From an economic perspective, the value of a good or service is what a consumer would be willing to pay for it. This interacts with supply in the market to determine the market price for a given good or service, which, in turn, can serve as a signal to potential suppliers. It’s very important to provide appropriate incentives to those in the health care industry who are developing and supplying new treatments and drugs so that society as a whole continues to benefit from scientific progress and innovation. Of course, rewards need to be commensurate with the value that we receive. We want to guide developers to focus on the health problems that matter most and are amenable to improvement. But that economic value is affected by a multitude of different elements, many of which are challenging to quantify.

Because innovative pharmaceuticals are purchased in large share by health insurers acting on behalf of all health plan members, health economists have developed methods to assess an intervention’s value based on measuring the incremental cost of gaining an additional QALY [quality-adjusted life year], or similar measures, such as disability-adjusted life years [DALYs]. This traditional cost-effectiveness analysis (or CEA) is, however, often limited in its ability to assess value from the perspective of the entire health care ecosystem and our entire society. By focusing on an individual patient’s heath gain and health care costs, it neglects several alternative aspects of benefits and costs that should also be considered.

Could you provide an example of an alternative consideration that might be included in an expanded value assessment framework such as the ISPOR “value flower” that you helped develop?

Traditional CEAs do not directly account for the impact of uncertainty or risk on people’s well-being. Most human beings are averse to risk – we naturally prefer certainty. But there are very few sure things when it comes to treating disease. So, beyond health gain, which is the key driver of the economic value of innovative medicines in many instances, we can also derive value from having some hedge against uncertainty, which lowers perceptions of both financial and health risk. Economists call this “insurance value.” Formally integrating this concept into value assessment leads to some interesting and important insights.

For example, let’s consider two competing interventions to treat hypertension that achieve the same average reduction in blood pressure, but one has much greater variability in response to treatment. That is, it’s not considered less effective, but you have less certainty about the effectiveness. A traditional CEA would rate the value of these two treatments similarly. For individual patients, however, some may prefer the one that offers more certain outcomes. The value of the more certain treatment would be higher for that subgroup in that case.

What other novel elements of value may be considered in such an augmented assessment?

A different, though related, element from which patients might also derive value has been called the “value of hope” – a potentially ambiguous label for an important idea. The distribution of health outcomes is characterized not only by variance but also by skewness – that is, the degree to which the distribution tends more toward one end of a curve or the other.

To illustrate, say a treatment has only a modest effect on survival across all patients on average but is curative for a small subset. While the curative outcome is not typical, empirical research suggests that people derive value from anticipating, or hoping, that they have a chance to end up in the smaller subset that is cured. So, in technical terms, interventions may have more value when they provide an increase in the positive skew of the distribution of treatment effects.

Are there any other notable departures from traditional CEA frameworks in such assessments?

Traditional CEA, as often applied, likely undervalues innovation in the way that price is handled over a product’s life cycle. The traditional framework for an individual patient over their lifetime assumes that a branded launch price will remain in effect forever. But our system – what you might call market design – is regulated to encourage generic entry after patent expiry, when the price can fall dramatically. Price assessment needs to take this broader and longer perspective. You can’t simply assume that the prices of drugs or their comparator treatments will remain the same over the long patient time horizons that are considered in a CEA, which can be more than 30 years.

Additionally, traditional frameworks do not account for variation in the baseline severity of illness experienced by patients. That is, a slight increase in the number of QALYs could be perceived differently depending on the patient: Those who are very sick are likely to value a small gain in QALYs more than relatively healthier individuals. This can lead to a change in the cost-effectiveness threshold that is applied in an assessment.

What challenges do you face in accounting for the incremental value arising from these kinds of concepts?

There are legitimate difficulties in measuring things like insurance value or the value of hope, and other complexities may arise, particularly in the context of precision medicine and advances in diagnostic technology. Patients may value the chance of long-tailed survival, but what if we developed a companion diagnostic for a specific condition that perfectly identified patients who would benefit? Would we then have to remove the value of hope for those who have learned that they cannot benefit from a given treatment for that condition? And how much of the innovation reward should be allocated to the diagnostic for providing what has been called the “value of knowing” – the additional value provided by the knowledge that such an accurate diagnostic exists?

How do these additional elements affect the challenge of valuing and pricing next-generation treatments, such as gene therapies, that may have longer-term benefits?

Let’s take gene therapy for hemophilia as an example, which is one of the areas I’ve researched.² Gene therapy is designed to produce health gains for patients over many years after only a single treatment. Rewarding that value creation for a one-time administration intervention, however, implies a large up-front cost. This cost can only be justified by long-term health benefits or cost savings that make it just as, or more, cost effective than conventional treatments. And that can raise difficult financing issues in our complex insurance system, which includes many small employment-based plans. Especially for ultrarare, health-catastrophic conditions, the interaction of insurance value, value of hope, and disease severity could contribute to an argument for greater value and, in effect, a higher cost-effectiveness threshold for some next-generation treatments. In my view, this is arguably the most important adjustment that could be made to traditional CEA.

Furthermore, uncertainties about the long-term health benefits of these new cell and gene therapies can make it challenging to assess clinical and economic value at their launch. We often don’t have enough data for gene therapy treatments, so we need to be careful in assessing their lifetime values. However, waiting for more comprehensive data could also result in depriving patients of access to important and innovative therapeutic advancements.

Pivoting to health care concerns that affect broader swaths of the population, how can these novel elements of value help with assessments of vaccines, such as those developed for COVID-19?

We can apply some of these same concepts to the vaccines that are now available for COVID-19. The availability of an intervention that is known to be efficacious provides value insofar as it protects individuals against the physical risk of the disease. That availability also provides value in that it converts the physical risk of becoming very sick – from contracting COVID-19, for example – into a measurable financial, or insurance, risk. In the first stage of the pandemic, our entire population was in great fear of its potential lethality, and we did not know whether we could rapidly develop an effective vaccine. The emergence and validation of the mRNA platform for COVID-19 created massive economic value by lessening these fears in the entire population.

In short, the availability of these vaccines allows for protection against COVID-19, as well as for peace of mind for individuals fearful of contracting the disease. This same logic can be employed in the context of newer vaccines, such as those developed to protect against RSV.

With an eye toward emerging health care technologies, how could augmenting traditional value frameworks with the novel elements we’ve discussed change the way coverage and usage decisions are made going forward?

Whether we are talking about emerging or existing technologies, it is critical to examine treatments with a broader concept of value for society as a whole. Traditional cost-effectiveness analyses may not consider the full picture of how a treatment could affect its users and potential users. In addition, the ISPOR value flower recognized other elements that operate at a societal level such as the sharing of scientific knowledge and impacts on health equity across subpopulations. Some progress has been made on measuring these.

Value frameworks are just that – they provide the setting in which we may consider the full societal value of an intervention. But they are only one input for a complex deliberative process that allows for additional elements of value to be contextualized with the aim that we send the right signals to innovators about what can be considered valuable, what should be covered by insurance, what physicians may be willing to prescribe, and what is desirable for individual use. ■

 

Endnotes

1. Novel elements of the value flower: fake or truly novel? (n.d.-b). ISPOR | International Society for Pharmacoeconomics and Outcomes Research. https://www.ispor.org/publications/journals/value-outcomes-spotlight/vos-archives/issue/view/navigating-the-changing-heor-publishing-landscape/novel-elements-of-the-value-flower-fake-or-truly-novel

2. Garrison, L. P., Pezalla, E. J., Towse, A., Yang, H., Faust, E., Wu, E. Q., Li, N., Sawyer, E. K., & Recht, M. (2021). Hemophilia Gene therapy Value Assessment: Methodological challenges and recommendations. Value in Health, 24(11), 1628–1633. https://doi.org/10.1016/j.jval.2021.05.008