How to Make Value Assessment More Relevant to Healthcare Decisions

Jennifer Bright is the executive director of the Innovation and Value Initiative.

In American healthcare today there are certain hard truths, starting with this one: far too many prescription drugs are shockingly (and often unjustifiably) expensive. There is a second truth as well—the cost of healthcare, drugs or otherwise, is seldom based on its actual value. But while payers understand value largely in the context of utilization, efficacy, and cost, we still lack the ability (and the will) to understand and account for value from the perspective of patients, especially those living with chronic diseases.

The Institute for Clinical and Economic Research (ICER) has helped us confront the question of value in the face of rising prices by providing analyses of new and often expensive drugs to posit if they are “worth it.” Most recently, ICER released its final report analyzing the value of a new class of therapies for rheumatoid arthritis (RA): Janus kinase inhibitors (JAKs).

The analysis is important. Unfortunately, it doesn’t answer some of the most important questions about the value of these drugs to patients and payers, including which therapies may deliver the most value to specific patients—and in the context of a lifelong chronic disease, at what point in a sequence of different treatments over time. This isn’t the fault of ICER but rather the shortcomings in the ways we have approached value assessment to date.

Typically, we have relied on narrowly-focused models that evaluate drugs in a head-to-head comparison, based on clinical trial results and healthcare cost data.

The reality is, real-world treatment paths don’t often work this way, and there are other variables in the equation that are often overlooked—starting with the patient experience. Drug X may be a new drug on the market that is relatively inexpensive and performs better in clinical trials than existing therapies. On paper, according to how we currently analyze value, it’s “a winner” for payers. But what if, in the real world, taking the drug is unpleasant or onerous for patients—because they have to miss work to get the treatment, for example—and many patients discontinue treatmentdespite its effectiveness? What if it doesn’t work as well for patients with certain comorbidities, or over a certain age?

Alternatively, what if Drug X is a “loser” when compared head-to-head, but in reality, it is highly effective for patients for whom a previous sequence of drugs has failed? These are the types of issues that can have significant impacts on both patient-level outcomes and health system costs -- and that drive questions important to patients, payers, and physicians that current models can’t answer.

That needs to change. As a first step, we need to form new partnerships across the healthcare industry to explore models and data that can answer questions that matter. The nonprofit Innovation and Value Initiative (IVI), has taken the first steps in building a model that supports evaluation of treatment sequences and includes factors defined by patients as important to both their clinical goals and quality of life goals. It’s a small but critical advancement. IVI’s open-source RA value assessment model could enhance the work ICER has done. Our model could answer critical questions like how patient heterogeneity affects treatment response and optimal treatment sequence and how total costs of care and economic burden on families and caregivers can show us a 3-dimensional picture of “value.”

I emphasize the word could because we haven’t answered these questions yet, due to challenges in accessing and “fitting” real-world data into value assessment models such as our RA model. Partnerships—with researchers, employers, health plans, or analytics vendors—could explore how to collect (or begin to collect) and bring together multiple streams of data to give insights into critical aspects of drugs such as:

• Frequency and reasons for treatment switching
• Differences in effectiveness across the population
• Total costs associated with treatment failure or sub-optimal sequencing

Together, we could offer patients, their providers, and payers so much more insight about the full range of RA treatments and about how different therapies work for patients with very different needs.

Despite our collective concern about high drug (and healthcare) prices, of course, the health industrial complex as a whole has a vested interest in the status quo. Yet while change is hard, it is not impossible. Public commitments to more transparency of data and fair pricing are encouraging, as is growing interest and awareness that approaches to, and methods for, assessing value must evolve. If everyone including current and former FDA leaders and patient leaders know that the key lies in harnessing data created in real-world treatment experience, what is holding us back from finding every way to access, aggregate, and deploy such information for the public good?

We need to form new and innovative partnerships—now—to accelerate this next level of value assessment. And this work needs to happen irrespective of our other current debate about how we pay for drugs, with the convoluted structure of health plans, pharmacy benefit managers, launch prices, and rebates. If we focus only on the concept that value equals best price, we may create effective policy that lowers the unit price of treatments, but still end up paying far too much for treatment that may be suboptimal for patients that don’t match a clinical trial profile.

Let’s make 2020 the year we all agree the status quo isn’t good enough, and the time we finally come together to improve value assessment in ways that are most meaningful for patients, payers, and providers.