Dan Ollendorf, Ph.D., M.P.H.,of the Center for the Evaluation of Value and Risk in Health at Tufts University Medical Center, is co-author of a new book, The Right Price, A Value-Based Prescription for Drug Costs.
Senior Editor Peter Wehrwein interviewed Dan Ollendorf, Ph.D., M.P.H., recently about The Right Price, A Value-Based Prescription for Drug Costs, the new book he co-authored with Peter J. Neumann and Joshua T. Cohen. Ollendorf is director of value measurement and global health initiatives at the Center for the Evaluation of Value and Risk in Health at Tufts University Medical Center in Boston. Neumann and Cohen are also at the center.
Wehrwein and Ollendorf discussed drug prices, quality-adjusted life years (QALY) as the metric for establishing the value of a drug (he is “pro QALY” with qualifications), genericization, international price indexing, and whether a U.S. government agency might be empowered to assess drug value and set prices (Ollendorf doesn’t think it is likely).
Prior to Tufts, Ollendorf was chief scientific officer for the Institute for Clinical and Economic Review, a drug pricing and value think thank in Boston.
You’ve called your book The Right Price. So let's start off with kind of a kind of a stupid question, but maybe one that will reveal something in the answer, which is, are the drug prices right in the United States? You sort of embedded in that title is a question.
It is the case that at times the price is right, because the price is matching up relatively well to the benefit that a drug is bringing to the patient. It is more often in the case, however, in the U.S. that because the price is not set necessarily with any inclination toward the benefit that's brought to the patient, that it's not right. Prices are very high in the U.S. We've seen lots of studies and publications showing that they're very high here — much higher than they are in other countries. And it seems as though when the price is well aligned with the value to the patient, it's almost as if it's an accident.
So prices are too high. In the book you discuss price in a standard sort of supply and demand way. You discuss some of the distortions that exist in the US. On the demand side, we have multiple buyers — the individual, their insurer — buying the same thing and sort of the same time. On the supply side, of course, we have the drugmakers, who by government fiat have monopoly pricing power here, and we do that on purpose, correct. We do that because of the research and development costs.
Right, we do it to encourage innovation. So there is this recognition that during the patent life of particular product that company will benefit from having that monopoly pricing power. At the same time, there is recognition — and I think it's growing recognition — that price cannot be unlimited because it's starting to take a larger and larger a chunk out of the patient's pocket to pay for it.
So it's a situation in which there is that pricing power. But it's now getting to a point where it's too much, essentially, to be affordable for our health system, and, in particular, for the patients and families who have to pay some portion out of pocket for these drugs.
So is the problem really that the price is too high for the individual and therefore it has to do with benefit design and high deductible plans and so on?Or is it deeper than and is it a question that society— even if people were totally insulated from the cost of these drugs — we'd still have a problem.
We do still have a problem because in addition to the patient share, spending on prescription drugs is a growing percentage of the overall spending for both private health plans and public payers. We know that there have been discussions around the future solvency of the Medicare program, for example. And those discussions are getting more intense as spending on prescription drugs is growing in the Medicare program as well.
Now there's an argument from pharma that that the spend on drugs results in less expenditure in other parts of the healthcare system. If people take drugs appropriately, they won't get hospitalized, and hospitalization is fantastically expensive. So isn't there a trade- off going on here?
I think if we were to see spending on drugs really taking a big chunk out of other areas of the health system, we would see growth and overall spending flatten. But that's not the case, either. Spending is increasing in other sectors of the health system as well.
So let's stipulate the prices are too high for the most part. There are, of course, discussions now — a proposal by Democrats primarily — that Medicare began to negotiate prices directly with drug companies. And that is held out as a way to bring prices down and maybe bring them closer to this value that you're talking about. Do you see that as a way of getting there? Are there some policy prescriptions here?
There are. I think one of the challenges with the current bills that are currently in discussion is that one of the ways in which they are proposing to empower Medicare to negotiate prices is to allow Medicare to fall back on an average of — or some sort of arithmetic calculation — of the price paid in other countries — countries with similarities to our health system, similar in terms of economic indicators, etcetera.
The challenge with that is that prices are generally too high across the board. What that equates to is a general price cut for medications that Medicare is paying for without regard to whether or not those medications are bringing significant benefit to their patient populations. So you're essentially cutting the price paid for a product that may cure a rare cancer in a child at the same level as you're cutting the price for a new medication for toenail fungus. So it's essentially an across the board cut without relationship to the benefit being brought to the patient.
In addition, this allows policymakers and politicians to sidestep the thorny issue of whether the use of cost-effectiveness analysis — the use of a formal analysis to understand how the costs relate to the benefits. They don't have to do that if they're simply referencing the price paid in other countries. But most of the other countries that are being proposed as reference countries do use cost-effectiveness analysis to set their prices. So we're essentially importing their approach to value assessment here. That's not something, obviously, that's part of public debate, but it's a key nuance in what's being discussed.
I take your point. I guess the question is, what would be so wrong about that borrowing? Our policy would be reflecting the sort of values that they've arrived at, so maybe that's not such a good idea.
So let's pivot a little bit to this discussion of value. The metric that's used most often in the discussion of drug value is the QALY — the quality-adjusted life year. And you discussed this at some depth in the book, and I highly recommend those chapters. You go into some of the interesting history QALYs. But you basically come down as saying, and I think you invoke Winston Churchill here, who I guess was invoked by Peter Singer, that QALYs are a terrible thing, but it's the best of the available alternatives. QALYs take into account both the length of life and morbidity. So are you pro QALY?
Yes, we are. And we recognize that it has its limitations, [but] it is going to provide some measure of the average experience for a patient with the disease and the potential effects of treatment on that disease. But we recognize that most diseases are quite heterogeneous, and patients will have very different experiences. It is a summary measure and won't necessarily take into account all aspects of value that a patient may consider important. So it's a relatively rough measure. But it is an approach to, as you say, estimate, the effect of disease and treatment on the length of life and quality of life while alive.
It's an important way to think about how the benefits of two different treatments might compare. And so there are critics out there who say that the quality [aspect] discriminates on its face against those with severe disease or disabilities by valuing their lives at a lower level. But really, its use in practice is to compare the effects of two different treatments. And so if you have a treatment that improves quality of life and extends length of life that difference will be apparent in a QALY-based measure, and it is really an important way to think about those benefits.
Now, another thing that we often say is that while the QALY and cost-effectiveness analysis is an input into decision-making, but it's only one. There are other important aspects of new treatments that we should be considered. Is this the first new treatment that's become available for very severe disease in a very long time, for example? Is there a significant burden of disease? So if we're paying a lot of money, for example, for new treatments for hepatitis C, given the public health impact of that spending, that might be worth it, on its face, regardless of what the cost-effectiveness analysis says.
So there are always other considerations that should be part of the conversation, it should not be the sole input into that conversation. But we think it's an important one, because it allows you to compare two treatments to each other.
I think you mentioned in the book that in France, I believe that they have a sort of a simple numerical system, but it for new therapies for a given disease and whether it's a big improvement, a minor one improvement, or the same. But it creates problems with respect to any sort of comparison across conditions.
I understand what you're saying about taking other factors into consideration — the newness factor, which I think is part of ICER’s formulation, I believe. But are there some qualifications to the QALY itself? In other words, not just QALY as one input and we’ve got to take into account these other things, but messing around with the QALY itself?
It is an active area of research. There are studies ongoing to try to understand if there are different domains of value that are not currently represented in the quality that could be added in. So as a very simple example, since we're in the age of COVID, and we're talking about vaccines, there has been a lot of work done to try to understand what the benefits are of herd immunity. It's not often something that's included in economic arguments for vaccines that are submitted to health authorities. But it's obviously a key component of the potential benefits of the vaccine to wider society. And so those kinds of effects around an infectious disease are examples of elements that are not currently folded easily into a quality measure that are under investigation at the moment.
The other thing to keep in mind is that the QALY is one building block of a cost-effectiveness analysis. Those analyses themselves are subject to a good deal of uncertainty because what you're doing is extrapolating the experience from a short-term clinical trial and trying to understand how those effects play out over time. So there are uncertainties there.
So as we discuss in the book, there's always a recommendation to conduct a rigorous set of sensitivity analyses to test that uncertainty, and understand whether, regardless of the range of estimates you're putting in, if you're coming to similar conclusions, that gives you some comfort as to the robustness of your approach.
And, of course, one can always revisit your analysis as new evidence comes in. I believe there's also a discussion, and it does relate to the COVID-19, about whether the QALY is too narrowly focused on health issues and doesn't take into account sort of the societal knock-on effects of some treatments. I believe in your world there's an active debate about how to factor into societal benefit and whether it should be included. So where do you come down on that? And how would you do it if you are in favor — baking in societal value?
We do talk about this. And the gut feeling is generally that the societal perspective is important to estimate if information and data are available, which is not always the case. It's oftentimes difficult to estimate societal impacts with a lot of precision. But it's important to understand what the results are looking like from a societal perspective and compare that to the health system perspective, in part, just to see if there's a difference. It may make a significant difference. In other cases, it may not.
It's also important to reflect on what the decision problem is and who the decision maker is. So you might think that there is significant benefit to, let's say, a therapy that improves the ability to return fully to work and be more productive. That seems like a no brainer. But if you think about the way the health system works in our country, that's much more important to the self-insured employer who's going to be fully at risk for the costs and benefits of the health plan to their employees, then it would be to an employer who's contracting with the health plan is not necessarily all that focused on productivity benefits; they're more focused on where the savings might be to the health system in direct medical costs.
So the perspective is important to think about whether it's societal or health sector, but it's also important to think about who's actually taking that information and making a decision based on it.
And who is at risk of the financial consequence.
You also discuss in the book, in terms of pricing and value, the dynamic of drugs becoming generic. This is a fraught area. As we know, drugmakers have come up with incredibly ingenious ways to keep things from becoming generic — the patent thickets and whatnot. And then we've seen a phenomenon where even generic drug prices are increasing fantastically. So that the notion that generic pricing is some sort of Holy Land for pricing is eroding. So could you just rehearse your argument [about generic drug prices] that you present in the book.
It is kind of a minefield right now, genericization. At the same time, despite the fact that taking into account changes in price is a recommended methodologic approach in the health economics community, it really has not been done very much.
And I know when I was at ICER, we got some challenges around our lack of consideration of potential generic pricing in our own modeling efforts. And our response was, in a very similar vein to what you just described, it's very difficult to predict what will happen with pricing, when generic entry will actually happen, because some of the gaming that industry does. What will then happen after a drug goes generic, will there be enough competition from generic manufacturers to really drive the price down? There are lots of unknowns. But at the same time to not actually explore that in a modeling approach seems like a missed opportunity.
We think about this in a very expansive way, not necessarily just about genericization, but about all price dynamics, because as we know, during the patent life, the price tends to increase. And sometimes, just before a drug goes generic, the price increases a lot. Then when generics come in, the savings may be limited with one entrant. But then when the two or three entrants come in, that's when you see the real savings. And so there are ways to explore this in modeling, just like you would with any other parameter. So why not do it and see what the effect is?
We've been discussing distortions in price and coming up with something that's a little bit more rational, systematic, a QALY-based system with all the qualifications and other things that have to be factored in. But this doesn't really address the question of who or what should do this work.
We're in a funny situation now when ICER does seem to have staked out this territory and is influential, but it's a private entity. And, you know, it makes reports and nobody has to follow them. So I'm curious, your ideas about who should do this work and with what sort of authority? Should there be a federal board or some entity? Or might ICER be deputized in and take on a sort of governmental agency? Who should do this work and what authority should they have?
Readers of the book will, if they didn't already know, will learn that we had a technology assessment agency in this country — an official government agency that focused not only on health, but also cost-effectiveness questions in other sectors as well. It was a quasi-governmental agency, did receive government appropriations, but had a very balanced board in terms of stakeholders and special interests. It was an interesting model and it did that a lot of important methodologic work, but in its independent role questioned the scientific credibility of Ronald Reagan's Star Wars program and then started to die a slow death afterwards.
So the idea of an official government entity that would use cost effectiveness and the QALY to inform public policy seems very far fetched at this point, given the history with not only the Office of Technology Assessment but the discussions around cost effectiveness and the QALY that have happened politically since then.
So I think our feeling is that this kind of work, if it remains in the private sector, could be coordinated by some sort of quasi-governmental entity that provides some guidance on methods and certifies the approach that's taken. And possibly there could be some ability for Medicare and Medicaid to certify, for example, that they've at least paid attention to the reports that are being produced, that they've used that in their own considerations about coverage and negotiation.
But I think it would probably have to stop there. Because I don't think we're ready as a country to officially embrace a formal, governmental approach to health technology assessment and cost effectiveness for the purpose of price negotiation.
And is it that's because we're not ready for it as a country? Or do you think it's because of the power of the stakeholders who would resist it?
The country's ready for some drug pricing relief, there's no question about that. But I think that we argue in the book that this is another example of American exceptionalism. The idea that we should not have our choices dictated to us. And so whether or not the majority of the populace feels that way, we know that people will take political advantage of that idea.So I think that the notion that there could be some group of bean counters, making some determination about what the right price looks like, is anathema to many in this country.
But if there was an entity with some that offered up ways of doing this, sort of a Good Housekeeping Seal of Approval model, that might fly.
Right, as well as some robustly conducted analyses with some important data that could be used as part of a conversation, again, as we describe, a critical input into the conversation, but not the only one.