Do Discrepancies Exist Between Payer Drug Policies and FDA-Required Labeling?

James D. Chambers, PhD, MPharm, MSc, associate professor of medicine at the Tufts Medical Center Institute for Clinical Research and Health Policy Studies, discusses his recent study that compared FDA-required labeling with payer drug coverage polices. 

First Report Managed Care:  Can you tell us about yourself and your research interest?

James Chambers, PhD, MPharm, MSc:  My name is James Chambers, and I'm an associate professor at Tufts Medical Center in Boston. I work in a center called the Center for the Evaluation of Value and Risk in Health, which we look broadly at any aspect of value in the healthcare system.

A lot of what we do is look at medical technology. My research interests are predominantly to look at how the commercial payers and government payers are covering and reimbursing specialty drugs for high-cost medical technologies in general.

This study which we're talking about today is based on an initiative we have called SPEC, or the Specialty Drug Evidence and Coverage Database, which is our attempt to quantify how commercial payers are covering specialty drugs including the types of restrictions that the payers may put on coverage decisions for specialty drugs.

Specific to this study, the evidence base that the payers cite and the decisions, whenever they're posting them online. My focus is on that specialty drug space.

First Report Managed Care:  Can you briefly discuss the objective of your recent study and why was it important to conduct?

Dr Chambers:  We had been planning this study for a while. We think it's quite important in that payers have a very difficult job in formulating these coverage policies. The evidence base that's available to them forms a fundamental component of their formulation of these decisions.

However, there's limitations in terms of how they can communicate with product manufacturers. Product manufactures may have additional information that's not in the public domain or have a better understanding of the clinical trial data that supports a product. Insurance companies are limited in terms of how they can communicate with manufacturers in regards to that evidence.

To make it somewhat straightforward, manufacturers are typically, or historically have been limited to the contents of the FDA label for products in what they can communicate with insurance companies.

What we wanted to do was to compare the evidence base that payers were citing in the coverage policies, meaning the evidence that payers were telling us that they reported reviewing when formulating that coverage decision and to compare that evidence with the FDA label in terms of clinical endpoints.

In other words, we're trying to determine how often payers were looking at evidence that was not represented in that FDA label. We were looking at clinical endpoints, patient-reported outcomes, economic utilization-type endpoints and so on.

What we were trying to do is trying to quantify, I think for the first time, through a number of case studies -- five different diseases -- just how often these insurance companies were going beyond the FDA label in terms of the evidence that they were considering when formulating their decisions.

First Report Managed Care:  Can you highlight the coverage policies you identified for drugs indicated for MS, RA, juvenile idiopathic arthritis, ankylosing spondylitis and non-small cell lung cancer?

Dr Chambers:  Whenever we show it to the industry, we wanted to choose some representative diseases. We wanted to choose diseases or indications that had a number of drugs indicated for them. In other words, it just wasn't a single product that payers would find it hard to not cover.

We chose these five different diseases -- ankylosing spondylitis, juvenile idiopathic arthritis, multiple sclerosis, non-small cell lung cancer, and rheumatoid arthritis -- because they were very well represented in our database. Each of those conditions, at the time we did the study, had at least five drugs indicated for them. Multiple sclerosis I think had 13.

There were lots of drugs, and therefore lots of coverage policies with which to perform the analyses. In terms of the coverage policies themselves, what do we include? We include in our database the SPEC database and publicly available coverage policies.

We go to payers' websites, we identify the publicly available coverage policies. That's the source of the information with which we build the SPEC database. In SPEC we include 17 of the largest 20 commercial payers in terms of covered lives. These are the largest national, and sometimes regional payers. You can guess who they are.

Like I say, we go to their websites, identify the coverage policies and that's how we built this study. We looked at the citations that the payers included in their coverage policies for the drugs indicated for those five conditions. That's the evidence base that we scrutinized to determine this difference in regards to the endpoints included in the study under the FDA label.

First Report Managed Care:  How many drugs were used in the study in total? Can you explain how these drugs were selected and how these relate to payer policies?

Dr Chambers:  I'm not sure exactly how many drugs we included because we have five indications here. Some of the indications are somewhat overlapping in terms of the drugs. I'm not sure how many drugs we included. We included 41 drug-indication pairs.

Because, say Humira is indicated for rheumatoid arthritis, ankylosing spondylitis and I think juvenile idiopathic arthritis, it appeared three times in the sample. There were 41 drug-indication combinations. Humira was in that three times.

We selected these diseases because they're very well represented in the database. We have comprehensively covered the autoimmune drugs indicated for those diseases in our database, which is how we selected them.

First Report Managed Care:  Can you detail the findings for rheumatoid arthritis drugs in relation to your study endpoints?

Dr Chambers:  Rheumatoid arthritis is an interesting example. There were 10 different drugs indicated for rheumatoid arthritis that were included here. We found that overall, when we looked at randomized control trials...we stratified the sample by randomized controlled trials and all the clinical studies.

Just to recap what we did here -- we looked at the randomized control trial in the FDA label, we identified the endpoints that were included in the registration study that was reported in that FDA label and then compared those endpoints to randomized controlled trial information that the payer cited in the coverage policy.

We found that roughly a quarter of the time, 25 percent or so of the time, the payers were looking, or at least reported reviewing studies that included endpoints that were not in that FDA label. That number increased when we looked at the other clinical studies. These were maybe single-arm studies or other clinical studies that wouldn't be reported in the label.

We found that, it was 41 percent of the time those studies included endpoints that were not in the FDA label. There was a lot of information in those studies that the payer would not have garnered from reading that FDA label.

In terms of what those endpoints were, we found four of those endpoints were clinical endpoints -- your typical clinical endpoints related to rheumatoid arthritis. Interestingly though, 23 of them were patient-reported outcomes and 6 of them were economic utilization endpoints.

A lot of the additional information that the payers were reviewing when formulating the coverage decisions actually was patient-reported-type endpoints. In my paper -- if I can find it here -- we found that, oftentimes there were quality-of-life endpoints. There were a lot of utility measures in there. The EQ-5D and SF-36 featured extensively.

There were also endpoints specific to sleep and fatigue, for instance, meaning that these studies...We're showing that these drugs are having an impact on these patients beyond your traditional clinical endpoints.

First Report Managed Care:  How can communication between the FDA and payers be improved and why is that important?

Dr Chambers:  First of all, I think it's very important. There's a lot that can be done to improve the efficiency of the process of a drug being approved through to a patient receiving our product.

There was an interesting policy initiative, and that's still ongoing. It was implemented a number of years ago, called Parallel Review. This focus on Medicare and the FDA. What it set out to achieve was to reduce the lag between a product being approved and a Medicare beneficiary receiving the product.

It had Medicare review the product even prior to the product's approval, meaning that the FDA and CMS -- the Centers for Medicare & Medicaid Services -- would review the clinical data supporting the product at the same time. Therefore it wasn't a sequential FDA approval, then CMS coverage decision, before the patient would have access to it.

There's lots of benefits of that -- one is that patients would gain more rapid access to technologies. But also, it moves communication away between the payer -- whether it be Medicare or the commercial payer -- and the manufacturers.

If we have greater clarity in terms of the evidence that payers require or would like to see when formulating the coverage decisions, then it is possible then for product manufacturers to better design evidence -- not just to meet the requirements of the FDA, but also to meet the requirements of payers.

As we see payers' role becoming increasingly important in terms of guiding which patients have access to which products, it's almost like a handbrake sometimes in terms of coverage of these new products.

If we better understand what payers require whenever they make those decisions, hopefully we could more rapidly develop evidence that better supports those products and ultimately, I would hope, speed patients' access to these products.

First Report Managed Care:  What can payers take away from your study?

Dr Chambers:  It's a good question. It was an examination of what payers were doing in some way. For payers, it really does emphasize that information that the FDA produces were extremely valuable -- to emphasize that point, of course.

Very robust randomized control trial evidence is critical for us understanding whether products are truly safe and effective. If we didn't have that, of course, we'd be making inappropriate coverage decisions, potentially harming patients and providing them access to products that they maybe shouldn't have access to.

It emphasizes that there's a lot of information not on the FDA label that is helpful to them, including clinical endpoints, including more softer patient-reported outcome-type endpoints. Also, the economic utilization-type studies that would not be included in the FDA label, but would help payers to make more efficient coverage decisions.

First Report Managed Care:  Finally. Is there anything else you'd like to add?

Dr Chambers:  A couple of things, maybe. One is that payers have a very difficult job at the moment when deciding how to cover and reimburse these technologies. We have an influx of wonderful technologies, breakthrough technologies that have huge potential.

If we're unable to afford them it makes the payer's job exceedingly difficult, increasingly difficult, to absorb the cost of them and to provide patients appropriate access to them.

Ultimately, if the payers are going to be making decisions that are value-based, allocating resources in an efficient manner, they need to have evidence to help them do that. This study was trying to have an insight into the types of evidence that they are basing those decisions on.

The second thing is that, one of the reasons for doing this study is because there's a lot of debate about appropriate communication between the health plans and product manufacturers.

Proponents of the status quo -- when there's more limited interaction between manufacturers and payers -- would argue that if you relax the rules, there's a potential there to be miscommunication of quite important clinical data which may ultimately lead to patients receiving products that may increase the risk of adverse health outcomes.

Opponents of the status quo would argue that there's great benefit to increase communication. By insurance companies interacting much more thoroughly with product manufacturers, they have a better understanding of the products and can make more appropriate decisions.

The findings of our study would seem to suggest that payers are going beyond the FDA label in terms of the endpoints that they consider, which would suggest that there would be benefit in greater communication.

However, we need to be careful to make sure that this is regulated and that we're trying to insist...we're trying to put the patient at the center of this and make sure that decisions are appropriate.

The very last point -- with the 21st Century Cures Act there has been some relaxation in the rules of communicating healthcare economic information, which I think is a good thing. Hopefully that will allow payers to more information and help them make more value-based decisions.