Navigating Coding Changes and Rare Disease Genetic Testing Challenges 

The rapid growth of genetic testing, the challenges it presents for the rare disease community, clinicians, and payers, and the impact on coding practices are explored with Jason Bush, executive vice president of product at Avalon Healthcare Solutions. 

Please share your name, title, affiliations, and a little about yourself. 

I’m Jason Bush and I've been with Avalon for 8 years now. I am currently the executive vice president of our product area. I’m responsible for product strategy, management, and delivery. I joined the company with the role of clinical operations as an SVP and was responsible for a lot of our internal processes and capabilities development to enforce a lot of the policy enforcement on the claims adjudication side.  

I’ve worked in health care for 20 years in both drug discovery and clinical laboratory management spanning from research and development in the clinical lab to clinical trials, business operations, sales, and field specimen collectors for diagnostic testing. Joining Avalon, I’m now a bit closer to the payer side than the provider side but it’s been a good hybrid of both.  

Why has genetic testing experienced a period of rapid growth?  

We could go way back to the 1970s when PCR was actually developed as a laboratory tool but that’s probably a bit too far. Instead, let’s go to the early part of the century when the Human Genome Project was kicked off to map every gene in the human genome. That project was thought to be the end-all and be-all of everything we would need to know about humans. The Human Genome Project did quite a lot to further our understanding of humans and biology. That project led to a lot of technology being built and developed to measure DNA sequences and missing parts of DNA. We realized there’s a significant amount of genetic variation between people, even though we’re all about 99% the same. Variations are what causes all of us to look different, behave differently, and have different susceptibility to diseases. 

With the advent of that additional technology, we can now measure gene deletions, variations in genes, variance of genes, methylation of genes, single nucleotide polymorphisms. All of those items can be measured now, and they all lead to a measurement of something in the genetic space or the genetic makeup of a person. 

Technology has increased our ability to measure, what we can and want to measure has increased, and the cost through the last 20 years to measure something has significantly decreased. As a result, you’ve got a confluence of “I want to know more about what’s going on in my body” as well as it’s easier to measure. So as a result, you’re getting a significant growth, and a lot of that is really trying to be turned into clinical diagnostic tests to correlate a gene with a particular disease state. 

What challenges does genetic testing present for the rare disease community, clinicians, and payers?  

Great question. When you think about genetic testing, and specifically I’ll start with the rare disease community, it is by definition a rare disease. This means the number of people susceptible or presenting with this disease is limited in number. And if you’re going to build a test that is then eventually going to be a reimbursable covered test by a health plan and meaningful to a physician, you really need to ensure that you’ve got the utility of that test and you’ve correlated that genetic test result with the outcome of that disease.  

Clinically, this means the absence or presence of a gene means you have this particular disease. That’s very challenging to test. And to do that test authoritatively, you have to have a certain number of samples. For people with rare diseases, you've got a small pool of people to begin with and then you have to worry about all the other things that go into a laboratory test development: is it analytically valid? Is it clinically valid? Is it clinically useful, meaning will the physicians do something different in their treatment of the individual when they have that information? All of that has to be bundled together to get to a worthwhile test. 

So what does that mean for clinicians? Well, there is no set standard for clinical utility. It is. I don't want to say it is subjective, but it is certainly a little bit more open to interpretation and variation amongst medical directors across plans and across medical directors across laboratories. You're going to have a variety of different opinions. So, it's difficult for clinicians to know if something is a good test for them to treat a patient if they don't have a significant amount of background on how that test was developed, what the test is, what it does, who's providing it, the quality of the laboratory providing it, etc.  

On the payer side, they ask similar questions: Is this information I'm paying for going to be meaningful? Is the physician going to treat the member differently? Is it a worthwhile test that I should cover? All of these factors create testing challenges in the rare disease community. Not impossible, because we're seeing more and more of that come out, but it makes it a bit more challenging to get the necessary numbers and have a high-quality laboratory involved.  

This leads me to the bigger point: all these laboratories are seeking reimbursement for these tests. A laboratory’s goal as an independent business is to get revenue and reimbursement. So, the payers need a way to understand that it's a good test, and the laboratories need to know a standard that's going to be achievable for that test to be reimbursed and covered. That's a challenge. Not all laboratories are operated the same way or have the same quality systems in place, so that provides an additional level of uncertainty.  

Developing a laboratory test is not easy. I've worked on esoteric tests in my life and had to do clinical trials specifically working on that utility component and if the utility component isn't there, then you really don't have a business. But I've also seen laboratories that have tried to do marketing and selling of their tasks before the utility was completed. There are potentially strategic reasons to do that as a business, but from a payer perspective, that's not something that they're necessarily interested in funding. 

Could you share an example of how these changes have impacted coding practices?  

Over the years there have been a variety of attempts to figure out how to code this. The standard coding within AMA for genetic testing is approximately 500 CPT codes. There are about 175,000 tests commercially registered. That means there's not enough coding opportunities to build for the tests that you're going to need and that presents a challenge. Not only that but depending on how the coding system is set up and the number of analytes, you may need to code in multiple different ways. 

There is no standard that the payers, insurers, and generals advise on how they want particular genetic test panels to be coded and billed. There's significant coding variation between payers that laboratories have to manage. There's also a lot of tests that don't quite fit into the coding schema and they have to use a miscellaneous code, and nobody knows what it is until you get some documentation or look into the actual claim record to see what test was performed. And that code is important for a plan to determine if there's going to be coverage and reimbursement.  

There's a situation where there are too many tests and not enough procedure codes. And in the last 10 years or so there's been some novel coding systems that have come out to try to overcome this problem specifically for genetics. 

About 12 years ago out of Palmetto GBA, Medicare Administrative developed the MolDX program and the DEX Registry that provided a unique Z code for every test, but the tests have to be registered. Concert Genetics has done GTUs as a way to provide uniqueness as well. And then with some re-up of the Pamela legislation, I believe in 2014, they required AMA to produce proprietary laboratory assays, PLA codes, which provide uniqueness within the AMA coding system. Each one of those has its advantages and uses in the market. 

All of these coding systems are focused on specifying what is the uniqueness of that particular test. But only one of them considers the quality assessment of a test, and that is what MolDX and Palmetto built in the DEX registry from the beginning. Their design was made to make sure it was a Medicare-approved test. They do require an assessment of the quality of that test prior to that test, actually being reimbursed or covered by medicare fee for service. 

So that's one distinction between the 3 novel coding systems I mentioned. They do provide a technical assessment and a quality metric associated with that. If it is not a quality test, and if it doesn't have utility for Medicare fee for service, they will not cover it.  

How can payers better manage genetic test evaluations, understand spending, and offer value to benefit patients?  

Great question. From a payer's perspective, they're seeing a lot of new tests come onto the market that are being pushed through the reimbursement system, or even through the prior authorization system. And as we mentioned, there's a great ability to measure new things and correlate that with clinical disease and this is part of the promise of the genetics revolution. Can you diagnose more? Can you treat patients better? Can you get them on the appropriate care first?  

And there's a lot of real-world examples where that's valid. Oncology testing is one, testing for mutational-directed therapy. So, there's a lot that's facing the health plans on that front. They need to recognize what they're facing, which is going to be an increasing number of tests coming to the market seeking reimbursement. 

All of that revolves around a good understanding of the science, the coding, the lab operations, and the quality. Physicians are going to need some education on it. They will have pricing and network challenges with laboratories, all those things. So, they need to recognize that this is going to be a growing body of work and potential value to their patients and their membership. So, to manage this, they need to invest in the necessary capabilities, or they need to partner with somebody who's already got those capabilities and can lend that to them. They need to evaluate these new technologies and assess the quality and viability of tests. Are the tests clinically supported? Are the physicians going to change their behavior with this information from the test since a lot of these tests can run in the multiple thousands of dollars to reimburse? It's a significant investment and you can't do it for just everybody, in all cases, because of the expense. 

This also begs the question of how are the tests going to be coded? What's the quality of the test? How do I roll that into my coverage policy? Under what conditions do I wanna say I can cover it? And then how do I cover it, do I? Or how do I enforce that? Do I do that through prior authorization, or do claims adjudication, or some blend of the 2? All of that, if done well and managed well, can speed the test through the process so that laboratories, physicians, members, and patients can understand if it's a covered test. That way there are fewer frustrations from the laboratories and the physicians and greater clarity on pricing and coverage. For the health care system, they’re getting information from the result of that test to the patient earlier with less headache and hassle for most of the parties.  

Physicians are having a hard time keeping up as well. We see a lot of genetic test requests that come in through our prior authorization program at Avalon. Some of what the physician is asking for is not appropriate for the situation, not because they're not focused on doing the right thing for the patient, but because there may be a better test alternative that they were unaware of. Or maybe they're asking for more than what they need to care for the patient. There's a lot of education that goes on through the prior authorization process. 

Is there anything else you would like to share?  

Genetic testing traditionally is managed through prior authorization because these are expensive tests. You don't necessarily want everybody to receive these tests because of the expense and because in a lot of cases, it's not warranted and useful for those members. So prior authorization is the standard way that genetic testing is managed. However, over the last couple years there's been an increased focus at the Federal and State levels to try to reduce when prior authorizations are appropriate. In some cases, this is targeted at genetics, in other cases it's more broadly about the general abrasion that PA produces as a side effect. That is one trend in the market.  

What I've also seen in the market on the health plan side is that traditionally genetic tests have been solely the realm of prior authorization and there hasn't been a lot of involvement of the claims adjudication side. However, now I'm seeing that distinction blur over the last couple of years. In conversations we've had with plans where they're starting to look at genetic testing and saying, “I probably don't need to truly prior off every single test. There are some that I can just do through adjudication. Some laboratories excel at making sure that the specimens are the patients are appropriate and compliant with policy. So can I expedite the process and look at it on the claims adjudication side as opposed to a prior authorization pre-service situation.”  

That blending does sometimes require novel coding systems that I've mentioned previously and all of that links me back to my previous point about health plans needing to invest in the ability to do this or partner with somebody who already does.