Downcoding, electronic prior authorization and Medicaid work requirements: What practice leaders should watch in 2027

Medicare Advantage and Medicaid plans must support electronic prior authorization by Jan. 1, 2027, under a Centers for Medicare & Medicaid Services (CMS) rule, and some payers have raised concerns about meeting that deadline.

Madison Hynes, M.P.P., an associate director of government affairs at the Medical Group Management Association (MGMA), said no hands went up when she asked attendees at two of her sessions whether their electronic health record (EHR) vendors had contacted them about it.

Hynes and James Haynes, J.D., also an associate director of government affairs, first delivered MGMA's Washington update Sept. 27 at the association's 2026 Annual Conference in San Antonio. Medicaid work requirements also take effect Jan. 1. Two days later, at a government affairs town hall, practice leaders told MGMA that an approved prior authorization is no guarantee of payment.

MGMA has also drafted legislation that would bar Medicare Advantage plans from automatically downcoding claims, the issue Hynes said the association has heard about most from members over the past year. She has been shopping the bill to congressional staff from both parties.

Before joining MGMA, Hynes was a senior consultant at Duty First Consulting, where she worked with CMS on Marketplace policy implementation.

Physicians Practice caught up with her in San Antonio, where she explained how practices can tell whether a claim has been downcoded, what happens if payers miss the Jan. 1 deadline, what Medicaid work requirements could mean for paperwork and uncompensated care, and how members can join MGMA's new work groups.

Our conversation, lightly edited for length and clarity, follows.

For a practice leader who couldn't make it to your Washington Update session, what's the one federal issue practice leaders should be watching right now?

Madison Hynes: If it's one thing in the next month, or maybe six weeks at this point, it's going to be the 2027 fee schedule. We have seen cuts to the conversion factor for both advanced alternative payment model (APM) participants and non-qualifying participants, and we know that has a huge impact on practices across the country.

So if it's one thing, it's looking at the final conversion factor, as well as a few other payment policies in the fee schedule that could be finalized, including changes to the 25 modifier and G2211. That will have the biggest impact immediately, in the next six weeks.

Then, looking forward, we're hoping for extensions of other policies, like the APM incentive payment and the geographic practice cost index floor, as we get toward the end of the year. That will be our focus moving forward.

The Patients First Act would give physicians an annual Medicare pay update tied to inflation. In plain terms, how would that change things for a practice compared with the fight over cuts every year?

What we've been pushing for for a while, as most folks probably know, comes down to this: The fee schedule is one of the few payment mechanisms that's not tied to inflation. Other payments are tied to things like market basket inflation, but Medicare payment for physicians isn't right now.

That's created an enormous problem overall and built in continuous cuts. Even when we do see an increase, like we've seen in some recent years, it's still not keeping up with inflation.

What this would do is tie the conversion factor every year to an inflationary update. There will still be financial challenges, and we can assume there will be tweaks to this legislation and policy over the years.

But at least for now, this would give a little bit of relief, so we aren't always fighting this end-of-year payment cut and always relying on Congress to come in and intervene at the very last minute. So this would be a huge opportunity. We know that Medicare payment does not cover the cost of care, and this would be the first step toward addressing that and making sure we don't fall any further behind.

You said there's some hesitation among payers about whether electronic prior authorization will be ready by Jan. 1. What happens if they miss that date?

That's the rumbling we're hearing in D.C. a little more recently: concerns from payers that they don't have enough people to test things, or that they aren't sure everything's ready to be implemented. They have had a few years to get this underway, so we do have concerns about why we're just hearing about this three months before the implementation deadline.

In terms of what happens if they don't, there are three main components of electronic prior authorization. The first is the ability for practices to use their EHR system to determine whether a service needs prior authorization.

The second is providing the template, and the third is the actual transmission of the electronic prior authorization request. There are some concerns that not all three steps could be implemented by the Jan. 1 deadline, and the Department of Health and Human Services could have some authority to enforce any delays or punishments for delays.

On our end, here's what we're most concerned about. You were in my session, and when I asked who's heard from their vendor about electronic prior authorization, no hands went up. The same thing happened today. For all this to work, there needs to be testing, as with any new implementation of technology or process. We aren't hearing from practices about any testing being done or about them being contacted by their EHR vendor.

We'll continue engaging with groups through the end of the year and into 2027 to get a better understanding of what's actually happening on the ground. This regulation was put into place to have real impact, and if it's not having that impact, I'm sure there will be additional concerns from both the administration and Congress.

The WISeR model brings prior authorization into traditional Medicare in six states, including here in Texas, and pays technology companies to run it using AI. Some lawmakers in Washington have already raised concerns. What worries MGMA most about that initial setup?

Overall, the model seems like it could be a slippery slope. Right now, it's only six states and only a handful of services. In some states, we have heard it's going OK and relatively smoothly.

In other states, we've heard it really hasn't gotten off the ground, that there continue to be delays and that these vendors are not abiding by the timelines CMS set. So right now, it may not be the loudest, biggest issue, and many practices really aren't impacted by it.

What we're concerned about is that CMS is testing WISeR in six states with six different vendors, and this is really just a small sampling to extend either the number of states in which WISeR is implemented or the number of services. For us, that's the biggest concern.

We also have concerns about the use of AI to review claims. I think we'll talk about downcoding in a little bit as well, but that's a major concern of ours. We hear a lot from members who say they're submitting all the documentation they need for prior authorization, and they think it's just being automatically reviewed by some system or algorithm and denied based on that.

More fundamentally, the idea of the WISeR model is to create savings. There's not a lot of transparency on the payment to the vendors, but in theory, they're being compensated for saving the government money. So are they being compensated for denying more care? That's a huge concern for us as well in terms of access to care.

We'll have to see where things move. As you mentioned, there's been a lot of talk in Congress about this, especially from folks in those individual states. Washington's delegation has been pretty vocal against it, and we'll have to see how much they're able to push back.

More information has been released, I believe through a recent lawsuit or a Freedom of Information Act request, that exposed a lot more about the model, the delays and the concerns, including documentation of vendors not implementing the portal to submit prior authorization on time. The ramp-up was very quick as well. With more of that exposed, hopefully we're able to shine a light on it.

Prior authorization is the largest burden in Medicare Advantage. It's one of the biggest administrative burdens across the health care industry for practices, and expanding it into traditional Medicare is pretty concerning for us.

Downcoding is the complaint you've heard most from members this past year. What does it look like when it happens, and how can a practice tell if it's happening to them?

This is such a black box, and I think that's one of the biggest issues with it overall. With prior authorization, you generally know that something needs a request, and you can go through and figure out what might be subject to that. Downcoding is really challenging because it's happening in the dark of night.

Any payer is able to release a downcoding policy, assuming they release it publicly or in a way that's sent to practices. I talk to policymakers and their staff all the time, and I print out some of the policies. One policy I print out as an example basically just says these four codes may be downcoded from a level five to a level three. Nothing else.

Other downcoding policies from plans say they're really only targeting the bad actors, but we don't know what qualifies someone as a bad actor.

Sometimes they allow the provider to appeal whether they're subject to downcoding, but we don't really know what that process is or how many providers are impacted, and there's no transparency on the thresholds for who might be impacted. And that's for the policies that target bad actors instead of being blanket policies. That's one of the bigger challenges.

For practices, it's really challenging, too. Obviously, they also deal with denials and delayed payments. But with downcoding, they aren't receiving a notification on any claim or portal that says, "X, Y, Z code has been downcoded from a four to a three or a five to a four."

Really, the only way to tell is to go line by line and look at the actual payment for each service. It's extremely burdensome to figure out what's being downcoded and then, once you identify it, to actually go through the appeals process.

Our concern is that practices may not know they're subject to this, and they may not have the capacity to appeal. But generally, when they do appeal, they're often finding success.

We are very concerned about the automatic nature of this. If an insurance company were reviewing materials and saying the severity is really a level three instead of a five because of X, Y and Z, and they were communicating that, we would understand that concern. But the automatic nature of it is very concerning, because we don't know that any material is actually being reviewed. That's what we're trying to address in our draft legislation.

We don't want this automatic downcoding to exist, because it's extremely unfair to practices, especially considering it's very rarely communicated to them directly.

Could you tell me a little more about MGMA's draft legislation?

This is one of the things we hear most from members right now, in addition to prior authorization, obviously. We worked with our attorneys in D.C. to draft legislation, and we thought about it in terms of the dream state, which would be a clean prohibition on downcoding.

When we write that legislation, the important thing for us is to understand that today it's downcoding, but yesterday it was prior authorization, and we don't know what utilization tactic plans will come up with tomorrow.

So what's important for us is not just to say downcoding, but automatic downcoding or anything similar that uses an algorithm, artificial intelligence or a protocol, which is another term we have in our draft legislation.

It's really important for us to include language that covers all scenarios as much as possible, so we don't exclude one word or one option. We know insurance companies will likely try to find a way around it, so we're making sure we cover all our bases as much as possible. That's really our goal.

We also understand there could be some pushback from insurance organizations. We have talked with some policymakers and staff in offices on Capitol Hill about what to do if there needs to be an exception. Really, these policies should only last 90 days, and they should only target bad actors who meet certain thresholds, which should be set in the legislation as well.

The climate on Capitol Hill is that there's a lot of concern about Medicare Advantage overall. We see that in a lot of bipartisan prior authorization legislation, and in the last year, we've seen it in a lot of bipartisan legislation addressing fair and prompt pay in Medicare Advantage. I think we're going to see downcoding woven into that more, hopefully, especially in the next Congress. I think it could be a top health care priority for them.

MGMA's draft downcoding legislation specifically names AI and other algorithms, and you're shopping two versions of it. What's the difference between the two, and why do you feel the need to have a backup?

To clarify, the two versions of our legislation are a clean prohibition and a prohibition with an exception that targets specific bad actors meeting a certain threshold, with only 90 days, certain communications and certain denial and appeal processes for physicians.

In terms of the prohibiting language and what we're trying to prohibit, it's automatic protocols, algorithms and AI. Again, we're really trying to broaden the language. We didn't have AI in there the first time, and we thought about it a little more and wanted to be very explicit about it.

That gets back to including language that covers any type of program or strategy that leads to automation. That's our thinking there, just to make sure we're covering all our bases.

States have to start Medicaid work requirements by Jan. 1. What should practices expect in paperwork and unpaid care?

It's going to be a pretty big change at the state level, and holistically, a few things will happen with work requirements in communities across the states.

First, there's going to be confusion over who's subject to work requirements. That may lead folks to simply not reenroll in Medicaid, or to not understand that they might qualify for an exemption or that they actually aren't subject to work requirements.

This is for the expansion population in Medicaid expansion states. It's not every single person, so there are quite a few caveats. If states don't do enough in how it's marketed and implemented to explain who's really subject to this, that will lead to turnover among Medicaid beneficiaries and folks who are accessing care.

So more holistically, we expect practices to see a rise in uncompensated care. I was speaking to someone who works in an emergency department, and they're already preparing for an increase in emergency department visits. In primary care, too, you might see a rise in uncompensated care, or people just not coming in to seek care, which is also really concerning overall.

For work requirements specifically, one thing we're urging practices to keep an eye on is the medical frailty exemption. The One Big Beautiful Bill Act set a few exemptions to the Medicaid work requirements, like being a student or volunteering.

Then there's medical frailty, which is essentially an exemption that says, "I'm too sick to work." CMS is not super explicit about exactly what qualifies as medical frailty, so it's falling on the states to determine that.

What CMS has released is general, three-tier guidance on how states could evaluate medical frailty based on certain coding and information they could gather from an electronic health record. That ultimately puts the determination of medical frailty on a provider, and we have some concerns about how those codes would be used.

There's also a cross-state concern. I live in Northern Virginia. A beneficiary who is exempt from Medicaid work requirements for medical frailty in Virginia could move across the street to Maryland or D.C. and not be exempt.

Third, and what we really want practices to be on the lookout for, state Medicaid agencies have three months to get this in place. We're concerned that if they aren't able to set up back-end systems to evaluate these codes, they might make documentation requests to practices, and practices could be required to respond.

We really don't know how it's going to play out, but we're thinking through the scenarios of how it could affect practices, financially in terms of uncompensated care and fewer patients coming in, but also in terms of potential administrative burden.

What's one tip you would give a practice administrator that they can put into place first thing when they're back from San Antonio?

I think it's just really important to be in the know and stay up to date on all these policy changes. Each one of these may sound very small. The conversion factor is one component. But if you bill a lot of G2211 codes and that changes, all these small things are really going to have an impact.

So it's really important to understand what's potentially coming down the pipeline, as well as to be able to step back and see the broader picture of opportunity.

A lot of our presentations are a laundry list of not-great things and policies that we know are presenting challenges to practices.

So perhaps not a tip, but a call to action: We love to hear from members about what's impacting your practice. We can't do our job unless you give us stories to share with policymakers to make change. So the tip is to stay on top of your news and understand the latest developments, and the call to action would definitely be to reach out to us and share your story.

At the end of your session, you mentioned specific work groups members can join. How can they get involved and share their experiences with you?

MGMA members can always reach out to us at govaff@mgma.org, and we're happy to talk with you and get you set up in a work group. Essentially, we're standing up a few ad hoc work groups to make sure we're getting more robust member feedback on certain issues.

Before we submitted our comments on the 2027 proposed fee schedule, we met with a few groups on the Medicare Shared Savings Program, the Merit-based Incentive Payment System (MIPS) and MIPS Value Pathways, and billing and payment. That gave folks a chance to gather virtually, understand what the proposals are and give their feedback, and we could ask them some live questions.

Looking forward, we'll also be trying to convene a group around Medicare Advantage, which I think will be a huge priority both in D.C. overall and for MGMA moving forward, as well as one on health IT. So there are a lot of great opportunities, and we'd love to engage.

We're here in San Antonio for MGMA's 100th anniversary. How has your conference been, and what are you enjoying so far?

It's been great. We've had a lot of great conversations with members, and that's always the most rewarding thing for us: being stopped in the hall by someone saying, "This Medicaid thing is happening in my state. Do you think it's tied to other things?" We're really happy to see the enthusiasm.

We're also really happy to be here talking about opportunities. Sometimes it feels like every year the fee schedule comes out and it's, "What's cut? What's cut?" While we're also addressing that and advocating on it, this is an opportunity to talk to folks about how things like the Patients First Act could really move things forward.

Tomorrow [Sept. 29], we're also going to have a town hall, and we're really looking forward to hearing more from members there to understand what else we should be prioritizing for 2027. I think that'll be a great opportunity for everyone.

Physicians Practice was in San Antonio at the MGMA Annual Conference, Sept. 27-30, celebrating 100 years of MGMA, attending sessions and speaking with industry leaders. Follow our coverage on our MGMA conference page.