Strategies For Reducing Gap and Chase Time Frames

In the context of risk adjustment gaps, there are two overarching goals—to identify and accurately predict member conditions. To do this, health plans are reliant on the amount and quality of the data they have—primarily sourced from clinical and claims data. Plans need this data to produce intelligence and directives to providers on how best to proceed on a proactive care path.

Due to COVID, plans experienced a decrease in access to data and that resulted in unique challenges for risk adjustment and their ability to supply predictive analytic recommendations to providers. As we return to normalcy, plans can expect a huge influx of data in 2022 and 2023 as members return to their normal care and procedures. This is going to be great for generating greater intelligence and actionable insights, but it’s also going to present some unique challenges and strain on some plan’s operational systems, especially if they’re manual.

For plans who have the technology and capacity to process large amounts of data, these coming years may be very fruitful for risk adjustment. The large data sets will provide a lot of intelligence for predicting member gaps and creating strong member segmentation models.

Host: Welcome, this is Episode 4, Strategies For Reducing Gap and Chase Time Frames. I’m excited to welcome Abe Chhaya to the show. Abe is a global IT leader with expertise in quality assurance, DevOps, and product management for healthcare IT. Welcome Abe. 

Abe: Great to be here.

Host: Let’s start by talking about the gap and chase timeframes for risk adjustment. Why does it matter and how does it impact risk adjustment operations?  

Abe: First and foremost, when we talk gap and chase in the context of risk adjustment operations, it all boils down to identifying and accurately predicting member conditions. Whether that is member conditions that the member already has or based on the claims and clinical data, member conditions that would predict what the member could have in terms of their overall health. So, when we talk about the gap process there’s really two aspects of it, and this is very foundational to risk adjustment and Medicare Advantage. First, the gap and chase process ensures that health plans are appropriate    So, that’s important in the first lens, because it allows member premiums to remain both competitive and affordable. If the health plan is able to predict accurately, they can take care of their member demographic and improve the member’s health by accurately identifying what type of gaps the member could have.  

For example, I’m a health plan, and I know that Jane Doe has diabetes. I’ve looked at the claims data, and I know they haven’t had an A1C check in six months. Well, that’s a big problem. I will send an alert to the provider to say, “This is a gap, and you need to connect with this member to ensure they get an A1C check.” By not having regular checkups on A1C and blood glucose levels, this member is at risk for other downstream illnesses. From CMS’s perspective, this shows that the health plan and provider network are taking care of the members appropriately, so CMS will reimburse the health plan for their part of the member care. This allows the health plan to keep their member premiums competitive and affordable and also have access to an array of different provider networks that are mature. When you select a health plan, you select it because of its provider network and the cost of care to the member. You want to have the best-in-class providers. You also want to ensure that the health plan knows how to take care of its member populations and keep its member premiums affordable, like everything in life, a good balance of both. So, that’s the first piece of it.

The second part is about ensuring the members get the care they need from the provider network. There’s a component that involves holding the providers accountable for member outreach—for both chronic conditions and looking at areas that indicate the potential for developing a chronic condition.  

All of this ties back to the speed and timeliness of member outreach. If you wait too long, the member could become more ill. Or, if you don’t do anything, the member could go to another health plan because they’re not getting the care they need.      

The gap and chase process is multi-faceted, with prospective gaps and retrospective gaps. It’s all founded on the claims and clinical data that comes through health plan operations. Prospective gaps are risk adjustment operations that look at claims data for the current year and interpret the data so the health plan can make recommendations to the provider, as well as their member outreach campaigns. If they’re discharged patients, the recommendation might be to make sure they get an in-home follow-medical record from the claims and clinical data and work with the providers to make sure that member outreach is solid.  

This is why it’s essential to accelerate the speed and time for gaps to identify them faster and proactively. And we have to adopt tools and technology to get it done. The human eye cannot compete with things like natural language processing and machine learning algorithms. These technology tools and capabilities, coupled with robust software logic, can look at millions of lines of claims data (instantly) and indicate to providers what the next best action is per member. For this member, follow up with an A1C. For that member to follow up with an in-home assessment. Another member might just do a PCP follow-up. So the quicker you can do that, the faster you can close out the member gap. The quicker the health plan can get reimbursed from CMS, the more efficient the health plan becomes. This leads to CMS Star ratings and risk score accuracy, which leads to controlling member premiums so health plans can be competitive and access more robust provider networks. The provider doesn’t have time to dig through medical charts. They’re looking for health plans to follow up and provide the providers with intelligent automated solutions.

The other aspect is retrospective gaps. Now, retrospective gaps are also important because they’re dealing with prior year services that are already rendered to ensure that no gaps were missed. This is where predictive analytics comes in for the gap and chase program. You want to look at both current year, prospective, and past year, retrospective, together to say what have we missed? And let’s provide the provider with a member 360 to recommend the next best action. For example, last year, X was missed, and this year, I would look out for XYZ because this member is trending towards chronic conditions. This is why gap and chase is so foundational to the success of improving and caring for member populations—within Medicare Advantage—and controlling member premiums and having access to a robust provider network.  

Host: Thanks Abe, that really sheds some light on why timeframes are important. What unique challenges do you think will occur in 2022 in the risk adjustment space?

Abe: Many providers are returning to providing the normal care and services that members need because we’re getting back to normalcy. In 2020 and 2021, due to COVID, the seniors in the Medicare Advantage population weren’t getting normal care and procedures done. So what we’re going to see in 2022 and 2023 is an influx of medical claims because people who have postponed their care or were bumped due to COVID are all going to be playing catch up. We’re going to see a massive spike in provider visits and procedures that were delayed. This will result in a lot of clinical and claims data that simply wasn’t available in 2020 and 2021.

Host: Well, having more data is good. What do you think this will do for health plans and risk adjustment operations? And how do you think health plans will handle this huge influx in data?

Abe:  If they have access to automated tools and technologies coupled with a robust data strategy to be able to intelligently parse through clinical and claims data, they will be in a good position to accurately predict member gaps and create strong member segmentation models. This will enable them to provide directives to providers on that next best action.  

In the past we had medical claims and coders parsing through the claims manually. But, you’re not going to be able to do that when you have 10-15 times the number of claims coming in. The human eye and the human being can only do so much. This is why in 22 and 23, CMS is going to enforce health plan reimbursement and risk score accuracy, and they’re going to rely on health plan’s risk adjustment operations to leverage technology to identify these gaps and make sure they’re following up with their providers appropriately. At the end of the day, health plans can’t just keep adding bodies. They have to leverage tools and technologies to automate and work with CMS to get reimbursements and control member premiums. In the past, human beings could do it. But these post-COVID times are going to strain systems that aren’t automated.  

It’s essential for plans to leverage key technologies and tools—such as natural language processing for medical coding. It allows medical coders to pre-identify conditions more effectively and then route that intelligence to the providers through a sub-automated mechanism so the provider can confidently take the next best action. These are just some of things that are going to be trending over the next few years.  

Host: Looking back over the past few years, with COVID, how did health plans navigate risk score accuracy in the absence of routine care, and therefore the absence of data?  

Abe: In the true COVID year, which was 2020, a lot of health plans couldn’t accurately predict member health and gaps because the data just wasn’t available. What a lot of health plans had to do was increase member premiums because they weren’t getting reimbursed by CMS. Health plans can’t look at historical data beyond the current and prior year. This impacted member premiums and Star ratings for that year. It was a really tough time for health plans. The health plans that got through it kept close provider relationships and offered a lot of different incentives to providers to keep in touch with member health. Some strategies that worked were in-home member visits, and for the plans that were more tech-savvy, they amped up telehealth solutions and partnered with vendors for in-home assessments so they could make up for the lack of available data. We will start to see a more balanced shift in 2022 and 2023.  

Host: What strategies do you think plans be using to reduce gap timeframes in the coming years?  

Abe: The first way to reduce the gap and chase time frame is through natural language processing. In risk adjustment operations, prospective and retrospective gaps, everything comes from the claims and clinical data that is available. Explanation of Benefits, Explanation of Payments for service that have already been rendered which comes in the form of adjudicated claim services rendered and their associated (HCC) Hierarchical condition category and associated chronic condition codes. All of this is essential.  

In health plan technology solutions, they’ll run a gap and chase analytic engine that looks through all the claims and produces a chase list. The chase list says that for each member in the health plan’s population, we are seeing XYZ gaps. The chase list then gets fed to the medical coders to confirm or deny those gaps by doing a medical record retrieval and associated charts, which document and tie out medical data which acknowledges that these gaps exist. However, they can also identify other types of conditions or gaps that could arise if member care is left unchecked. All of this is manual. If we want to reduce this timeframe, we have to introduce technology and solutions like NLP and Machine Learning. NLP and machine learning have some key algorithms built based on artificial intelligence data insights. For example, if we know that the member segment in the state of Florida has a high tendency for diabetes—after the gaps engine is run and before the chase list is produced, machine learning will proactively identify gaps based on algorithms that are set to review millions of lines of claims details—in seconds, automatically finding the right, specific chronic conditions, and associated potential gaps in care for each member. Once this is run, the chase list is produced automatically and submitted for medical record retrieval for review and processing. These medical records and associated charts are run through the NLP engine to further identify, and accurately link to the original claim line detail—key indicators with pre-identified gaps that are both foundational and predictive to determine the next best action for this member. Then, the medical coders will review and approve for downstream, extremely accurate, and timely submissions to both the provider network—via a CDI alert—and eventually CMS. This all occurs automatically as the gap is closed and confirmed by the provider. This closes the timeframe for the entire end-to-end process for risk adjustment. Things that could take 4-6 weeks to identify can now be identified and completed within 24-48 hours, and the throughput of concurrent processing increases significantly. It’s more timely and efficient. Rather than parsing through files, PDFs, and charts, it’s one way to speed up the gap and chase process.  

The other way to speed this up is through electronic medical record retrieval. CMS requires evidence of the gap identified before they issue payment. They need documentation that the health plan has correctly identified gaps based on current care and the prior year’s care. Electronic medical record retrieval is something that will be foundational in the next couple of years. Because today, in order for the medical record coder to attach evidence and proof, they have to sift through the claims data that the provider needs to follow up on. The evidence is the medical record. So the medical record has to be attached to the chart that the medical coder codes that says, "This claim ties to these conditions, and these are the gaps that the provider needs to follow up on.”  

How do you get a medical record that resides in the hospital or the provider’s office? Well, that’s via fax, hand-written notes, or it’s in their system. The way it used to work, or sometimes still works, is via a phone call or a field technician goes out to the provider office or hospital, and then, they need to fax and make copies of the system records for these members to close gaps so the health plan can get reimbursed. Now, there is an automated way to directly connect to the provider’s electronic databases to grab the electronic medical record in a secure way, in real time. This allows the gap to be closed in real time within 24-48 hours. So you have the gap identification in 24-48 hours. And then, the electronic medical record which is evidence for CMS, is produced within another 24-48 hours. So, what used to take weeks is now done in a couple of days. At the end of the month, CMS requires submission of evidence to issue payment on a monthly basis for the health plans to get reimbursed. So the quicker you can do that, A—it benefits the member because they’re getting the care they need quicker and more efficiently, which improves or maintains their health, and B—Health plans are getting paid faster, and member populations are being taken care of better and faster.  

It’s important to bring a human element back into the process. At the end of the day, what’s important for a health plan? In my opinion, the most important thing is the provider network, taking care of our members and keeping member premiums competitive and affordable. Because in the Medicare Advantage space you’re dealing with seniors. They’re retired, and they’ve worked really hard to get there. And they shouldn’t have to worry about high medical expenses and paying high member premiums to get the care they need. As you get older, you need more care. The fact of the matter is—keeping the member premiums affordable and ensuring the health plan gets reimbursed by CMS translates to making sure the senior population is taken care of. It’s a win-win situation to speed up the gap and chase process from both the member health perspective and the health plan-provider-hospital, and operational efficiency perspective.  

Host: Abe thank you so much for being here today. I’ve enjoyed talking to you and I know our listeners have a lot to gain from what you’ve shared today. Thanks to everyone joining in today. Don’t forget to subscribe to the podcast. If you like this episode, share it on LinkedIn with your colleagues.