Date Posted: Tuesday, March 19, 2024

 

A new year brings excitement, fresh starts, and the latest changes to the CMS-HCC (Centers for Medicare and Medicaid Services hierarchical condition category) risk adjustment model, version 28 (V28). And even if you're excited about 2024, the guideline updates are probably stressful and a bit deflating. It can take months for coding and revenue cycle teams to get up to speed with the latest updates, and the training and accompanying expenses – as well as the inevitable mistakes that come with any coding update – cause billing delays, underpayments, and claim denials. Most healthcare organizations likely still have last year's evaluation and management changes for emergency departments fresh on their minds. Will you experience the same struggles again with CMS-HCC-V28?

 

2023 brought about the mass adoption of AI. But even though AI has only recently moved into the mainstream, the technology has been used in healthcare for decades. According to Cedars-Sinai Medical Center in California, AI medical applications were first explored as far back as the 1970s. Today, the technology is used in everything from diagnosing patients to transcribing medical documents and even dealing with guideline updates. There are five ways AI can make transitioning to CMS-HCC-V28 more efficient. Before we get to those, let's look at some of the changes ahead.

 

What Are the Changes in CMS-HCC-V28?

 

The overall goal of the CMS-HCC changes is simple: Capture more accurate and complete data on chronically ill patients. Unfortunately, these changes are quite complex in practice. They impact how HCCs map, coefficient HCC values, and HCC code names and numbers. Diagnosis codes are also affected: The changes removed 2,294 codes that map to an HCC payment and added 268 previously unmapped ones.

 

Factor in specific updates to various condition categories, including heart disease, blood disease, and diabetes, among others, and you have an overwhelming number of changes that could slow down your revenue cycle team for months. Thankfully, AI can help.

 

Handle Guideline Changes With AI and Save Time, Mistakes, and Money

 

It might sound too good to be true, but AI can dramatically improve your transition to version 28 by increasing efficiency and guaranteeing that you receive proper reimbursement under the new model. Here are five ways it does just that.

 

1. Updates with the press of a button: What if implementing coding guideline changes was like updating your computer? All you had to do was click a few buttons, drink a hot cup of coffee, and wait for a few short hours for the changes to apply. The speed at which AI can implement guideline updates varies depending on the complexity of the changes and the technology itself. Still, the process is similar to updating a system's configuration, which means it's substantially easier, faster, and cheaper.

 

2. Trains and collaborates with coders: Just because AI can accelerate the transition to new guidelines doesn't mean you can eliminate human coders or forgo training. In fact, AI can play a critical role in training your staff and speeding up the process, thus reducing costs. How does it work? While your team codes, AI provides real-time guidance and feedback based on the new update rules. And the system works both ways: coders can give feedback to the AI, so it learns and increases accuracy.

 

3. Flags insufficient documentation: AI plays a crucial role in improving documentation accuracy, which is of vital importance when adjusting to the latest guidelines. If a diagnosis lacks enough supporting documentation, AI points out this error and acts as an early warning system to clinical teams. This forewarning allows staff to rectify and augment the documentation promptly. When healthcare organizations can catch a problem so early on, it prevents a whole flood of negative consequences further downstream in the revenue cycle, such as denied claims, delayed reimbursement, and underbilling. Avoiding this pitfall saves time, headaches, money, and ensures proper reimbursement from the start.

 

4. Comprehensively captures diagnoses: AI's ability to thoroughly review a patient's medical records means it excels at identifying nuanced diagnoses. Consider an encounter where a patient is primarily diagnosed with hypertension. A comprehensive AI analysis may uncover additional conditions assessed, such as peripheral artery disease or hypertensive heart disease, contributing to a more detailed patient profile. This detailed assessment permits the AI system to capture HCCs for accurate billing. Overall, the comprehensiveness and specificity of AI coding help to ensure that risk adjustment factors (RAFs) are determined accurately, enabling proper reimbursement.

 

5. Implements proper combination coding with ease: Comorbidities can be easily overlooked in a patient encounter. For example, if a patient who has chronic kidney disease comes in for an insulin refill, a physician may review his or her bloodwork and eGFR but not specifically account for the CKD with a code. Because the kidney disease is being monitored during the encounter, a combination code such as E11.22 can be used that captures both diabetes and CKD, leading to a higher-valued HCC. This is where AI can help. It will rigorously review the patient's medical records and spot this oversight, securing proper reimbursement.

 

Don't Let Guideline Changes Slow Down Your 2024

 

Coding guideline changes take a lot of work and can drag any healthcare practice down and hurt staff morale.

 

Working with AI can remove that burden and give healthcare organizations a head start. The reduced workload empowers staff to devote themselves to high-value tasks and progress on vital initiatives, instead of being bogged down with complex coding updates.

 

While this vision may sound hard to believe, it's already happening. AI is changing the medical coding game, not just for CMS-HCC-V28, but for all guideline updates in the years ahead. After reading about the five concrete ways AI can help healthcare organizations transition to new guidelines, are you ready to see what it can do for you?

 

Taylor (Ross) Webster is the Head of Coding Quality at Fathom, a health technology company that uses deep learning artificial intelligence to automate medical coding for a wide range of specialties and practices. Webster manages strategic analysis, client analytics, and reporting. www.fathomhealth.com