In recent years, payers have increasingly turned to artificial intelligence (AI) as a tool to streamline claims processing, accelerating the deployment of rules related to prior authorizations and medical-necessity assessments. Providers believe this has been a key driver behind a corresponding surge in claims denials, as well funded payers leverage technology to rapidly render these determinations and send them back to hospitals.
This trend has providers looking to respond with a tried-and-true strategy: Time to beat ’em at their own game.
Industry consolidation has given larger payers more capital to invest in payment-integrity technologies, including advanced algorithms and rules-based systems to systematically deny claims. Payers justify this approach as a response to “bad actor” providers who they believe order unnecessary tests and procedures to inflate revenue.
The growing revenue-integrity mindset among payers has led to heightened scrutiny and denials, with payers now harnessing AI to identify patterns of potentially unnecessary care. For hospitals and other healthcare providers, AI has transformed from a mere “nice to have” capability to a strategic imperative for revenue cycle management (RCM).
Waste and inefficiency in the broader healthcare system represent significant costs, and financial viability is vital for providers to effectively serve their communities. By utilizing AI to drive waste reduction, maximize revenue, and improve RCM efficiency, hospitals and healthcare systems can recover payments owed to them and strengthen their financial footing.
The technology can streamline claims processing, improve coding accuracy, and extract critical data from medical records and payer contracts to pinpoint the root causes of denials. AI can assist with a wide range of RCM functions, from front-end tasks like scheduling optimization and prior-authorization automation, to mid-cycle activities like clinical documentation and coding, to back-end duties like denials modeling and appeals automation.
With the largest payers outspending the largest health systems by a factor of 10 on technology investments, providers cannot compete on their own—they must leverage AI and machine learning (ML) to catch up with payers. However, integrating AI and automation into RCM workflows poses technical, data, talent, and operational challenges for many healthcare providers. To ensure a successful AI strategy, organizations must have these four key cornerstones in place:
The power of use cases
Success with AI in revenue cycle management hinges on a solid understanding and mastery of complex and dynamic use cases. Some examples:
- Automated coding and billing
- Predicting claims denial
- Forecasting and optimization
Hospitals have the potential to realize a strong return on their AI investment from drafting appeals letters, a time-consuming process that typically necessitates the involvement of several high-cost specialized staffers. Hospitals can train the models on clinical documentation, clinical care guidelines, payer policies, and managed care agreements to produce detailed letters citing clinical evidence.
Another strong use case is training machine learning (ML) models on historical data to look for the likelihood of succeeding in appealing denied claims and improving workflows. This approach dovetails with hospitals’ ongoing labor shortage by freeing up RCM staff to focus on higher-value tasks.
The power of data
With the digitization of healthcare over the last 15 years, there’s been a massive increase in data generation within the industry. Having access to robust data is critical to create reliable AI solutions that can make accurate predictions, mitigate biases, and improve over time. Examples include claim-level data, clinical documentation, payer policies and guidelines, and metadata.
Only then can AI be used to submit claims properly, identify the right arguments to make in an appeal, and automate the task of appealing low-balance denials so skilled analysts can focus on more appropriate, complex claims.
The power of an advanced platform
An optimal platform provides the tools, infrastructure, and capabilities to develop, deploy, and manage AI solutions effectively. Purpose-built platforms adept at digesting unstructured documents play a pivotal role.
The underlying data infrastructure must be capable of routing the right operational, claim-level, clinical, and contract data to the large language models (LLMs)—and then route the results into operational workflows. Getting RCM teams to adopt these results is yet another challenge, requiring change management, training, and a product-driven approach to implement these models and ensure they create value for the business.
The power of effective talent
Developing, applying, and managing AI in the revenue cycle depends on highly skilled talent to ensure success.
Building AI models depends on hard-to-find data scientists with sophisticated talent who can effectively harness data and identify the right use cases for machine learning. Skilled talent comes from an array of areas, including:
- Data scientists and engineers
- Product managers
- RCM operations subject-matter experts
- Learning and development specialists
Integrating these four key pillars is akin to optimizing a powerful engine for peak performance. The technology platform serves as the engine itself—the core machinery driving the RCM process. The data is the essential fuel for the engine, providing the raw ingredients the system needs to operate efficiently. The use case represents the race the organization is trying to win, demanding the right combination of engine power and fuel supply. And ultimately, the skilled RCM professionals and data scientists—the talent—act as the expert drivers, navigating the race course and extracting maximum performance from the engine-fuel system.
The way forward
The reality is that hospitals often lag behind payers when it comes to the technological and financial resources required to successfully implement AI-powered revenue cycle management. However, that doesn’t mean providers are without options. One effective strategy is to seek out collaborative opportunities across the industry. By partnering with RCM vendors that have already built robust AI/ML models using large healthcare-specific data sets—and crucially, adhere to the highest standards of information security—providers can accelerate their own AI adoption. Additionally, tapping into industry groups like state hospital associations to establish multi-organization pilot projects is another smart tactic. This allows providers to pool resources and scale AI initiatives more efficiently.
While payers may have a technological and financial head start, providers can still find creative ways to harness the power of AI to enhance their RCM performance. But it’s about more than just building an impressive engine. Simply having the most advanced technology platform doesn’t guarantee success.
Providers must ensure they have the right data to power the engine, the optimal use cases mapped out, and the talented workforce capable of pushing the system to its full potential. Only then can healthcare providers truly accelerate their RCM efforts and outpace the competition.
About Jim Bohnsack
Jim Bohnsack is Chief Strategy Officer at Aspirion where he is responsible for leading the development and execution of the Company’s long-term growth plan including oversight for strategy, mergers and acquisitions (M&A), and corporate development.