Healthcare is in the midst of an AI revolution, and that’s a good thing. At a time of rising hospital costs, labor shortages, and operating margins well below historical levels, revenue cycle executives need to evaluate their denials management strategy through the lens of artificial intelligence.
The steady increase in insurance claims denials has become one of the most significant financial trends affecting hospital revenue cycle performance. Recent analyses from Kodiak Solutions, formerly Crowe, revealed initial claims denials have jumped 11.9 percent year-over-year. Drawing on data from over 1,800 hospitals, the research highlights the mounting challenges, especially in aged accounts receivable, notably for patients covered by Medicare and commercial plans.
A separate report by the American Hospital Association (AHA) leveraging financial data from more than 1,300 hospitals and health systems, found a 56 percent hike in Medicare Advantage denials between January 2022 and June 2023, reiterating the mounting significance of denials management to the overall financial viability of a provider.
Yet another report reveals that almost half of healthcare providers in the US have witnessed a surge in claim denials, with errors in patient access and registration emerging as the primary culprit. Once more, these results highlight the persistent hurdles providers encounter in securing reimbursement.
For hospitals still grappling with the financial aftermath of Covid-19, managing denied claims couldn’t be more dire. Ongoing labor shortages and supply chain disruptions persistently impact hospital margins, emphasizing the urgent need for a more proactive approach. To achieve this, adopting AI methodologies similar to those employed by major insurers in claims processing becomes a prudent step forward.
AI’s influence on claims denials
AI holds enormous potential to revolutionize the denials landscape, and its impact is already evident. Through the use of AI, healthcare providers can streamline claims processing, enhance coding accuracy, and extract essential information from medical records and payer contracts to tackle the root causes of denials.
While the expertise of seasoned revenue cycle professionals remains crucial for denials resolution and prevention, denials will still occur with higher frequency and complexity based on trends from recent years. The integration of AI and automated workflows empowers these professionals to operate with maximum efficiency and effectiveness. This empowerment extends to not only preventing denials in the first place but also successfully overturning them. Examples of the broad-spectrum use of AI in the revenue cycle Include the following.
- Selection models for prioritization: In the realm of AI, selection models are specifically designed to categorize or choose specific items or entities based on predefined criteria. These models, a subset of machine learning, automate the prioritization of accounts, ultimately helping to boost revenue yields.
- Workflow acceleration: AI can also streamline workflows and expedite response times by automating the extraction of critical information from various sources, such as claims, medical records, contracts, and guidelines. This extracted information is then applied to accelerate various tasks. For example, AI can generate draft appeal letters for a revenue cycle professional to subsequently review and edit, reducing manual labor, minimizing errors, and enhancing overall workflow efficiency resulting in improved revenue outcomes.
- Improved price accuracy: AI proves invaluable to Revenue Cycle Management (RCM) teams by harnessing the value hidden in unstructured data from sources like medical records and managed care contracts. It identifies discrepancies in reimbursement rates and ensures adherence to contractual terms. This proactive approach helps prevent denials and underpayments while equipping healthcare providers with essential data to guarantee accurate reimbursement per negotiated agreements.
- Predictive trend analysis: With denial rates on the rise and evolving coding policies, the process of submitting claims becomes increasingly complex, presenting challenges for under-resourced provider organizations. AI steps in by identifying and forecasting trends related to denials and payments. This foresight enables RCM teams to make necessary operational adjustments to avoid denials. Additionally, AI’s predictive capabilities allow RCM teams to allocate their limited resources more effectively by identifying which denials are more or less likely to be overturned when they do occur.
Deploying AI for denials management: Key considerations
When healthcare organizations incorporate AI and automation into RCM processes, they encounter a range of technical, data, talent, and operational considerations.
Creating effective AI models is a challenging task, necessitating a substantial amount of high-quality data to train and mature models. It also requires data scientists with specialized skills who can harness data effectively while identifying meaningful use cases for machine learning. Moreover, it demands robust underlying platforms and engineering capabilities for efficient model deployment and a deep understanding of evolving standards and regulations to ensure ethical and compliant usage. Collaborating with established RCM innovators in the industry can expedite a healthcare system’s path to stronger financial health.
Large Language Models (LLMs) or foundational models represent a groundbreaking innovation, offering immense potential and substantial uncertainties. Regulatory frameworks are catching up with this technology, but its rapid evolution introduces ongoing uncertainties. Collaborating with strategic partners skilled in this becomes a vital strategy. Notably, major players like Google, Microsoft, AWS, and other cloud providers are developing healthcare-specific LLM platforms, contributing to risk mitigation in this evolving landscape.
Furthermore, engaging with industry groups such as state hospital associations and collaborating with multiple healthcare organizations and technology companies for pilot projects are advisable approaches. Provider organizations should begin small, leveraging AI to address their most challenging issues, such as patient collections.
Interoperability is paramount, requiring a robust data infrastructure capable of efficiently routing claim-level, clinical, contract, and operational data to AI models and integrating the results into operational workflows. Implementing these models also entails change management, training, and a product-driven mindset to ensure they generate tangible business value.
As the healthcare industry grapples with various challenges, from workforce shortages to complex coding intricacies, embracing AI can empower revenue cycle professionals to envision a future with reduced denials, optimized revenue, and fortified financial health. The urgency is clear, as delays can result in potential revenue loss. The path ahead promises streamlined processes, improved coding accuracy, and proactive denials prevention, marking a transformative shift where AI becomes an indispensable asset in strengthening the financial revenue cycle structure for healthcare providers.
The utilization of AI by payers has already contributed to a rise in denial volumes, emphasizing the urgency for healthcare providers to promptly harness AI as a fundamental asset to enhance their revenue cycle.
Photo: Filograph, Getty Images
