How AI is Changing the Conversation Around Hospital Denials – and Improving Financial Implications in the Process

News
Article

In an early 2023 brief published by KFF, CMS transparency data for 2021 revealed an average denial rate of 17% for all in-network claims, with authorization denials reported as high as 24% of denial volume for some plans and medical necessity denial rates as high as 37% for others.

Of the many complexities within the healthcare industry, medical necessity determinations can be the most challenging – and pose one of the greatest barriers to efficient revenue cycle processes and appropriate reimbursement.

Complex medical necessity policies that differ between payer, diagnosis, and presentation have muddled utilization review (UR) and revenue cycle processes for years, driving waste and friction between providers and payers that has only intensified since the pandemic.

In an early 2023 brief published by KFF, CMS transparency data for 2021 revealed an average denial rate of 17% for all in-network claims, with authorization denials reported as high as 24% of denial volume for some plans and medical necessity denial rates as high as 37% for others.

When compounded by denial issues prior to service, at the time of service, and after claim submission, these subjective and inefficient processes have diverted valuable clinician attention away from patient care and into administrative functions – routinely identified as a critical driver in unsustainable, wasteful and rising health expenditures.

The impact is only magnified as provider and payer resources are forced into lengthy appeal processes over subjective and differing guidelines and misalignment on appropriate level of care.

As the healthcare industry seeks ways to make sustainable change that supports healthcare cost containment initiatives and relieves provider and clinician burnout, this is a barrier that must be addressed.

Solving for misalignment between payers and providers

When focusing on the hospital and acute care implications, CMS transparency data showed 25% of all hospital claim denials were related to medical necessity and accounted for 30% of all denial write-offs, resulting in significant losses for care already provided. With an estimated $33 billion in denial write-offs annually for hospital systems alone, that’s a conservative estimate of $10 billion in lost hospital reimbursement related to medical necessity for care already provided.

In addition to lost revenue, the cost of the denial and appeal processes adds an estimated $7.2 billion in administrative cost for hospital and health systems annually.

With increasing awareness of the waste in the system that needs to be eliminated, more organizations are exploring artificial intelligence (AI) and machine learning (ML) to better align resources and assist with complex decision making.

As providers and payers collaborate more on safe and effective uses for AI and ML, this is a perfect opportunity to realign UR processes back to medical standards of care rather than subjective criteria sets that are different for every patient scenario and inconsistent between payers and providers.

When leveraging AI and ML to synthesize and predict outcomes based on robust data, traditional UR and medical necessity determinations are transformed from a subjective, time-consuming activity to an objective, standardized one with real-time data insights into each patient’s clinical picture as it evolves and a clear view of potential mismatches in patient status.

This reduces ambiguity – removing duplicative and unnecessary reviews for both providers and payers, preventing avoidable denials and delays in determinations, and putting more clinicians and dollars back into improving patient outcomes. It also helps ensure appropriate reimbursement for care provided via a consistent, compliant framework that is aligned with evidence-based medicine and the standards of care, as opposed to reimbursement policies.

Addressing friction within healthcare organizations

Almost as complex as navigating the appropriate level of care between providers and payers, misalignment can exist within healthcare systems, as financial and clinical teams struggle to ensure appropriate reimbursement while maintaining efficient revenue cycle processes.

Unlike preservice denials during the prior authorization process, hospital denials related to medical necessity or level of care are typically received after patient care has been initiated or rendered, creating a significant financial burden for the provider. When not resolved prior to discharge, the misalignment between the provider and payer can carry over to backend revenue cycle teams who are focused on resolving claims as efficiently as possible – but aren’t involved in the clinical care of the patient, thus relying on claim-related data and denials received from the payer.

Because medical necessity denials involve complex decision making for the treating providers and the clinical staff supporting utilization reviews, they have become some of the most complex and contentious denials to avoid and resolve. This complexity not only adds potential delays in patient care on the front-end, but it also creates administrative waste at the point of care and delays in AR resolution and rework on the backend.

Based on financial data from US hospitals, an estimated $405 billion in provider payments were delayed more than 90 days in 2022, and an estimated $67 billion in payments were delayed more than six months, an increase of $34 billion from 2021. With clinical appeal time frames that can extend beyond two years, that associated financial impact to backend revenue cycle teams and the financial metrics they’re responsible for further complicates the friction, rework, and misalignment within healthcare organizations.

AI and ML can improve partnerships and handoffs between UR and back-end revenue cycle teams as they leverage the AI and ML to see appropriate status and expected reimbursement through the same objective lens for a consistent and effective approach to denial resolution and payer communication. Combined with a hospital’s denial and financial data from point of service to final outcome, the result is more robust reporting and analytics to support decision making among UR, revenue cycle, and financial management teams.

This end-to-end visibility allows for better root cause analysis, provides insight into gaps in processes, and facilitates better collaboration between teams — preventing denials further upstream while still ensuring the appropriate reimbursement for the care provided. By strengthening denial analytics with objectivity and visibility into inpatient medical necessity, providers can reduce risk, improve revenue assurance, and strengthen denial prevention and resolution.

Within the healthcare industry, everyone's goal is to improve patient outcomes and create a sustainable healthcare system for all communities. This is no easy task, and it can’t be accomplished on an individual institution basis. Instead, we must push for continuous forward momentum across the board with both providers and payers. A crucial step in the right direction is ensuring that the bulk of healthcare spend is going toward care and not administrative work – a cause that both entities can support in the face of inflating healthcare administration costs.

By employing AI and ML to support utilization review and medical necessity determinations, we have an opportunity to relieve administrative overhead and align providers and payers with a shared goal of providing better care and improving health outcomes.

Tanya Sanderson, RN, BSN, MBA, MHA, CLNC, is the Senior Director of Revenue Integrity at XSOLIS.


Recent Videos
1 KOL is featured in this series.
1 KOL is featured in this series.
1 KOL is featured in this series.
1 KOL is featured in this series.
Jennifer Seminerio, MD, a gastroenterologist
Jennifer Seminerio, MD, a gastroenterologist
Related Content
© 2024 MJH Life Sciences

All rights reserved.