AI That Pays: Lessons from Revenue Cycle — Nathan Wan, Ensemble Health

The Overlooked Financial Side of Healthcare 00:19

  • Nearly 40% of hospitals operate at a negative margin, mainly due to broken and manual revenue cycle processes rather than clinical costs.
  • Revenue cycle management (RCM) encompasses the entire financial process throughout the patient's journey in the healthcare system.
  • Administrative costs and complexity have surged over decades, outpacing growth in clinical staff, with a 30-fold increase in healthcare administration roles versus only a doubling of clinicians.
  • The majority of healthcare administrative costs stem from inefficient communication and friction among payers, providers, and patients.

Breakdown of Revenue Cycle Challenges 06:45

  • The "friction" in RCM contributes substantially to healthcare costs, particularly around delayed or denied insurance claims.
  • Denials are time-consuming and costly for hospitals, which operate on slim margins; denials often result from technical rather than medical issues.
  • A claim example: one provider had to appeal four times and waited 200 days for payment after a procedure.
  • Both providers and payers are using AI to improve their processes, with payers also leveraging AI to increase denial rates and adjudication volume, intensifying provider burdens.

AI Applications and Solutions at Ensemble 09:46

  • Ensemble Health, as an end-to-end RCM provider, leverages its comprehensive data access to identify and fix errors across the entire process, aiming to prevent issues rather than merely react to them.
  • Prior authorization is a significant pain point, often hindered by unclear or shifting requirements from payers—leading to unnecessary denials.
  • AI is being used to predict and flag likely denials in advance and to automate the gathering and submission of documentation for prior authorizations.
  • For clinical denials (when payers and providers disagree on medical necessity), AI assists in rapidly generating well-supported appeal letters by analyzing voluminous and varied medical records, guidelines, and policies.

AI Effectiveness and Measured Outcomes 14:52

  • Ensemble's AI initiatives have led to a 40% or greater reduction in the time required to process appeals, with higher quality as judged by increased overturn rates for denials.
  • The ROI of AI solutions is directly measured, tracking quantifiable impact on revenue cycle operations.
  • AI alone does not resolve all challenges; the field remains complex, inconsistent, and relies on data from disparate systems.

Infrastructure, Coordination, and Future Vision 16:16

  • Ensemble has built unified data infrastructure (EIQ platform) to integrate data from various formats and systems, facilitating more effective AI interventions.
  • The company continues to develop AI agents for all RCM aspects but recognizes that automation must be paired with reasoning and upstream error prevention.
  • The ultimate aim is to build smarter, more coordinated systems that reduce waste and enable healthcare resources to shift from administrative friction to clinical care.
  • Ensemble's unique position—comprehensive data, expert teams, and operational scope—supports leading this transformation in healthcare administration.