One young businesswoman making a presentation speech.

Advanced AI streamlines prior authorization and saves  $200,000

Natural language processing and digitized clinical guidelines standardize procedure

Project Summary

Client: One of the largest integrated health care delivery and financing systems in America.

Challenge: Manual intervention in authorization process bottlenecked patient care.

Solution: Introduce natural language processing to automate and optimize authorization process.

Results: $200,000 in savings, 150 hours saved, more capacity for 30+ care managers

Challenge

In the intricate ecosystem of healthcare, prior authorizations carry a complicated legacy.

For certain complex, high-risk, or expensive medical treatments, healthcare providers require pre-approval from patients’ insurance companies. In a process called prior authorization, payors examine a patient’s clinical history to determine if they will cover a prescribed procedure or medication. Meticulous, detail-driven, and holistic, the prior authorization evaluates the medical necessity of prescribed care.

When prior authorizations are efficient and accurate, they benefit all stakeholders. For payors and providers, they ensure that medical recommendations are cost-effective – for patients, they certify that procedures and prescriptions are safe and necessary for wellness.

But when prior authorizations are inefficient or inaccurate, they stagnate the healthcare process. Deadlocking negotiations between payors and providers, delayed authorizations become insurmountable obstacles between patients and the critical care they need.

Lumevity’s client – one of the largest integrated health care delivery and financing systems in America – knew that it fell into the second category. To determine whether medical procedures met the criteria for authorization, the client’s care management team personally compared patients’ clinical data with medical policy guidelines.

The manual review of lengthy clinical histories and granular medical policies was maddeningly time-consuming, and it led to unnecessarily delayed authorization cycles.

Further, the authorization process was deeply subjective – individual care managers applied their own interpretations of insurance policies to patients’ medical histories. Combined with the tiresome monotony of the documents themselves, this subjectivity sometimes led to errors in authorization decisions, delaying the process even more. 

Tedious, subjective, and constantly risking natural human error, the client’s prior authorization procedure desperately needed an upgrade.  

Solution

We worked with the care management team to develop a natural language processing engine that quickly improved their authorization review procedure.

Natural language processing is a subfield of artificial intelligence. A combination of linguistics and computer science, it enables machines to understand, interact with, and communicate in human language.

Applied to the care management team, this innovative technology simplified, standardized, and optimized the client’s prior authorization process.

The engine first identifies the relevant medical policy for the requested procedure and reduces it to its most pertinent parameters. Then, it scans the patient’s medical history and clinical charts for data that matches the relevant policy. Finally, with all information gathered, the engine guides a human care manager through a decision tree to help determine whether the member qualifies for authorization.

Takeaway

This collaboration between the care manager team and Lumevity highlights what makes Lumevity so unique: We believe that integrated problems require integrated solutions. The client’s old, inefficient authorization process negatively impacted employee engagement, customer satisfaction, and revenue collection. By focusing on holistic transformation, Lumevity implemented a new system that simultaneously addressed all three areas of measurement. The natural language processing engine activated a more engaged community of employees, improved patient care, and drove significant financial impact – all at the same time.

$200,000 in savings

After two months of use, the increased efficiency of the natural language processing engine generated $200,000 in annualized savings.

Saved 150 hours

Removing the need for human care managers to manually authorize claims, the new process saved the team 150 hours of work – in just two months.

More capacity for care manager team

Freed from the drudgery of repetitive work, care managers had more capacity to operate at the top of their licenses, leading to more effective utilization management and better patient outcomes.

You may also like

Analytics governance model liberates resources and activates cross-department engagement

Streamlining analytic procedures and facilitating advanced analytics with organizational/operational modeling

Read More

How to put innovation to work for you

An inside look at a program that’s on track to save $1B in five...

Read More

Team-built automation reduces hours to seconds

Robotic desktop automation creates folders required for biweekly payroll processing

Read More