Cerebral is a virtual mental health provider with a mission to democratize access to quality care for all. With several full monthly subscription plans to choose from, members get online access to therapists and prescribing physicians.
Visits are conducted via video chat, and members can message directly with their care team. Members can have their medications delivered directly to their door.
PROBLEM
Unfortunately, Cerebral’s chart audit process was entirely manual, time-consuming, and steep.
Manual internal audit and monitoring processes required hundreds of hours per month from a team of clinical auditors responsible for individually identifying audit notes. To do this, they would have to go through a process of creating a selected sample of clinicians to review, issues or topics to review for compliance or adherence, and ensuring that previous audit findings were not repeated in the environment.
“To obtain an adequate, curated sample, manual monthly data creation involved reviewing the calendars and appointments of hundreds of physicians to identify and copy medical record links into the EHR to locate the actual session notes for review,” said Julia Dolan, associate director of behavioral operations at Cerebral.
“Then, the results and observations were separately recorded in another document, analyzed and summarized in a report for management and individual physicians, and finally synthesized separately for senior management and the company’s executive management,” she added.
In addition to being time-consuming, the process was also prone to human error. Successful completion and evaluation of supplier audit results depended on the auditor’s detailed attention to detail.
“They had to enter the physician’s name and medical record number into a scoring template used to document the review,” Dolan explained. “That naturally created typical audit hazards, such as name mismatches, incorrect links, or omissions to apply an audit component to the chart.
“While we remained committed to continuous improvement and iteration to ensure high-quality audit feedback, we identified gaps in the manual process and remained diligent and proactive with the resources available – this was not an insignificant task for clinical leadership,” she added.
APPLICATION
Supplier Brellium has launched AI-powered chart viewing technology aimed at simplifying management of unstructured clinical notes by using artificial intelligence to conduct comprehensive audits in real time. This technology aimed to solve a major problem in healthcare administration: inefficiencies and errors in clinical documentation caused by manual processing.
“The core functionality of the system was to pre-screen all unstructured clinical notes for various types of errors and inconsistencies that could potentially impact the quality of patient care and/or insurance billing, such as client sentiment, successes and successors, evidence-based interventions, and whether treatment goals were measurable and time-bound,” Dolan explained.
“Specifically, the AI has been trained to identify clinical errors and spot trends in notes that may require attention,” she continued. “This proactive analysis is also designed to ensure that billing and/or coding errors can be identified and corrected before an insurance claim is submitted, significantly reducing the risk of claim denials and the need for costly rework.”
What’s more, by automating the review process, the technology promised to reduce the time healthcare providers spend on manual auditing by 98%, she added. While customary methods typically only allowed for a diminutive portion of patient visits to be thoroughly reviewed, this AI-powered approach allowed Cerebral to expand the percentage of notes it reviewed, increasing the scope of reviews while also enabling a drastic reduction in the labor-intensive task of manual screening—with a 78% reduction in time, she said.
“These significant efficiency improvements have freed up significant resources, allowing healthcare providers to focus their time and efforts on more customer-centric activities and other administrative areas,” she noted.
MET THE CHALLENGE
In December 2023, Cerebral began using an automated chart review process for its behavioral health department. Faced with the challenge of processing thousands of therapy notes each month, Cerebral needed a scalable system to meet internal quality control and compliance requirements, as well as insurance requirements—without the unsustainable burden of manual reviews.
“Our clinical team worked directly with the supplier to deliver “a customized AI system that integrates Cerebral’s unique session note formats, service times, and insurance partner specific audit requirements,” Dolan explained.
“By creating customized question sets and parameters for each note template type, the Brellium system automatically searches the session notes in Cerebral’s EHR and highlights necessary changes, significantly reducing resources spent on administrative audits and freeing up time for patient care, while also reducing the risk of financial loss from insurer recovery,” she said.
RESULTS
Within six months of implementing AI technology, Cerebral saw a significant 81% improvement in meeting key underwriting criteria and reduced the time spent on chart audits by 78%, demonstrating significant progress in operational efficiency.
Additionally, between March and June 2024, Cerebral saw its average physician audit scores improve by 8%.
ADVICE FOR OTHERS
“AI-powered chart auditing has allowed us to expand our documentation audit in an objective and operationally efficient manner,” Dolan said. “It has enabled Cerebral to more effectively review trends and error patterns with precise feedback that might otherwise be missed during a manual audit.
“This has allowed us to more easily identify areas for improvement and more effectively implement additional training across our physician teams in a meaningful and effective manner, contributing to continuous improvement,” she added.
She concluded: AI-assisted chart auditing significantly increased the efficiency of work processes at Cerebral and improved the quality of note documentation.