Analyzing Electronic Medical Records to Reduce Insurance Denials
The Situation
A chain of residential rehabilitation centers providing treatment for emotional trauma, addiction, depression, and other disorders wanted to reduce the rate at which patients’ insurance claims were denied. To do so, they needed to identify clinician notes and records that had a higher chance of denial. Merilytics (an Accordion Company) partnered with the client to analyze Electronic Medical Records (EMR) in order to identify a correlation between EMR text and authorization denials.
Services
Reporting & Analytics
The Execution
- Integrated and consolidated data from different EMR systems such as KIPU, Easystep, and Sigmund.
- Levered initial hypotheses to identify key patterns in the claims with higher denial/approval rates (such as no medical justification, recurring phrases, the same set of clinicians, etc.).
- Predicted the denial probability of a claim using Artificial Neural Network Algorithms.
- Designed BI reporting infrastructure to monitor the quality of the EMR documentation.
- Created daily reports for the EMR team to provide access to the consolidated EMR data and provide visibility into the records/clinicians with a higher potential for a denial at a patient level.
The Results
As a result of the engagement, the client was able to identify the information needed to train clinicians to improve the quality of clinical documentation and reduce the denial rate by ~45% — which resulted in an acceleration of >$2M in collections and improved the predictability of cash flows. What’s more, the client is now able to take a corrective course of action on predicted potential denials on a near real-time basis.