Few receivables processes are more complex than health payments. With the highly variable and complex explanations of payment/benefit (EOP and EOB) documents trapping data required to reconcile and verify the payment, the effort is time-consuming and fraught with opportunities for error.
No two remittances are alike, even from the same payer meaning that valuable time is spent simply visually scanning the document to locate required data that then needs to be entered into a system.
And apart from data entry, there is also the cumbersome process of matching and then comparing identified remittance data with the original submitted claim. Some remittances require more than an hour to process.
Simple, Streamlined Reconciliation
Using advanced computer vision coupled with machine learning, Parascript software understands these complex health remittances and can even learn different payer types to quickly identify the required information.
Both header and service detail data are quickly identified and exported in a simple workflow that can run in a hybrid unattended mode for accurate data along with a guided assistive mode to present suspect information to a reviewer.
There is no need to search for data, it is presented allowing for a quick check or correction. Data can be automatically balanced against the check payment amount as well as the original matched claim meaning that payment errors are handled more efficiently and quickly while valid payments can be processed.
The result is a process that is more streamlined, modern, and accurate enabling true digitization of a once fully manual process.
Automated identification and separation of remittances within a stream of documents.
Automated reconciliation with matching CMS-1500 claim.
Machine learning-based data location and extraction of claim and line-level data:
- Claim Level: Member, HIC, Coinsurance, Bill Type, Covered, Deductible, Service Dates, Non-covered.
- Line Level: Multiple service line data that includes the above and procedure codes and modifiers, billed amount, RARC codes, Reason.
Software learns and optimizes on your data providing a “custom fit” without the typical costs.