Automated accounts receivable management system with a self learning engine driven by current data
View Patent ↗An automated system is disclosed for managing accounts receivable for outstanding healthcare accounts. In particular, the system in accordance with the present invention is an automated system which preemptively assesses the risk of denials of outstanding healthcare accounts by way of a self-learning engine driven by current data and prioritizes those accounts for follow up according to the risk profile of the account. As such, the system is able to minimize denials by the respective payers thereby improving optimizing the efforts of healthcare accounts receivable personnel while improving the revenue yield and minimizing the revenue cycle.
1. An automated system for managing healthcare patient accounts comprising:
a computer system including a computer which is programmed with a self learning engine, said computer system further programmed to perform the following steps:
(a) receive patient account input data including type of service; denial history for similar claims; and payer for each patient account;
(b) based upon the patient account input data, the computer system is further programmed to automatically generate probabilities of claim denials defining risk levels for patient accounts based upon the history of denials for similar claims, the balance of the patient account and the age of the patient account; and
(c) based upon the risk levels, the computer system is programmed to automatically analyze patient accounts as a function of their respective risk levels and prioritize those analyzed accounts as a function of said risk levels so that patient accounts with the lowest risk level of denial are assigned the highest priority and patient accounts with the highest risk level of denial are assigned the lowest priority, said priorities also assigned as a function of a combination of the balance of the patient account and the age of the patient account; and
(d) said computer system further programmed to prioritize said patient accounts in a work list for follow-up according to their respective risk of denial in order to improve the revenue cycle.
2. The automated system as recited in claim 1 , further comprising:
said computer system further programmed to follow-up of on said patient accounts as a function of the risk level of said patient account and the level of skill of the back office personnel.
3. The automated system as recited in claim 1 , further comprising:
said computer system further programmed to automatically generating probabilities of claim denials defining risk levels for patient accounts based upon payment behavior of a particular payor learned as a function of a history of denials of similar claims from said particular payor.
4. The automated system as recited in claim 1 further comprising:
said computer system further programmed to automatically generating probabilities of claim denials defining risk levels for patient accounts based upon attributes of said input data that are expected to result in a write-off based upon previous denials of similar claims with one or more common attributes, said risk levels learned as a function of the claim denials and attributes of previous patient accounts.
5. The automated system as recited in claim 1 further comprising:
said computer system further programmed to automatically generating probabilities of claim denials defining risk levels for patient accounts based upon attributes of said input data that are expected to result in an adjustment based upon previous denials of similar claims with one or more common attributes, said risk levels learned as a function of the claim denials and attributes of previous patient accounts.
6. The automated system as recited in claim 1 further comprising:
said computer system further programmed to automatically generating probabilities of claim denials defining risk levels for patient accounts based upon attributes of said input data that are expected to result in a predetermined outcome based upon previous denials of similar claims with one or more common attributes, said previous patient accounts with attributes resulting in said predetermined outcomes defining rules.
7. The automated system as recited in claim 6 further comprising:
said computer system further programmed to continuously correlate said rules to said predetermined outcome.
8. The automated system as recited in claim 7 further comprising:
said computer system further programmed to determine the degree of correlation of said rules to said predetermined outcome.
9. The automated system as recited in claim 8 further comprising:
said computer system further programmed to applying said rule to said input data to predict the probability of claim denial only when said degree of correlation is greater than a predetermined threshold.
10. The automated system as recited in claim 1 further comprising:
said computer system further programmed to receive patient account input data including type of service;
denial history for similar claims; and payer for each patient account before a claim is submitted to a payer.
11. The automated system as recited in claim 1 , further comprising:
said computer system further programmed to schedule follow-up of third party insurance claims based upon approval of a Medicare claim.
12. The automated system as recited in claim 1 , further comprising:
said computer system further programmed to mine patient data to determine if the patient has co-insurance; and
add the claim to the work list for follow-up on any unpaid portion of a claim.
13. The automated system as recited in claim 1 , further comprising:
said computer system further programmed so that said self-learning engine continuously analyzes the input data to develop rules in order to predict the behavior of the patient accounts based upon learned attributes of the input data.
14. The automated system as recited in claim 13 , further comprising:
said computer system further programmed so that said self-learning engine can predict the behavior of the patient accounts based upon the input data and categorize the rules into one or more of the following categories: “Has Denials”; “Has Write-off”; “Has Write-off or Denial”; or “High Adjustment”.