Systems and methods for determination of patient true state for risk management
Systems and methods for managing audit risks utilizing the true state of the patient are provided. A number of medical records for a patient are subjected to predictive modeling for various conditions (known as patient ‘true state’). The true state is then cross referenced by the eligible Medicare documentation, and any findings that are being submitted to MediCare for reimbursement. The result of this cross referencing is the ability to classify each finding and/or true state into a “green, “yellow”, or “red zone”. The green zone is where the finding, documentation and true state are in good alignment. A red zone is where the finding and the true state are entirely at odds. The yellow zone is where the findings and the true state are in agreement, but where there is still audit risk that may be resolved through one or more “opportunities”. Examples of opportunities include bolstering the documentation for the reimbursement, getting additional evidence to improve the confidence of a true state inference, or including additional documentation for a finding that exists in the true state, but hasn't been previously identified.
1. In a health information management system, a method for managing audit risk, the method comprising:
receiving a plurality of medical records for a patient;
identifying terms by cross referencing language nodes in the plurality of medical records against a wiki-like database;
identifying medical concepts through machine learned relational clustering of the terms;
inferring a true state for the patient using a processor by applying a predictive model to the medical concepts, wherein the true state is a condition of the patient;
cross-referencing the inferred true state with at least one coder finding for the patient and MediCare eligible documentation;
classifying each coder finding into one of at least three confidence groups responsive to alignment between the true state with at least one coder finding for the patient;
generating a personalized care recommendation based on the patient's true state, care history and personal information; and
executing the personalized care recommendation.
2. The method of claim 1 further comprising generating a confidence level for the inferred true state.
3. The method of claim 2 wherein the at least three confidence groups include a green zone, a yellow zone and a red zone.
4. The method of claim 3 wherein the green zone is populated with findings that are aligned with the documentation and the patient true state.
5. The method of claim 3 wherein the red zone is populated with findings that are misaligned with the patient true state.
6. The method of claim 3 wherein the yellow zone is populated with findings that have an opportunity.
7. The method of claim 6 wherein the opportunity is at least one of where additional MediCare documentation is needed to reduce audit risk, a finding does not exist for the patient true state, and the true state has an intermediate confidence level.
8. The method of claim 2 further comprising validating the true state.
9. The method of claim 8 wherein the validation updates the predictive model and updates the confidence level for the inferred true state.
10. The method of claim 8 further comprising optimizing routing of the evidence for validation based upon greatest impact the evidence has on the true state inference.
11. A health information management system for managing audit risk comprising:
a records database configured to receive a plurality of medical records for a patient; and
a first pass analyzer including a processor configured to:
identify terms by cross referencing language nodes in the plurality of medical records against a wiki-like database;
identify medical concepts through machine learned relational clustering of the terms;
infer a true state for the patient by applying a predictive model to the medical concepts, wherein the true state is a condition of the patient;
cross-referencing the inferred true state with at least one coder finding for the patient and MediCare eligible documentation; and
classifying each coder finding into one of at least three confidence groups responsive to alignment between the true state with at least one coder finding for the patient;
generating a personalized care recommendation based on the patient's true state, care history and personal information; and
executing the personalized care recommendation.
12. The system of claim 11 wherein the first pass analyzer is further configured to generate a confidence level for the inferred true state.
13. The system of claim 12 wherein the at least three confidence groups include a green zone, a yellow zone and a red zone.
14. The system of claim 13 wherein the green zone is populated with findings that are aligned with the documentation and the patient true state.
15. The system of claim 13 wherein the red zone is populated with findings that are misaligned with the patient true state.
16. The system of claim 13 wherein the yellow zone is populated with findings that have an opportunity.
17. The system of claim 16 wherein the opportunity is at least one of where additional MediCare documentation is needed to reduce audit risk, a finding does not exist for the patient true state, and the true state has an intermediate confidence level.
18. The system of claim 12 further comprising a validation system configured to validate the true state.
19. The system of claim 18 wherein the validation system updates the predictive model and updates the confidence level for the inferred true state.
20. The system of claim 18 further comprising a routing optimizer configured to optimizing routing of the evidence for validation based upon greatest impact the evidence has on the true state inference.