Alert optimizer

• US 10,607,728 B2
• Filed: 10/06/2015
• Issued: 03/31/2020
• Est. Priority Date: 10/06/2015
• Status: Active Grant

• Current Assignee: Cerner Innovation, Inc. (Oracle Corporation)
• Original Assignee: Cerner Innovation, Inc. (Oracle Corporation)
• Inventors: Agassi, Natalee, Agassi, Emin, Kim, Jay, Haley, John
• Primary Examiner(s): Sereboff, Neal
• Application Number:
  US14/875,800
• Publication Number:
  US 20170098037A1
• Time in Patent Office: 1,638 Days
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: Y02A 90/10 Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Abstract

An alert optimizer subsystem for a HIT system modifies, reconciles, and/or prioritizes candidate clinical alerts. The optimized alerts may be filtered, prioritized, enriched, and/or formatted so that the alerts are more relevant and/or more actionable for a system user.

Claims

1. A method for preserving computer processing capacity in a Health Information Technology (HIT) system, the method comprising:

receiving one or more clinical orders from a clinician, the one or more clinical orders specific to a patient;

checking the one or more clinical orders against a set of clinical alert rules;

identifying two or more candidate clinical alerts related to the clinical orders and the set of clinical alert rules;

optimizing the two or more candidate clinical alerts to create an optimized clinical alert, the optimizing comprising;

accessing a patient-specific alert history and a patient-specific Electronic Health Record (EHR), the patient-specific alert history comprising a past clinical alert associated with the patient and an indication that the clinician overrode the past clinical alert, wherein the past clinical alert was overridden prior to receiving the one or more clinical orders from the clinician;

identifying a first clinical condition associated with the past clinical alert;

identifying a second clinical condition associated with a first candidate clinical alert of the two or more clinical candidate alerts;

determining that the first clinical condition associated with the past clinical alert is the same as the second clinical condition associated with the first candidate clinical alert; and

filtering out the first candidate clinical alert such that the first candidate clinical alert is not presented to the clinician on a user interface of a clinician device; and

presenting the optimized clinical alert, for display by the user interface of the clinician device, wherein the optimizing reduces processing capacity required to present the candidate clinical alerts.

2. The method of claim 1, wherein checking the clinical alert rules comprises:

identifying clinical information relevant to checking the clinical alert rules;

accessing the EHR associated with the HIT system; and

extracting the relevant clinical information from the EHR.

3. The method of claim 2, wherein the relevant clinical information comprises patient demographics, clinical test results, clinical observations, medical conditions, medical risk factors, recent clinical orders, patient medication history, active patient medication, patient home medication, patient allergies, patient diagnosis, patient genetics, prior patient clinical choices, patient clinical directives, or combinations thereof.

4. The method of claim 3, wherein optimizing the two or more clinical alerts comprises filtering out a second candidate clinical alert of the two or more candidate clinical alerts, wherein the second candidate clinical alert is inconsistent with the patient medication history, active patient medication, and/or patient home medication, such that the second candidate clinical alert is not presented.

5. The method of claim 3, wherein optimizing the two or more clinical alerts comprises filtering out a second candidate clinical alert of the two or more candidate clinical alerts, wherein the second candidate clinical alert is inconsistent with a patient care directive or prior patient clinical choice, such that the second candidate clinical alert is not presented.

6. The method of claim 1, further comprising assigning to each candidate clinical alert of the two or more candidate clinical alerts a level based on a likelihood of an undesired outcome and a severity of the undesired outcome.

7. The method of claim 6, wherein the level distinguishes at least between urgent, precautionary, and advisory clinical alerts.

8. The method of claim 7, wherein if a total number of optimized alerts exceeds a threshold number, advisory clinical alerts are not presented.

9. The method of claim 8, wherein if the total number of optimized alerts exceeds a second threshold number, precautionary clinical alerts are not presented.

10. The method of claim 1, further comprising generating alert text targeted to a variable audience.

11. The method of claim 10, wherein the variable audience is defined by a role of a system user.

12. The method of claim 11, further comprising formatting an alert targeted to the system user.

13. The method of claim 10, wherein the variable audience is defined by a role of a clinician who generated the clinical order underlying the optimized clinical alert.

14. The method of claim 13, further comprising formatting an alert targeted to the clinician.

15. The method of claim 1, further comprising:

accessing a prescribed actions list; and

filtering out any alert which has a prescribed resolution, such that the first candidate clinical alert is not presented.