ANALYTIC MODELING OF PROTECTED HEALTH INFORMATION

• US 20150213458A1
• Filed: 01/28/2014
• Published: 07/30/2015
• Est. Priority Date: 01/28/2014
• Status: Active Grant

• Current Assignee: Solventum Intellectual Properties Company (Solventum Corp.)
• Original Assignee: 3M Innovative Properties Company (3M Company)
• Inventors: Wolniewicz, Richard H.
• Granted Patent: US 10,803,466 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06Q 30/018 Certifying business or products
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records

Abstract

This disclosure includes techniques for analyzing patient data. In one example, a method includes accessing, by a computer system, one or more databases comprising health information for a plurality of patients, wherein the health information includes protected health information, randomly selecting, by the computer system, a subset of the health information from the one or more databases, wherein the subset of the health information corresponds to a subset of the plurality of patients, removing, by the computer system, the protected health information from the subset of health information to produce a de-identified analytics subset of patient data suitable for analytical model construction and evaluation, and storing, by the computer system, the de-identified analytics subset of the patient data in the one or more databases.

Claims

1. A method for analyzing patient data, the method comprising:

accessing, by a computer system, one or more databases comprising health information for a plurality of patients, wherein the health information includes protected health information;

randomly selecting, by the computer system, a subset of the health information from the one or more databases, wherein the subset of the health information corresponds to a subset of the plurality of patients;

removing, by the computer system, the protected health information from the subset of health information to produce a de-identified analytics subset of patient data suitable for analytical model construction and evaluation; and

storing, by the computer system, the de-identified analytics subset of the patient data in the one or more databases.

2. The method of claim 1, further comprising:

presenting, by the computer system, the de-identified analytics subset of the patient data via a user interface for analytical model construction and evaluation.

3. The method of claim 1, further comprising:

receiving, by the computer system, an analytical model based on the de-identified analytics subset of patient data; and

applying, by the computer system, the analytical model to the health information with the protected health information for the plurality of patients.

4. The method of claim 3, further comprising outputting, by the computer system, a summary of the results of the application of the analytical model to the health information with the protected health information for the plurality of patients via a user interface restricting user access to the protected health information.

5. The method of claim 3, further comprising:

accessing, by the computer system, a result of the of the application of the analytical model to the de-identified analytics subset of patient data;

comparing, by the computer system, the result of the of the application of the analytical model to the de-identified analytics subset of patient data to a result of the of the application of the analytical model to the health information for the plurality of patients; and

presenting, by the computer system, a summary of the comparison of the result of the of the application of the analytical model to the de-identified analytics subset of patient data with the result of the of the application of the analytical model to the health information for the plurality of patients via a user interface restricting user access to the protected health information to facilitate evaluation of the suitability of the analytical model for the health information for the plurality of patients.

6. The method of claim 1, wherein removing, by the computer system, the protected health information from the health information for the randomly-selected subset of the plurality of patients to produce the de-identified analytics subset of patient data suitable for analytical model construction and evaluation comprises:

removing data from fields identified as including protected health information; and

filtering unstructured data to remove unstructured protected health information.

7. The method of claim 1, wherein producing the de-identified analytics subset of patient data suitable for analytical model construction and evaluation further comprises:

receiving, from a human user, indications of additional protected health information within the health information for the randomly-selected subset of the plurality of patients after removing, by the computer system, the protected health information from the health information; and

removing, by the computer system, the additional protected health information indicated by the human user from the health information for the randomly-selected subset of the plurality of patients.

8. The method of claim 1, further comprising:

receiving, from a human user, a request to access protected health information removed to produce the de-identified analytics subset of patient data;

presenting the requested protected health information to the human user; and

storing a record of the request to access the protected health information.

9. The method of claim 1, further comprising:

following an update to the one or more databases comprising health information for the plurality of patients, the update including health information for additional patients, randomly selecting, by the computer system, health information from the one or more databases of a subset of the additional patients;

removing, by the computer system, the protected health information from the health information for the randomly-selected subset of the additional patients;

adding the health information for the randomly-selected subset of the additional patients without the removed protected health information to the de-identified analytics subset of patient data to produce an updated de-identified analytics subset of patient data; and

storing, by the computer system, the updated de-identified analytics subset of the patient data in the one or more databases.

10. A computer system for storing and analyzing health information for a plurality of patients comprising:

one or more databases comprising health information for a plurality of patients, wherein the health information includes protected health information, and health information for a randomly-selected subset of the plurality of patients within the protected health information removed for the randomly-selected subset of the plurality of patients; and

a user interface that facilitates user access to health information for a randomly-selected subset of the plurality of patients within the protected health information removed for the randomly-selected subset of the plurality of patients for analytical model construction and evaluation.

11. The computer system of claim 10, further comprising one or more processors configured to apply an analytical model to the health information with the protected health information for the plurality of patients in the one or more databases.

12. The computer system of claim 11, wherein the computer system is configured to output a summary of the results of the application of the analytical model to the health information with the protected health information for the plurality of patients via the user interface restricting user access to the protected health information.

13. The computer system of claim 11,

wherein the one or more processors are configured to compare a result of the of the application of the analytical model to the de-identified analytics subset of patient data to a result of the of the application of the analytical model to the health information for the plurality of patients;

and

wherein the computer system is configured to output a summary of the comparison of the result of the of the application of the analytical model to the de-identified analytics subset of patient data with the result of the of the application of the analytical model to the health information for the plurality of patients via the user interface restricting user access to the protected health information.

14. The computer system of claim 10, further comprising one or more processors configured to remove the protected health information from the health information for the randomly-selected subset of the plurality of patients to produce the de-identified analytics subset of patient data suitable for analytical model construction and evaluation.

15. The computer system of claim 10, wherein the one or more processors are configured to remove the protected health information from the health information for the randomly-selected subset of the plurality of patients to produce the de-identified analytics subset of patient data according to steps including:

removing data from fields identified as including protected health information; and

filtering unstructured data to remove unstructured protected health information.

16. The computer system of claim 10, wherein the one or more processors are configured to remove the protected health information from the health information for the randomly-selected subset of the plurality of patients to produce the de-identified analytics subset of patient data according to steps including:

receiving, from a human user, indications of additional protected health information within the health information for the randomly-selected subset of the plurality of patients after removing, by the computer system, the protected health information from the health information; and

removing the additional protected health information indicated by the human user from the health information for the randomly-selected subset of the plurality of patients.

17. The computer system of claim 10, wherein upon receipt of a request to access protected health information from the randomly-selected subset of the plurality of patients removed to produce the de-identified analytics subset of patient data via the user interface, the computer system is configured to:

present the requested protected health information via the user interface; and

store a record of the request to access the protected health information in the one or more databases.

18. The computer system of claim 10, wherein following an update to the one or more databases comprising health information for the plurality of patients, the update including health information for additional patients, the computer system is configured to:

randomly select, from the one or more databases, health information of a subset of the additional patients;

remove the protected health information from the health information corresponding to the randomly-selected subset of the additional patients;

add the health information corresponding to the randomly-selected subset of the additional patients without the removed protected health information to the de-identified analytics subset of patient data to produce an updated de-identified analytics subset of patient data; and

store the updated de-identified analytics subset of patient data in the one or more databases.

19. A computer-readable storage medium comprising instructions that, when executed, cause one or more processors to:

access one or more databases comprising health information for a plurality of patients, wherein the health information includes protected health information;

randomly select health information from the one or more databases corresponding to a subset of the plurality of patients;

remove the protected health information from the health information corresponding to the randomly-selected subset of the plurality of patients to produce a de-identified analytics subset of patient data suitable for analytical model construction and evaluation; and

store the updated de-identified analytics subset of patient data in the one or more databases.

21. The computer-readable storage medium of claim 19, comprising further instructions that, when executed, cause the one or more processors to:

receive an analytical model based on the de-identified analytics subset of patient data; and

apply the analytical model to the health information with the protected health information for the plurality of patients.

22. The computer-readable storage medium of claim 21, comprising further instructions that, when executed, cause the one or more processors to output a summary of the results of the application of the analytical model to the health information with the protected health information for the plurality of patients to a user via the user interface restricting user access to the protected health information.

23. The computer-readable storage medium of claim 21, comprising further instructions that, when executed, cause the one or more processors to:

access a result of the of the application of the analytical model to the de-identified analytics subset of patient data;

compare the result of the of the application of the analytical model to the de-identified analytics subset of patient data to a result of the of the application of the analytical model to the health information for the plurality of patients; and

present a summary of the comparison of the result of the of the application of the analytical model to the de-identified analytics subset of patient data with the result of the of the application of the analytical model to the health information for the plurality of patients to a user via the user interface restricting user access to the protected health information to facilitate evaluation of the suitability of the analytical model for the health information for the plurality of patients.

24. The computer-readable storage medium of claim 19, wherein in removing the protected health information from the health information for the randomly-selected subset of the plurality of patients to produce the de-identified analytics subset of patient data suitable for analytical model construction and evaluation, the instructions cause the one or more processors to:

remove data from fields identified as including protected health information; and

filter unstructured data to remove unstructured protected health information.

25. The computer-readable storage medium of claim 19, wherein in producing the de-identified analytics subset of patient data suitable for analytical model construction and evaluation further the instructions cause the one or more processors to:

receive, via a user interface, indications of additional protected health information within the health information for the randomly-selected subset of the plurality of patients after removing the protected health information from the health information; and

remove the indicated additional protected health information from the health information for the randomly-selected subset of the plurality of patients.

26. The computer-readable storage medium of claim 19, comprising further instructions that, when executed, cause the one or more processors to:

receive, via the user interface, a request to access protected health information removed to produce the de-identified analytics subset of patient data;

present the requested protected health information via the user interface; and

store a record of the request to access the protected health information.

27. The computer-readable storage medium of claim 19, comprising further instructions that, when executed, cause the one or more processors to:

follow an update to the one or more databases comprising health information for the plurality of patients, the update including health information for additional patients, randomly selecting health information from the one or more databases of a subset of the additional patients;

remove the protected health information from the health information for the randomly-selected subset of the additional patients;

add the health information for the randomly-selected subset of the additional patients without the removed protected health information to the de-identified analytics subset of patient data to produce an updated de-identified analytics subset of patient data; and

present a representation the updated de-identified analytics subset of patient data via the user interface restricting user access to the protected health information for analytical model construction and evaluation.

20. The computer-readable storage medium of 19, comprising further instructions that, when executed, cause the one or more processors to present, via a user interface, the updated de-identified analytics subset of patient data for analytical model construction and evaluation.