Method and computer program product for predicting and minimizing future behavioral health-related hospital admissions

• US 20060224416A1
• Filed: 12/12/2005
• Published: 10/05/2006
• Est. Priority Date: 03/29/2005
• Status: Active Grant

• Current Assignee: Coventry Health Care of Missouri, Inc. (CVS Health Corporation)
• Original Assignee: Coventry Health Care of Missouri, Inc. (CVS Health Corporation)
• Inventors: Tan, Agnes W. H., Dancha, Lynne A., Koopmeiners, Michael, Lloyd, Karen D., Lindquist, Tammie J.
• Granted Patent: US 8,190,451 B2
• Time in Patent Office: Days
• US Class Current: 705/2
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G16H 20/70 relating to mental therapies, e.g. psychological therapy or autogenous training
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

An accurate predictive model that identifies the patient/members within the healthcare system at high risk of hospital admission for a wide range of morbidities, or co-morbidities, and that allows subsequent intervention to manage those patients identified as high risk to an acceptable level. There is a further need for such a predictive model that focuses on specific groupings of conditions, e.g., behavioral health predictive modeling. There is also a need for a method that provides for intervention to manage the risk to the identified patients/members. One embodiment of the present invention discloses and claims a method of high-risk patient identification and management. In one aspect, the inventive method may comprise compiling a listing including all individuals with any primary behavioral health diagnosis over a specified time period; merging the listing with at least one data source to extract at least one behavioral health-related predictive factor; generating, based on at least one predictive model, a predictive output comprising the probability that the individuals listed will require a future behavioral health-related hospital admission; identifying the high-risk individuals from the predictive model output; and intervening with the high risk members to identify and modify, to the extent possible, the risk factors that place the member at high risk.

Claims

1. A method for predicting and minimizing future behavioral health-related hospital admissions, comprising:

compiling a listing including all individuals with any primary behavioral health diagnosis over a specified time period;

merging the listing with at least one data source to extract at least one behavioral health-related predictive factor;

generating, based on at least one predictive model, a predictive output comprising the probability that the individuals listed will require a future behavioral health-related hospital admission;

identifying the high-risk individuals from the predictive model output; and

intervening with the high-risk individuals.

2. The method of claim 1, wherein the compiling further comprises obtaining behavioral health-related data for individuals from historical data sources.

3. The method of claim 2, further comprising obtaining the data from current data sources.

4. The method of claim 1, wherein the merging further comprises acquiring data indicating a secondary diagnosis of behavioral health conditions.

5. The method of claim 1, wherein the merging further comprises acquiring data indicating emergency room visits.

6. The method of claim 1, wherein the merging further comprises acquiring data indicating comorbidities for behavioral health-related hospital admissions.

7. The method of claim 1, wherein the merging further comprises acquiring data indicating inpatient hospital service utilization.

8. The method of claim 1, wherein the merging further comprises acquiring data indicating utilization of all professional services.

9. The method of claim 1, wherein the merging further comprises acquiring data regarding history of psychotropic drug use.

10. The method of claim 1, wherein at the least one predictive factor is inserted into the at least one predictive model to generate the predictive output.

11. The method of claim 1, wherein the at least one predictive model is specific to individual primary diagnoses.

12. The method of claim 1, wherein the at least one predictive model is specific to individual secondary diagnoses.

13. The method of claim 1, wherein the at least one predictive model is specific to individual primary and individual secondary diagnoses.

14. The method of claim 1, wherein the identifying further comprises selecting the patients in a selected percentage of the predictive output.

15. The method of claim 1, wherein the identifying further comprises selecting the patients in the upper 5% of the predictive output.

16. The method of claim 1, wherein the identifying further comprises selecting the patients in the upper 1% of the predictive output.

17. The method of claim 1, further comprising using a programmed digital computer for the compiling, merging, identifying and generating.

18. The method of claim 1, further comprising using a computer program product for the compiling, merging, identifying and generating.

19. The method of claim 17, further comprising using a computer program product for the compiling, merging, identifying and generating.

20. A computer program product for predicting and minimizing future behavioral health-related hospital admissions, comprising:

computer code for compiling a listing including all individuals with any primary behavioral health diagnosis over a specified time period;

computer code for merging the listing with at least one data source to extract at least one behavioral health-related predictive factor;

computer code for generating, based on at least one predictive model, a predictive output comprising the probability that the individuals listed will require a future behavioral health-related hospital admission;

computer code for identifying the high-risk individuals from the predictive model output; and

computer code for intervening with the high-risk individuals.

21. A programmed digital computer for predicting and minimizing future behavioral health-related hospital admissions, comprising:

a processor;

a memory operatively coupled to the processor;

a data input interface operatively coupled to the memory; and

a data output interface operatively coupled to the memory;

wherein the programmed digital computer operates to pull a list of individuals with any primary behavioral health diagnosis over a specified time period in response to at least one instruction and to store the list of the plurality of individuals in the memory;

wherein the programmed digital computer operates to merge the listing with at least one data source to extract behavioral health-related information in response to at least one instruction;

wherein the programmed digital computer operates to generate an output comprising the probability that the listed individuals will require a future behavioral health-related hospital admission based on at least one predictive model and in response to at least one instruction; and

wherein the programmed digital computer operates to identify the high-risk individuals from the predictive model output in response to at least one instruction.