Integrated health management platform

• US 8,200,506 B2
• Filed: 12/19/2006
• Issued: 06/12/2012
• Est. Priority Date: 12/19/2006
• Status: Active Grant

• Current Assignee: Accenture Global Services Limited (Accenture PLC)
• Original Assignee: Accenture Global Services Limited (Accenture PLC)
• Inventors: Kil, David H.
• Primary Examiner(s): O'Connor, Gerald J.
• Assistant Examiner(s): PHONGSVIRAJATI, POONSIN
• Application Number:
  US11/612,763
• Publication Number:
  US 20080147438A1
• Time in Patent Office: 2,002 Days
• Field of Search: 705 2- 3, 707373-740
• US Class Current: 705/3
• CPC Class Codes: G06Q 40/08 Insurance
• CPC Class Codes: G16H 20/00 ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

Apparatuses, computer media, and methods for supporting health needs of a consumer by processing input data. An integrated health management platform supports the management of healthcare by obtaining multi-dimensional input data for a consumer, determining a health-trajectory predictor from the multi-dimensional input data, identifying a target of opportunity for the consumer in accordance with the health-trajectory predictor, and offering the target of opportunity for the consumer. Multi-dimensional input data may include claim data, consumer behavior marketing data, self-reported data, and biometric data. A consumer may be assigned to a cluster based on the multi-dimensional input data and a characteristic of the consumer may be inferred. A cluster may be associated with a disease progression, and a target of opportunity is determined from the cluster and the disease progression. An impact of the target of opportunity may be assessed by delivering treatment to a consumer at an appropriate time.

Claims

1. A method for managing healthcare, comprising:

obtaining multi-dimensional input data for each of a plurality of consumers, the multi-dimensional data including, for each consumer, claim data, self-reported data, and consumer behavior marketing data for the consumer, the consumer behavior marketing data further including behavior, lifestyle, and attitudinal information about the consumer;

a computer using the multi-dimensional input data to iteratively generate a set of clusters for each of a plurality of outcome variables, wherein the clusters for each outcome variable are generated to maximize an output dispersion compression by comparing a ratio of a dispersion of each cluster and an overall dispersion of a population to a predetermined dispersion-compression threshold;

a computer assigning a first consumer to at least one cluster according to multi-dimensional input data for the consumer;

a computer determining for the first consumer at least one health-trajectory prediction from a model associated with the at least one cluster for the first consumer and the multi- dimensional input data for the first consumer;

the computer identifying a target of opportunity for the consumer in accordance with the at least one health-trajectory prediction; and

the computer offering the target of opportunity for the first consumer.

2. The method of claim 1, further comprising:

inferring a characteristic of the first consumer from a subset of the multidimensional input data for the first consumer.

3. The method of claim 1, further comprising:

associating a disease progression based on one of the at least one cluster, said one of the at least one cluster being associated with at least one attribute of the consumer.

4. The method of claim 3, wherein the target of opportunity is determined from said one of the at least one cluster and the disease progression.

5. The method of claim 4, further comprising:

assessing an impact of the target of opportunity for the first consumer.

6. The method of claim 5, wherein assessing the impact of the target of opportunity comprises using a Markov model.

7. The method of claim 1, further comprising:

extracting the target of opportunity from medical information using a set of rules for the multi-dimensional input data for the first consumer.

8. The method of claim 1, the multi-dimensional input data further including health risk assessment data.

9. The method of claim 1, the multi-dimensional input data further including biometric data.

10. The method of claim 1, further comprising:

forming a consensus view when the consumer belongs to a plurality of clusters.

11. The method of claim 1, further comprising:

identifying a previous event that occurred to a subsequent transition event; and

measuring a correlation between the previous event and the subsequent transition event.

12. The method of claim 1, further comprising:

sending a reminder to the consumer if the consumer does not utilize the target of opportunity, wherein the target of opportunity for the consumer is in accordance with the at least one health-trajectory prediction.

13. The method of claim 1, the assigning comprising:

determining a lifestyle of the consumer; and

combining a health condition of the consumer with the lifestyle for assigning the consumer to said one of the at least one cluster.

14. The method of claim 1, further comprising:

determining at least one unobserved trait of the first consumer from the multidimensional input data; and

tailoring an intervention strategy for the first consumer from the at least one unobserved trait.

15. An apparatus that manages healthcare, comprising:

a memory; and

a processor accessing the memory to obtain computer-executable instructions, and the processor executing the computer-executable instructions to perform the following operations;

obtaining multi-dimensional input data for each of a plurality of consumers, the multi-dimensional data including, for each consumer, claim data and consumer behavior marketing data for the consumer;

using the multi-dimensional input data to iteratively generate a set of clusters for each of a plurality of outcome variables, wherein the clusters for each outcome variable are generated to maximize an output dispersion compression by comparing a ratio of a dispersion of each cluster and an overall dispersion of a population to a predetermined dispersion-compression threshold;

assigning a first consumer to at least one cluster according to multi-dimensional input data for the consumer;

determining, for the first consumer, at least one health-trajectory predictor from a model associated with the at least one cluster for the first consumer and the multi-dimensional input data for the first consumer;

identifying a target of opportunity for the consumer in accordance with the at least one health-trajectory predictor; and

offering the target of opportunity to the consumer.

16. The apparatus of claim 15, the processor further executing the computer-executable instructions for performing the following operation:

forming a consensus view when the consumer is assigned to a plurality of clusters.

17. The apparatus of claim 16, the processor further executing the computer-executable instructions for performing the following operations:

associating a disease progression based on one of the at least one cluster, said one of the at least one cluster being associated with at least one attribute of the consumer, wherein the target of opportunity is determined from said one of the at least one cluster and the disease progression.

18. The apparatus of claim 17, the processor further executing the computer-executable instructions for performing the following operation:

assessing an impact of the target of opportunity for the first consumer.

19. A non-transitory computer-readable medium having computer-executable instructions that, when executed by a computer, cause the computer to perform operations comprising:

obtaining multi-dimensional input data for each of a plurality of consumers, the multi-dimensional data including, for each consumer, claim data and consumer behavior marketing data for the consumer;

a computer using the multi-dimensional input data to iteratively generate a set of clusters for each of a plurality of outcome variables, wherein the clusters for each outcome variable are generated to maximize an output dispersion compression by comparing a ratio of a dispersion of each cluster and an overall dispersion of a population to a predetermined dispersion-compression threshold;

a computer assigning a first consumer to at least one cluster according to multi-dimensional input data for the consumer;

determining, for the first consumer, at least one health-trajectory predictor from a model associated with the at least one cluster for the first consumer and the multi-dimensional input data for the first consumer;

identifying a target of opportunity for the consumer in accordance with the at least one health-trajectory predictor; and

offering the target of opportunity to the first consumer.

20. The non-transitory computer-readable medium of claim 19, wherein the instructions further cause the computer to perform operations comprising:

assigning the first consumer to another cluster based on the multi-dimensional input data; and

forming a consensus view when the first consumer belongs to a plurality of clusters.

21. A computer-implemented method, comprising:

obtaining multi-dimensional input data for a consumer, the multi-dimensional input data including claim data, self-reported data, and consumer behavior marketing data;

using the multi-dimensional input data to iteratively generate a set of clusters for each of a plurality of outcome variables, wherein the clusters for each outcome variable are generated to maximize an output dispersion compression by comparing a ratio of a dispersion of each cluster and an overall dispersion of a population to a predetermined dispersion-compression threshold;

assigning the consumer to a particular cluster in the set of clusters, wherein the clusters are defined according to an impact of treatment on consumers in each cluster;

determining a health trajectory prediction for the consumer according to the multi-dimensional input data and a model specific to the particular cluster, wherein the health trajectory prediction predicts a future status of a health of the consumer according to disease progression, engagement, and impact;

identifying a target of opportunity from the health trajectory prediction, wherein the target of opportunity comprises a treatment for a disease of the consumer; and

presenting the target of opportunity to the consumer;

wherein the obtaining, using, assigning, determining, identifying, and presenting are performed by one or more computers.

22. The method of claim 21, further comprising generating the set of clusters to maximize an output dispersion compression for the plurality of clusters.

23. The method of claim 21, further comprising determining at least one unobserved trait of the consumer from the multidimensional input data.

24. The method of claim 21, further comprising using a Markov model to estimate an impact of the target of opportunity on the consumer.