Method for market risk assessment for healthcare applications

Patent metadata

Field	Value
Publication number	US-2016321748-A1
Application number	US-201514699482-A
Country	US
Kind code	A1
Filing date	Apr 29, 2015
Priority date	Apr 29, 2015
Publication date	Nov 3, 2016
Grant date	—

Abstract

Exemplary embodiments of the present invention provide a method of health insurance market risk assessment including receiving first data including demographic and cost data for members of a health insurance plan in a current market, receiving second data including demographic data for the current market, and receiving third data including demographic data for a new market. The first to third data are used to transform a distribution of the plan members to account for differences between the current and new market demographic data and to estimate probabilities of enrollment in the new market. A statistical model is learned to predict risk in the new market using the transformed distribution and the estimated probabilities. The statistical model is used to determine risk of entering the new market.

First claim

What is claimed is:

1 . A computer-based method of health insurance market risk assessment, comprising:
receiving first data, the first data including demographic and cost data for members of a health insurance plan in a current market; receiving second data, the second data including demographic data for the current market; receiving third data, the third data including demographic data for a new market; using the first to third data to transform a distribution of the plan members to account for differences between the current and new market demographic data and estimate probabilities of enrollment in the new market; learning a statistical model to predict risk in the new market using the transformed distribution and the estimated probabilities; and using the statistical model to determine risk of entering the new market.

2 . The method of claim 1 , further comprising performing a privacy-preserving transformation on the first data.

3 . The method of claim 2 , wherein the privacy-preserving transformation includes a clustering procedure where each cluster contains at least a pre-specified number (k) of members followed by a probability transformation.

4 . The method of claim 1 , wherein the transform of the distribution of the plan members and the estimate of the probabilities of enrollment in the new market occur at the same time.

5 . The method of claim 1 , further comprising modifying the estimates of the enrollment probabilities.

6 . The method of claim 5 , wherein the estimates of the enrollment probabilities are displayed to and modified by a user.

7 . The method of claim 1 , wherein the statistical model is learned using a non-demographic factor of the new market.

8 . The method of claim 1 , further comprising using the statistical model to produce individual-level cost predictions of entering the new market.

9 . The method of claim 8 , further comprising aggregating the individual-level cost predictions according to user-defined criteria.

10 . The method of claim 9 , wherein the aggregation is performed using a computing device.

11 . A computer-based method of health insurance market risk assessment, comprising
transforming a distribution of existing plan members to account for differences between existing and new market demographics while estimating and accounting for probabilities of enrollment in the new market; learning a statistical model to predict risk in the new market using the transformed distribution and the estimated probabilities; and using the statistical model to determine risk of entering the new market.

12 . The method of claim 11 , further comprising:
receiving adjustments to initially estimated enrollment probabilities.

13 . The method of claim 12 , wherein the adjustment is received from a subject matter expert.

14 . The method of claim 11 , wherein using the statistical model to determine the risk of entering the new market comprises:
computing individual-level cost predictions.

15 . The method of claim 14 , further comprising:
aggregating the individual-level cost predictions according to user-defined criteria.

16 . The method of claim 15 , wherein the aggregated individual-level cost predictions are displayed on a computing device.

17 . The method of claim 11 , wherein the statistical model includes a plurality of predictive values.

18 . The method of claim 11 , further comprising:
applying a privacy preservation measure to data indicative of the existing plan members.

19 . The method of claim 17 , Wherein the privacy preservation measure is applied to the data which has already been removed of personal identifiers.

20 . A computer-based method of health insurance market risk assessment, comprising:
aggregate claims of current members of a health insurance plan to estimate demographic distribution of the current members; for each demographic group in the estimated demographic distribution of the current members, compute aggregate statistics of corresponding health costs; aggregate demographic data for the current member's market to estimate a demographic distribution of a current market; aggregate demographic data for a new market to estimate a demographic distribution of the new market; for each demographic group of the estimated demographic distribution of the current market and the estimated demographic distribution of the new market, compute a ratio of new market distribution; re-weighting the aggregated claims of the current members and the aggregated statistics of the corresponding health costs using the ratio of new market distribution; and leaning a model for predicting risk of entering the new market by performing a linear regression of cost on demographic variables using the re-weighted aggregated claims of the current members and the aggregated statistics of the corresponding health costs.

21 . The method of claim 20 , further comprising:
adjusting predictions made by the learned model by multiplying the predictions with a cost factor for the new market.

22 . The method of claim 21 , further comprising:
aggregating the adjusted predictions according to pre-defined criteria.

23 . The method of claim 22 , further comprising:
visually alerting a user to low-risk regions via a computing device using the aggregated adjusted predictions.

Assignees

• IBM

Inventors

• Mahatma Shilpa
• Mojsilovic Aleksandra
• Ramamurthy Karthikeyan Natesan
• Varshney Kush R
• Wei Dennis
• Yuen-Reed Gigi

Classifications

• G06Q40/04 Primary

  Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange · CPC title

• G06F19/328

  Physics · mapped topic

• G06Q40/08

  Insurance · CPC title