Predictive Models Won’t Replace People, But Can Help With Underwriting, Claims - Risk Management - Property and Casualty Insurance News

Predictive Models Won’t Replace People, But Can Help With Underwriting, Claims - Risk Management - Property and Casualty Insurance News:

"Breaking News
NU Exclusives

Predictive Models Won’t Replace People, But Can Help With Underwriting, Claims
New tech tools allow insurers to spot patterns or key correlations for human review
By DANIEL HAYS

Published 9/7/2009


Predictive Models Won’t Replace People, But Can Help With Underwriting, Claims
New tech tools allow insurers to spot patterns or key correlations for human review

By DANIEL HAYS
Published 9/7/2009

Orlando

While predictive modeling won’t replace knowledgeable insurance adjusters and underwriters, it is the wave of the future for the workers’ compensation industry, a trio of experts here stressed.

Jennifer Tomilin, senior vice president at Zurich North America, said she could not foresee human underwriters ever being replaced by automation because “there are areas where we don’t have enough data for predictive modeling.”

However, modeling will be a big part of every insurer’s future, while companies that are “stuck in the mud” and fail to take advantage of this evolving technology, “those folks are going to be left behind,” warned Steve Laudermilch, senior manager at Deloitte Consulting.

The three panelists offered their views and predictions during a panel session—“Tech To The Rescue: Can Predictive Analytics Save Workers’ Comp?”—here at the annual Workers’ Compensation Educational Conference. The session was part of the National Trends program put together by National Underwriter in partnership with the WCEC organizers—the Florida Workers’ Compensation Institute.

The speakers outlined a variety of ways in which modeling is used, what data types it employs and what it targets.

Ms. Tomilin said modeling—sometimes called data mining—is an analysis that finds unsuspected relationships that relate to data in novel ways.

For example, in researching a company, the modeler might look at how many employees in a particular ZIP code work from home, figuring a return to work program might be easier to employ for an injured employee who does not have to commute to their job.

Some data that is sifted through is ultimately tossed out, she noted, giving an example of an examination of sweet dessert purchases in a certain ZIP code for a possible link to truck accidents, because eating sugar can cause drowsiness.

She also noted that discriminatory data of the kind once used in redlining is rejected.

The value of predictive models, she said, is that they take “a lot of the subjectivity out” of analysis and eliminate “the gut-feel practices” of underwriters.

Predictive modeling can serve insurers by, among other things, targeting fraud, identifying claimants who will benefit by treatment from a specialist, and in helping to “triage claims,” according to Kaleb Adams, vice president of Predictive Modeling at Specialty Risk Services.

While he said data from an insurer’s underwriting department is not always that reliable or useful in predictive modeling, solid data is voluminous from the claims sector and can be used to provide facts concerning a loss, a worker’s claims history and co-morbidity factors such as obesity.

He said modeling can pin down cases that would benefit from having a special nurse assigned, those that indicate they will involve a large loss, and those that need less attention.

Mr. Adams said the system can use text mining to read over notes and identify a potential fraud situation, a health problem such as morbid obesity, or a risk management threat such as a wet floor.
Models that are used, Mr. Laudermilch noted, are built by testing the impact of hundreds of variables—such as lifestyle, age, employment history and pharmacy drug use—against closed claims.

The models, he said, will be based on data on injury groups, such as back strains, to identify which are the more serious cases in relation to all the others.

“This is not a replacement for adjusters,” he noted. “The idea is to give them claims they can work on, where they can make a big impact” accelerating best practices.

Explaining some of the areas where data is derived, Mr. Laudermilch noted that companies glean it from items such as warrantee forms that customers fill out.

Learning that a person has interests in outdoor pursuits or running for fitness can lead to a conclusion that if injured, “they are probably motivated to get healthy”—more so, perhaps, than someone who is a “couch potato,” he added.

Read Original Post: http://www.property-casualty.com/Issues/2009/September%207%202009/Pages/Predictive-Models-Wont-Replace-People-But-Can-Help-With-Underwriting-Claims.aspx
***********************************************************************
http://dreamlearndobecome.blogspot.com This posting was made my Jim Jacobs, President & CEO of Jacobs Executive Advisors. Jim also serves as Leader of Jacobs Advisors' Insurance Practice.