BI’s Article search uses Boolean search capabilities. If you are not familiar with these principles, here are some quick tips.

To search specifically for more than one word, put the search term in quotation marks. For example, “workers compensation”. This will limit your search to that combination of words.

To search for a combination of terms, use quotations and the & symbol. For example, “hurricane” & “loss”.

Login Register Subscribe

Predictive modeling pits work comp insurers against self-insured groups: Panel

Predictive modeling pits work comp insurers against self-insured groups: Panel

CHICAGO — Predictive modeling among workers compensation insurers is presenting a challenge to self-insured groups, who are facing off with insurers that are offering increasingly competitive pricing, according to panelists who spoke Tuesday at the Self-Insurance Institute of America Inc. annual conference.

Predictive technology can be helpful to identify “migratory catastrophe claims,” comp claims that seem relatively minor at first, but last for years and result in significant costs, said Steven J. Link, executive vice president and chief innovation officer for Midwest Employers Casualty Co. in Chesterfield, Mo.

Mr. Link gave the example of an injured worker whose claim costs had reached $388,000 and who was taking 11 medications before Midwest's excess comp coverage kicked in. The insurer entered the worker into a functional restoration treatment program, allowing the woman to return to work and to stop taking her prescriptions within six months of the intervention.

“Had we picked it off in six months instead of five years and the interventions occurred sooner, we would have saved that $388,000 or a big chunk of that, plus a lot of the pain and misery that person went through,” Mr. Link said, adding that Midwest Employers has begun using predictive modeling to catch such claims early on.

While models can help reduce claim costs and duration, they also are allowing traditional insurers to provide competitive comp pricing to some clients instead of throwing out entire books of business, such as construction firms, Mr. Link said. He said that dynamic has hurt SIGs during the soft workers comp market.


Insurers are “smarter about what they're hanging onto and they're smarter about what they let go, and that's negative to the SIG industry, who historically writes a whirlwind at this stage of the cycle,” Mr. Link said.

SIGs can work with consultants to take advantage of predictive modeling technology and improve their own pricing, said Stanford A. Smith, predictive analytics manager for the northeast region at consultancy Milliman Inc. in Wakefield, Mass.

“You'll be much more informed,” Mr. Smith said. “I look at it as managing the underwriting process with intelligence that's not been available before. It affects both the selection and guides the pricing process.”

Mike Crandall, founder and owner of Covenant Risk Partners Inc. in Norcross, Ga., said SIGs should see predictive models as a tool that can help make them more competitive with insurers.

“This is an opportunity to bring improvements though similar modeling techniques to both sides of the equation, so you have improved risk selection, more appropriate pricing and you're doing … smarter things sooner on the claims side to help over time improve the profitability of the business you have,” Mr. Crandall said.

The SIIA panel was moderated by Kimble E. Coaker, CEO and administrator of the Alabama Trucking Association Workers Compensation Fund in Montgomery, Ala.

Read Next

  • SIIA appeals Michigan court ruling on ERISA pre-emption

    The Self-Insurance Institute of America Inc. is appealing a federal court ruling that said the Employee Retirement Income Security Act did not pre-empt a 2011 Michigan law imposing a 1% tax on paid health care claims.