Although predictive modeling has been utilized by P&C insurers for years, the focus has primarily been on pricing and underwriting risks. We now are seeing increased interest and utilization of that process to manage claims. Current example applications of predictive modeling for claims include:

Identifying potentially fraudulent claims
Predicting which claims may benefit from having a specialist assigned to them
Developing logic to more scientifically triage claims to the right level of expertise in the organization
Determining what differences in the book of business are impacting claim severity

Process Can Reduce Errors

At P&C insurers and third party administrators, the claims management task is handled by teams of professionals consisting of claim adjusters, claim managers, nurse case managers, client risk mangers, and other claim and medical experts. While these professionals generally have years of experience and can identify claims that may need intervention, predictive modeling can help add discipline to this process and reduce the potential for human error by processing multivariate patterns that humans are incapable of recognizing.

At SRS, our current predictive modeling solutions rely on data from several areas: claimant demographics, employment data, injury details, medical billing information, and claim financials. We also use natural language processing to text mine and extract key information from claim handler notes, and then integrate this data into our models.

Predictive modeling can be especially useful in identifying large losses. While large losses account for only five to ten percent of open claims across the workers' compensation market, they account for 60 to 75 percent of the outstanding dollars for workers' compensation claims.

Early identification of large losses is difficult, as claims can take many different courses based on the claimant, the injury, and the treatment. When seemingly innocuous claims deteriorate, it often takes more than a year before their true costs become known. Failure to recognize the magnitude of these claims early can put an injured worker down the wrong path and lead to spiraling claim costs.

Assist in Early Identification

Predictive modeling can facilitate early identification of claims that would benefit from nurse case management. The process also can flag claims that may not currently have large losses but do have a high likelihood of adverse development.

At SRS, our nurse case management model estimates financial risk and evaluates claims shortly after intake and then performs another sweep at 25 days when additional data is available. We have the flexibility to utilize either a score-based threshold or a volume-based threshold to determine which claims are likely to need intervention from a nurse. The goal is to reduce overall claim costs through targeted early intervention and to get the injured worker back to work as soon as possible.

Industry estimates indicate that fewer than 10 percent of large losses are recognized at 90 days. At SRS, by using predictive modeling, we can identify up to two-thirds of large losses by focusing on those claims that have scored the highest.

It must be noted that no predictive model is perfectly accurate and that accuracy depends heavily on the manner of deployment. Nevertheless, those companies striving to lead the pack in claims management and mitigating losses would be wise to incorporate predictive modeling into their strategies and workflows.

Kaleb Adams is vice president of predictive modeling at Specialty Risk Services, Hartford, Conn. On Monday, Aug. 16, he and other modeling experts will examine predictive analytics at the FWCI 65th Annual Workers' Compensation Educational Conference and 22nd Annual Safety and Health Conference in Orlando. The discussion will center on how insurers and their risk management clients can get a better handle on their respective workers' compensation exposures using these latest technology tools and techniques. In addition to Adams, panel members include Jennifer Tomilin, senior vice president, Zurich North America, Schaumburg, Ill.; and Steven W. Laudermilch, senior manager, Deloitte Consulting, Philadelphia. Sam Friedman, editor-in-chief of National Underwriter P&C, Hoboken, N.J., will moderate. Conference information is available at www.fwciweb.org.

Special Reports /

Tech to The Rescue! How Predictive Modeling Can Save Workers&#039; Compensation Carriers

Related Stories

Recommended For You

Tech to The Rescue! How Predictive Modeling Can Save Workers' Compensation Carriers