Predicting a New Toolkit for Claims Professionals
Predictive analytics techniques are widely used in the insurance industry today, usually for actuarial and underwriting purposes. Using analysis of patterns in past events to model future outcomes or behaviors has made huge strides, largely thanks to advances in so-called Big Data technologies.
Now, some property/casualty carriers are finding that predictive modeling can be put to good use in claims management, supporting a claims handler’s own judgment and analysis.
Predictive modeling essentially compares factors associated with new and pending claims against those of past losses. This can include nature of injury, treatment, characteristics of the claimant, insured data, liability, attorneys involved and venue, among others.
For example, claims with the potential for high defense costs can be identified, as well as which defense firms were associated with favorable outcomes involving similar type cases.
Claims can be ranked according to their subrogation potential, so that the claims professional can take another look at losses that otherwise may not have been identified for possible recoveries and either rule out or take appropriate action before it’s too late.
Arguably, the most valuable use of predictive modeling in claims is identifying potential fraud, by flagging similarities that an ongoing claim may have with past claims that were confirmed to be fraudulent.
One of the more interesting uses of predictive modeling is its reported ability to provide early warning of potential “outliers" claims that appear routine but eventually develop into high value losses. In Workers’ Comp., for example, claim professionals sometimes find it hard to assess loss values associated with apparently routine claims that develop strongly over time - the “creeping Cats.”
It’s not yet clear if the use of analytics in claims management will eventually become standard operating procedure. But if the technology available today can enhance efficiencies and improve loss costs, it must be worth a closer look.