Database essentials: insurers are faced with the fairness-consistency challenge in settling bodily injury claims.

In advocating adherence to cost-based pricing as the surest path to strong underwriting performance, I always stress the importance of having access to vast quantities of high-quality data, and the right analytic tools to make sound business decisions.

Among property/casualty insurers it is widely accepted that deep reserves of high-quality data are essential to profitable rating, underwriting and reserving. But when it comes to claims adjudication--another critical component of the insurance value chain--insurers have been unable to preserve and manage essential information contained in bodily injury claims.

For ratemaking, fraud detection, specific property information and other essential decision support, ISO is able to amass, store and manage essential data.

For ratemaking alone, our statistical database contains 11.3 billion commercial and personal lines records.This enormous database grows as companies submit more than 2 billion policy and claim records each year. Insurers add more than 60,000 detailed building records yearly to our specific property database of 3 million records.And insurers fight claim fraud by adding 45 million claims yearly to the industry's all-claims database, which now totals 456 million records.

BI losses make up a major portion of dollars paid in both the commercial and personal lines. In monoline products liability, for example, direct-paid BI indemnity losses for the five calendar years 2000-2004 totaled approximately $6.3 billion, excluding allocated loss-adjustment expenses: slightly more than 50% of the $12.3 billion paid by the line in that period.

At $41.8 billion, BI indemnity losses (excluding allocated loss-adjustment expenses) represent more than 75% of all commercial auto liability losses for the same period.And in personal passenger auto, BI liability direct-paid losses were $119.7 billion, or nearly 52% of total personal auto liability losses for the same five-year period.

Thousands of BI claims are processed and settled every day. Once a claim file is closed, however, valuable information is locked away--information containing literally billions of data elements on injuries, treatments, settlement values, attorneys, physicians and economic regions.

Every day, claims handlers in most companies make decisions about complex injury claim payments without historic company or industry benchmarks because there is no database of prior settlement information.Without access to data, companies cannot use settlement trends to train claim handlers, develop predictive models or sharpen underwriting decisions.

For years, the industry has struggled to be consistent in paying injury claims. But without access to data on settlement patterns, litigation results and new treatment procedures, achieving consistency is difficult--if not impossible.

In a test of claims professionals, the ISO Claims Services unit concluded that any two adjusters looking at the same claim have a 60% or greater chance of being 100% apart in their assessment of a BI loss.

Insurers rely on adjuster training, file review and staff mentoring to achieve consistency and fairness in BI claim settlement, while some companies add injury indicators and injury classifications to their claims administration systems.

Other companies try to limit claims-payment inconsistency by modeling past payments for pain and suffering. Such modeling tools improve consistency in payments based on medical severity of BI claims. But other significant variables could be leveraged with additional predictive modeling to improve payment consistency even more.

With today's enormous computing capacity and sharply declining data-storage costs, data warehousing detailed BI claims settlement information is eminently possible.Amassing BI data can help insurers meet the fairness-consistency challenge of settling claims in several ways. Such a data source could be company-specific, capturing detailed injury claims information for insurer decision making and predictive modeling. Moreover, a large aggregate BI data source could enable companies to compare their results with industry benchmarks for decision making and modeling.

As the business environment becomes more challenging, insurers are finding new technology is key to survival and prosperity. Saving and maintaining detailed claims information is a good first step toward a more proactive system of handling BI claims and achieving fair, consistent and cost-effective settlements.

Frank J. Coyne, a Best's Review columnist, is chairman, president and chief executive officer of ISO. He can be reached at [e-mail].