Is Your Loss Run a Loser?

What do insurance loss runs and Rodney Dangerfield have in common? They both get no respect. Few insureds ever seem to see an insurance loss run they like. Talk to policyholders, and a pet peeve many have are those confounding printouts! Whether generated by insurers or by third-party administrators (TPAs), loss runs are sore points with many clients, satisfying almost no one.

Too much information, and customers say theyre incomprehensible.

Skimp on the data, and theyre meaningless.

Whats a happy medium?

On a fun scale, deciphering loss runs ranks with root-canal therapy and reading minutes of Federal Reserve Board meetings. Few seem to have a loss run they think is a good model or template. CIOs and tech professionals need not, however, resign themselves to the fact that there isnt a format out there that satisfies anyone. Loss runs can, however, be good for something other than functioning as oversized paperweights.
Lets see what insurance customers say about typical loss-run deficiencies before turning to ways to improve them.

Inaccuracy Woes

Sometimes the problem is not with the loss runs format but with data reliability. Precision and accuracy often are elusive. The operative expression is garbage in and garbage out (GIGO). According to Bob Morrissey, senior partner with MSA Insurance Operations Consultants, in New Jersey: My experience with loss runs is not so much their clutter with superfluous or blank sections, its the inaccuracy. Often, customers list loss-run inaccuracy as the major problem with their carrier and that they dont tie into payment screens. For example, he recalls one client had loss-run totals that were $4,000 over the payment screen. Four thousand dollars may seem immaterial, but multiply that by the number of claimants, and it is quite substantial. So, Morrissey concludes, while a simple format is necessary, accuracy is more critical.

Often, though, the concerns involve loss-run format and expandability. Dave Parker, Pima County, Ariz., risk manager, operates the computer system for all state claims, a custom program running on an IBM mainframe. Originally, Parker hated loss-run reports. After I was given the ability to run my own reports with the parameters I needed, he notes, we found a happy medium. Getting a run of all losses, or our favorite triangulation report, usually doesnt do anything except tell us how many cases we had and how the costs developed. You need the ability to ask, What if? It is the ability to tailor the printouts that makes the system valuable.

Thus, adaptability is a key loss-run feature. Like designing computer systems, crafting a loss run frequently is an experience where users later think of wants and needs that never occurred to them during the original planning process. If clients want to make changes later, they should be able to perform that task.
John Price, a consultant with Safety Management Services, recalls, Years ago, Travelers had a fair loss run for workers compensation. He adds, however, You had to be a big-time operator to get it. Then they came out with FOCUS, and that was an improvementlet the safety types design the runs tailored to the client. It was nice but time-consuming.

When I came to California, Price adds, someone had saturated the workers compensation market with Wang computer systems and a fairly standard workers compensation loss run. In short, it inhaled with great vigormost of them still do. The information is factual but one dimensional. Loss data is multifaceted and should be reported that way. Listing only by employee, department, cost, body part, time of day, etc., doesnt tell the whole story.

Doing ergonomic consulting, Price continues, Id take the loss runs, build my own database, and create my own reports using either dBase or Reflex. Time-consuming and expensive, but the data I got out was invaluable. I could pinpoint where in the plant what type of injuries were occurring. This let me focus my on-site time and identify the causes of the injuries.

Other concerns surface when loss runs are designed for the corporate bean counters.

Financial Overkill

Problems also arise when CIOs focus loss runs exclusively or predominantly on financial data. This is especially true when insurance customers want to use the printouts as loss-control tools. According to Price, Most loss runs are designed for financial information only. This is a serious mistake. It causes insureds to focus on costs and not injury reduction. The information is nice to have, can be used by CFOs to guess how much they will pay/get back at the end of the year, but the dollar numbers really dont help the poor safety person in the plant.

What I would like to see is a cross-tabulation of some of the data in addition to the dollar figures, Price continues. Something such as body part and injury type all indexed by departments. Unfortunately, this takes time and costs money for the MIS types to figure out. Thats why I never could get management to make the changes I wanted. All of the revisions I have seen come out of MIS departments for the past 10 years are simple format changes to existing financial information. Nothing substantive.

Speaking with consumers of loss runs, therefore, four tips for CIOs and tech professionals emerge:

Adopt the KISS principleKeep It Short and Simple. The worlds greatest loss run is worthless if no one can decipher it.
Customize. Let clients customize, within reason. Some clients want data sliced and diced one way; others want it organized and formatted differently. One client might want to know whats causing all the fleet vehicle accidents; another might want to know the time of day when most accidents occur. These nuggets may have profound safety and loss-control implications for all sides involved.
Offer it online. More and more customers are accustomed to being able to retrieve their financial information online. This includes checking- and savings-account data, bills, 401(k) statements, brokerage statements, and the like. Why not loss runs, too, on a read-only basis?
Seek customer/client feedback. Incorporate that feedback. Make clients a partner in the loss-run design process. Ask them: What are the three things you would change about our loss runs, if you could?

Design Considerations

If an organization opts to self-insure, or to absorb a high self-insured retention (SIR), reliance on insurer-generated loss runs diminishes or disappears. At this point, organizations internalize the loss-run function. Thus, risk managers may find themselves as loss-run consumers for those exposures they insure and loss-run manufacturers for those exposures they retain or self-insure. Then the risk manager must design a report that is intelligiblenot only to risk professionals but to upper management and all divisions. The latter is a particularly keen consideration when loss-allocation systems tie division operating resultsand manager bonusesto losses. At that point, the manager even may start empathizing with insurers that fielded gripes about loss runs being hard to read, reserves set too high, or numerical errors in printouts.

Another option for high SIR or totally self-insured clients is to externalize the task by hiring a third-party claims administrator to handle claims. Typically, part and parcel of the TPAs job duties include generating loss runs. The ability to generate prompt, error-free, and useful loss runs is a given on any risk managers shopping list when weighing proposals of competing TPAs.

In designing state-of-the-art loss runs, regardless of whether the focus is workers compensation claims, automobile, homeowners, or contractors liability, include certain bedrock data: policy number, claim number, claimants name, insured, loss date, brief loss description, damages/injuries, reserves, payments, jurisdiction. These often are must haves for any information-conscious company aiming to use loss runs as tools for managing risk and preventing future losses.

In the nice-to-have category: litigation status (pre-suit, in-suit, on appeal?), name of claimants/plaintiffs counsel, name of defense counsel, subrogation potential, punitive damages alleged, coverage questions, name/address of TPA handling file, diary date.

Of course, soon you can have a loss run that consumes a page or more per claim. At this point, detail may overwhelm loss-run managers and become mind numbing. In designing loss runs, a delicate balance exists between comprehension and detail. Sometimes less is more.

Too much information, and customers are baffled or turned off.

Too little information, and users consign loss-run printouts to the dumpster.

Loss-run design may not be the most exciting part of being a CIO or tech professional, but the state-of-the-art printout can become a useful tool for companies sifting through data to save money by preventing future losses.

Turn Loss Runs into Safety Weather Vanes

If youre a CIO seeking a back to basics grounding in loss runs, here are five items to look for on loss runs from the insurance company:

1. Specificity. Claims involving hurt back or injured hand will not help clients. (Keep in mind the people completing the accident reports may not be the same users who review the loss runs or use them as safety tools.) Determine exactly what happened in the incident.
2. What body part was injured? Left leg or right? Are multiple areas of injury captured or just the most prominent.
3. What happened? Get specific. If someone slipped at a construction site, did he fall or not? Did she slip on ice, grease, or water?
4. What was the cause? Within the bounds of brevity, the loss run should state reasons.
5. What was the injury? Was it a laceration, a bruise, a burning sensation in the lower back?

Kevin Quinley, CPCU, ARM, AIC, AIM is senior vice president, Risk Services, Medmarc Insurance Co., Chantilly, Va. He can be reached at kquinley@medmarc.com or www.kevinquinley.com.