In April 2013, the Insurance Fraud Bureau of Massachusetts (IFB), the automobile and workers’ compensation insurance fraud investigation agency, and the Automobile Insurers Bureau of Massachusetts (AIB), a member-owned auto insurance industry service organization, prepared and released a report that offered more than the quantification of insurance fraud investigation and prosecution. The report, “The Community Insurance Fraud Initiative (CIFI), A 10 Year Retrospective,” provided a measure of CIFI’s impact on insurance fraud activities, claims, and premiums as well as several takeaways that might be useful in other fraud prevention programs around the country.
CIFI arose from a staged accident that killed a grandmother in Lawrence, Mass., in September 2003. Subsequently, the IFB, Lawrence Police Department, and Essex County District Attorney’s Office created a task force to identify, investigate, and prosecute staged accidents in Lawrence. Within three months, the task force charged 17 people with insurance fraud—a significant number were involved in staged accidents.
The success in Lawrence led the AIB and IFB to develop a plan to identify other problem areas in Massachusetts. In that process, they noticed that, prior to 2003, Lawrence had the highest injury-to-accident ratio with 141 injuries to every 100 accidents. The ratio was nearly four times the statewide average of 38 injuries per 100 accidents. The AIB and IFB concluded that the only plausible explanation for the high injury-to-accident ratio was staged accidents, which tend to be high-occupant, multicar, injury-related accidents. Using this metric, the AIB and IFB identified other cities with high ratios and implemented the CIFI program in those areas with the cooperation of local law enforcement and prosecutors. Ten years later, they measured the results of their efforts.
The AIB and IFB quantified CIFI’s impact on fraud in five ways. First, they tallied the number of people charged with insurance fraud (1,917), which indicated that fraud existed in the CIFI areas.
Second, they examined the injury-to-accident ratio, which decreased in all CIFI areas as well as non-CIFI areas. The AIB and IFB concluded that the decrease in non-CIFI areas resulted from the “halo effect” created by its public awareness campaign and possibly the arrests of fraud perpetrators who operated in multiple areas.
Third, they looked for a decrease in the billing practices of “high-volume providers” in CIFI areas. To do this, the AIB mined its Detail Claim Database (DCD), which has tracked auto injury claims in Massachusetts since 1994, for medical providers whose billing patterns suggest unusual practices or those providers who are high-volume auto accident billers. For example, Lawrence had 22 high-volume chiropractors and physical therapists before initiating the CIFI program and seven chiropractors and physical therapists with significantly reduced billing after the CIFI program.
Fourth, they looked for a decrease in the percentage of PIP claims with attorney involvement, because fraudulent PIP claims often involve claim buildup with the assistance of an attorney. They found that attorney involvement in PIP claims dropped 4 percent in CIFI areas and 6 percent statewide.
Fifth, the AIB and IFB looked for a dollar reduction of claims and premiums in CIFI areas compared to the entire state. They calculated the average annual savings by taking the cumulative premium savings divided by the number of years of savings and the current number of vehicles in each community. They found that in CIFI areas, policyholders’ average annual savings were $185. By comparison, policyholders in non-CIFI areas realized an average annual policy premium savings of $139. The AIB and IFB concede that they cannot attribute premium savings directly to the CIFI programs; however, because CIFI reduced fraudulent activity—measured by reduced injury-to-accident ratios—they concluded that CIFI significantly contributed to premium savings indirectly. Therefore, the CIFI program improved conditions for both insurers and insureds and deterred fraud throughout the state.
The CIFI report illustrates several points. First, the insurance industry, state agencies, local law enforcement, and prosecutors together can reduce, perhaps even eliminate, certain fraud activity in targeted areas. Each entity plays a crucial role in the overall success of the fraud prevention effort. For example, the AIB interpreted data from the DCD to identify and target fraud prevention efforts in other areas of Massachusetts. While the DCD is a database unique to Massachusetts containing useful claims information including injuries, treatment, billing, payments, and individuals involved in claims, there may be a similar database in your jurisdiction that could yield useful information for identifying or targeting fraudulent activities.
Second, targeted investigations and public awareness campaigns have beneficial effects on adjacent nontarget areas as evidenced by the halo effect. For instance, targeting motor vehicle fraud or workers’ compensation schemes in a city may improve the insurance climate in a state or region.
Third, data mining for fraud indicators is an effective means to identify potential fraudulent activity and target fraud prevention efforts. The trick is to have the right database or databases and a person or two who know how to interpret the data.
Finally, targeted fraud prevention efforts lead to premium savings over time. In the end, premium savings is the goal.