Declining economic conditions are leading to an increase in insurance fraud. This is according to Paul van der Merwe, Director of Knowledge Integration Dynamics (KID). “Insurance fraud is costing the average South African household thousands of rands a year in the form of higher premiums and higher prices for goods and services. As a result there is an urgent need for proactive fraud tracking and prevention,” maintains van der Merwe.
"Sharp drops in stock markets and record levels of personal debt are underlying conditions for a potential increase in insurance fraud. Yet the public needs to understand that fraud increases the costs of their policies and consumer goods and services."
Van der Merwe says in South Africa it has recently been reported that short-term insurers pay about R1,2 bn in fraudulent claims every year. The industry pays out R15 bn in claims annually, and according to recent statistics released by the SA Insurance Association (SAIA), between 8 and 35% of those claims are fraudulent.
Barry Scott, CE of the South African Insurance Association, maintains insurance companies have had no way of verifying whether clients have previously committed insurance fraud, whether they are insured with more than one company and are enriching themselves by filing several claims for one loss, or whether they are in the habit of company-swapping to escape being found out. "As a result, premiums are pushed higher and higher and honest clients end up funding the dishonest ones," says Scott.
Analytic solution
Finding and stopping fraud on such a large scale requires a heavy-duty analytic solution, van der Merwe continues. "The South African market has dire need of analytical applications - applications that can take data from disparate sources and present an overview of all the patterns and relationships hidden in the data to assist in discovering and highlighting trends, potential new business areas and fraudulent transactions quickly and efficiently."
As an example van der Merwe says, fraud investigations that once took days at Prudential Insurance Company of America, have been reduced to about 15 min through the use of analytical software. The system allows the company to first cross-reference a claim with its own claims records, and then cross-reference it against 135 million other property, liability, auto, workers' compensation and other claims in the database.
In one case, a California auto claim was filed with Prudential and entered into the software system to cross-reference it against other Prudential claims. Immediately, the software showed that the claimant had used two different addresses when dealing with the insurer.
The information drew the interest of the investigator, who then went a step further in the cross-referencing process and discovered that the claimant had filed five previous claims with Prudential and had used two social security numbers, five addresses and four telephone numbers.
One more click of the mouse revealed that the claimant had links with 17 other claimants. For example, they had used one of his addresses or phone numbers or they were involved in one of his claims as a passenger or the other driver in an accident.
When the software was used to access the millions of claims with other insurers in the database, an even more suspicious picture emerged. The claimant was involved in 69 claims with 19 other insurers and had reported another four social security numbers, 25 addresses and three telephone numbers. When all that data was used to make a further search, it revealed that the single claimant was linked in some way with 148 other people. The information yielded by the software was reported to the relevant state authorities, who are conducting a criminal investigation.
The software, NetMap, was developed by Alta Analytics, represented locally by KID. "Netmap provides detailed analysis and data filtering functionality by identifying and graphically displaying patterns, trends and relationships contained in corporate data," says van der Merwe. "Managers can easily spot and dynamically filter this information to identify trends and patterns in their business processes.
NetMap analyses data from any source - an insurer's own data or an industry's comprehensive all-claims database. It reveals and presents potentially fraudulent claims in an intuitive, visual context that investigators understand immediately. "By combining the vast claims information on an insurance company's database and the searching and cross-referencing capabilities of the software, insurers are able to reduce considerably the costs of fraud," van der Merwe concludes.
For more details contact details Paul van der Merwe, Knowledge Integration Dynamics, on tel: (011) 787 0822, e-mail: [email protected]
© Technews Publishing (Pty) Ltd. | All Rights Reserved.
printer friendly version