How tech will help us win battle against insurance fraud

Last year, the media highlighted an insurance fraud case where a husband and wife duo decided to fake the husband’s death. They forged a death certificate so the wife could claim compensation from their insurer.

On the heels of this was another case involving 11 individuals. They were thought to have conspired to defraud a local insurer of Sh15 million by forging its medical cards and hospital prescription pads, and using them to acquire drugs from various health institutions.

Then there was the insurance investigator who was accused of pretending to be in a position to influence motor insurance claim payments.

Insurance fraud is a consistently growing concern for insurers in Kenya, with an increasing share of claims payments attributed to the vice.

According to the Insurance Regulatory Authority’s fourth quarter report for 2015, 106 suspicious cases were reported to the Insurance Fraud Investigation Unit (IFIU) last year, up from 87 cases in 2014.

The amount lost increased from Sh102.76 million in 2014 to Sh366.90 million. Of the 106 cases reported, motor underwriting led the list with 42 cases, followed by agent/broker fraud at 25 and then medical with 17 cases.

Risk survey

Insurance fraud is prevalent across the entire insurance value chain, with the most affected areas being claims and underwriting. But as technology evolves and fraudster tactics become more sophisticated, insurers are now having to deal with other forms of fraud, including internal fraud, money laundering and cyber fraud.

The KPMG East Africa Insurance Fraud Risk Survey 2015 recommends better assessment of risks at proposal stages, and improvement of internal controls to curb the threat of fraud risk. Another key focus area is data analytics.

In the past, insurance companies have relied heavily on fraud investigators to look into suspicious cases and determine if fraud has occurred. But as our world and technology evolves, we are witnessing a new breed of fraudsters (and fraud rings). This demands that the insurance industry shift to improved detection initiatives.

To combat fraud more effectively, the ability to collect and analyse huge volumes and varieties of data is essential. There are new technology tools and techniques in the market that can help insurers uncover complex or organised fraud activities using both structured and unstructured data.

These include data analytics, predictive modelling, link analysis, automated red flags/business rules, and geographic data mapping.

Predictive modelling enables insurers review historical fraudulent claims, and identify factors and elements that can help prevent future fraud. The main goal is detecting fraud early enough in the claims process.

Link analysis examines relationships among claims, people and transactions, thus helping link different players and identify the extent of the relationships between the parties, then giving off data that can be used to define indicators that point to possible fraudulent activity.

Anticipate trouble

Automated red flags/business rules can be inbuilt into an insurer’s core IT systems and are instrumental in helping anticipate certain types of suspicious claims activity based on past fraud. They identify anomalies or irregularities during the processing of claims.

Geo-mapping can help insurers evaluate risk and exposure during underwriting and claims processing. For example, an insurer can confirm that an incident actually occurred where the customer claims it did.

There are many other tools available in the market, but it is important to note that to take full advantage of any or even all of these technological capabilities, insurers must first address legacy technology issues and inefficient processes that have proven to be a great hindrance in the war against fraud.

Adapting modern technology significantly improves operational workflow and helps manage claims investigations faster.

With automated workflow solutions, insurers can track daily activities and expenses, and get at-a-glance reports to help flag any suspicious activities.

With modern systems, underwriters are also able to confirm information by comparing applicant reports with information from other databases and public records. This way, they can confirm prior coverage, discover undisclosed information and eliminate policy application fraud before it occurs.

The writer is CEO of software firm Turnkey Africa.

bizbeat@standardmedia.co.ke