Data Analytics for Insurance
Insurance fraud can be committed at various points in the claims transaction process by different parties inclusive of claimants, policyholders and other third-parties.
The Federal Bureau of Investigation (FBI) states that “the total cost of insurance fraud is estimated to be more than $40 billion per year. That means insurance fraud costs the average family between $400 and $700 per year in the form of increased premiums.”*
While many insurance companies accept claims fraud as an unfortunate cost of doing business, it doesn’t have to be that way.
In an industry known for tight regulations, new market entries and overall tough competition, insurance companies need sophisticated technologies that will allow them to stay ahead of threats. Alessa employs a combination of business rules and predictive analytics to detect and prevent fraud throughout the claims process, spanning multiple lines of business – including auto and health.
With up to 75 percent of claims data being unstructured and in siloed business systems, Alessa is able to present a consolidated view of all the requisite data sets, detect anomalies, and trigger alerts to personnel within the business, to resolve the issues as soon as they occur. The solution is able to independently monitor all facets and variables involved in the insurance process such as transactions, customer behavior and third-party data across multiple claims and lines of business.
Here is a list of analytics (rules) that can be employed in order to ensure compliance to policies and prevent fraud.
- Identify duplicated members and dependents
- Invalid dependent(s) such as multiple spouses attached to the same member
- Report excessive billing of same diagnosis, same procedures
- Determine excessive number of procedures per day or place of service per day/per patient
- Identify multiple billings for same procedures, same dates of service
- Analyze data for mismatched services to diagnosis codes
- Test for doctor and patient with same billing address
- Claims that are split to bypass adjudication limits
- Incomplete claim information submitted but claim still being processed for payment.
- Changes in claim details after the initial claim is made
- Claims entered by unauthorized users
- Claims approved for payments that exceed thresholds
- Identify duplicated electronic, outsourced or internally keyed claims
- New claims from suspended providers
- Late claims approved for payment
- Identify overpayment of claim based on business rules
More fraud scenarios and analytics can be found here.
Life Insurance Fraud
- Review transaction payments containing more than one type of payment type
- Determine patterns of overpayment of premiums
- Review payments from highly suspect banks or countries
- Test if customer is from a non-cooperative country or territory as identified by the international Financial Action Task Force (FATF)
- Report purchases of multiple products in a short period of time
- Analyze beneficiaries with multiple policies
- Analyze employees that are beneficiaries
- Determine agents/brokers with high numbers of death claims
- Calculate benefit payments paid for lapsed policies
- Find policy loans that are greater than face value
- Report any unauthorized policy changes
- Compare commissions paid based on recalculation and paid amount
- Generate reports on paid-up additions, dividends on deposit, policy loans, etc.
- Summarize monthly transactions by new business, death, endowments, etc.
To learn more about scenarios indicating insurance agent fraud and how to combat it, visit here.
- Calculate average yield by investment type and show detail variances
- Calculate total accrued interest receivable by type of security, agent, etc.
- Compare computed interest to amount actually received on investments