From pharmaceutical, healthcare and financial claims, to insurance and product warranty claims, identifying and monitoring fraud is a priority for many organizations. Not only is a monetary loss at stake, but there is also potential damage to a company’s brand, reputation, and trust. However, it can be a challenge for companies to understand and stay ahead of ever-evolving fraud risks and proactively identify and investigate active threats—let alone to navigate the new world of data analytics to assist in the process.
A study by The Association of Certified Fraud Examiners’ (ACFE) based on 1,483 cases of occupational fraud revealed that the typical organization loses 5% of revenues each year due to fraud. On a global scale, this translates to losses of approximately $3.7 trillion. The study also revealed that the median loss caused by the frauds was $145,000 and 22% of the cases involved losses of at least $1 million. According to James D. Ratley, ACFE President and CEO, “the analysis of these cases provides valuable lessons about how fraud is committed, how it is detected and how organizations can reduce their vulnerability to this risk.”
The return on investment for effective fraud detection solutions can be transformative for a business. Fraud prevention, recovery and restitution results in immediate savings, and once teams begin to use fraud solutions successfully, the results can provide insight for their long-term strategic planning. Consider these examples:
- Claims Fraud: The U.S. Postal Service Office of Inspector General deployed a custom fraud detection solution to identify and prioritize the investigation of questionable contracts and healthcare claims resulting in over $11 million in recoveries, restitutions, and cost avoidance in the first year, and the system has since contributed to $920 million in findings.
- Healthcare Fraud: A coordinated national healthcare takedown led by the Medicare Fraud Strike Force in 36 federal districts resulted in charges against 301 individuals, including 61 doctors, nurses and other licensed medical professionals, for their alleged participation in health care fraud schemes involving approximately $900 million in false billings.
- Workers’ Compensation Fraud: The Department of Labor Office of Inspector General deployed a predictive model to detect fraud in Office of Workers’ Compensation Program data. The model reveals abnormalities in the claims data that can be used to form the basis of future audits, and the system includes visualization tools to enable auditors to explore model results. Consolidating data sources into a centrally managed data repository and interactively visualizing information reduced case research times from days to hours, resulting in considerable savings.
- Service Provider Fraud: A Medicare and Medicaid Dental Insurance provider deployed an advanced analytics prototype to detect potential provider fraud, waste, and abuse in their network of service providers. The solution increased the fraud detection rate from 5% to 48% for the top 50 riskiest service providers identified by the model, and enabled the client to more efficiently target suspect claims for investigation, improving the efficiency of their investigative resources.
- Product Warranty Fraud: A major computer and electronics product manufacturer needed to reduce return rates and proactively identify cases of warranty fraud or claim anomalies indicative of fraud among its service provider base. They deployed a fraud analytics solution that performed intelligent, multi-sourced data analysis, returning single and consistent entitlement information that authenticates the claim. The customer credited the system with over $67 million in cost savings over five years.
Fraud detection solutions should look at the entire business sequence, uncover patterns of suspicious behavior, and provide actionable alerts to the organization. The many different actors and schemes involved, varying state regulations and oversight, and compliance with privacy laws can all contribute to the challenge of detecting and preventing fraud. Even advanced fraud detection algorithms are best at catching only the most extreme offenders. But, as shown in the examples above, eliminating those outliers can have a dramatic impact on fraud prevention and savings. Moreover, attention to fraud and abuse tends to deter other bad actors.
