The Value of Public/Private Partnerships in Preventing Fraud in Federal Government Databases

Topics: Artificial Intelligence, Cybersecurity, Data Analytics, Efficiency, Fraud, Government, Government Fraud, Medicare Fraud, Thomson Reuters


Verifying the authenticity of several million names in a government database is an enormous task. But that is exactly what one of the federal government’s largest healthcare agencies recently did.

Using investigative software, the agency authenticated 60 million names and addresses in time to meet a strict deadline for mailing new benefit cards. The project successfully prevented cards from being mailed to incorrect addresses, significantly reducing opportunities for fraud and abuse.

Indeed, fraud prevention was one of the main reasons the federal government decided to re-design its healthcare benefit cards in the first place. For the past 50 years, benefit recipients have been identified by their Social Security number, which was printed on their Medicare or Medicaid card. In the era of the internet, however, identity thieves steal, buy, and sell Social Security numbers to commit all kinds of fraud, so the system is not as secure as it once was.

Consequently, federal benefit programs have become a favorite target for fraudsters. In 2017, the Government Accountability Office estimated that Medicare alone had recorded $52 billion in so-called “improper payments” (down from $95 billion in 2016); and that same year, the U.S. Department of Justice prosecuted several large government fraud cases, recovering more than $3.7 billion in ill-gotten gains.

To create a more foolproof system, the federal government developed a new numbering protocol for benefit recipients and re-issued individual cards with no personal information on them other than the recipient’s name and their new government ID number. The new cards — 60 million of them — had to be mailed through the U.S. Postal Service. Before the cards could be sent, however, the agency responsible for distributing them had to make sure the names and addresses in their master database were correct and up-to-date.

Data Hygiene a Key Priority

To verify the names and addresses, the agency undertook what’s known as a “data hygiene” project — a systemic data review that compares information in a pre-existing database against external data sources to certify that the information is correct. Both the scale of the project and the need for extreme accuracy made it necessary for the agency to seek help from the private sector for independent verification. The agency used a proprietary public-records databases within the private sector to determine if there is a match.

For this project, identifiable matches were put in a “yes” pile, and new cards were mailed. But names that ended up in the “no” pile presented an additional challenge. To correct and verify these addresses, more information was needed. Working together with the agency’s own data analysts, private sector analysts helped develop an alternative method for identifying why the existing contact information was incorrect and how to fix it.

This extraordinary public-private collaboration resulted in an efficient and accurate process that was capable of verifying up to three million names per week. And in the end, all the new benefit cards were mailed months before the official deadline.

A Public/Private Win-Win

The success of this project offers compelling evidence that well-executed public/private partnerships can help improve the effectiveness and cost-efficiency of many government programs.

The accuracy of results saved the federal government potentially billions of dollars by minimizing the possibility of cards falling into criminal hands. With a mailing of 60 million people, even a small fraction of incorrect addresses could result in thousands of cards being sent to the wrong people. And because they weren’t, that meant the government did not have to expend nearly as many resources on fraud prevention, investigation, and follow-up.

To learn more about this extraordinary public/private partnership, download the complete case study here.