Texas Uses Data Visualization to Combat Medicaid Fraud

by   |   August 15, 2013 5:43 pm   |   0 Comments

LYNXeon Medicaid data visualizaiton illustration 650x630 Texas Uses Data Visualization to Combat Medicaid Fraud

The Texas Office of Inspector General used the LYNXeon visualization tool to track connections among government payments, health care providers and Medicaid recipients. Image above is an illustration, courtesy of 21CT.

Pinning down how much taxpayer money is lost to Medicaid fraud is difficult simply because the successful frauds go undetected. But the U.S. Government Accountability Office estimated that $32.7 billion (or 10 percent) of state Medicaid payments made in 2007 were improper. Other estimates are much higher.

It’s no wonder why. A huge federal program such as Medicaid — which provides health and medical services funding to poor individuals and families — involves a byzantine network of care providers, medical institutions, pharmacies, drug manufacturers and patients spread across 50 states.

Consequently, there are a number of schemes used by providers and patients to defraud Medicaid. Among them are:

  • Billing for services not rendered
  • Double billing
  • Billing for more hours than there are in a day
  • Substituting generic drugs
  • Billing for more expensive procedures than performed
  • Kickbacks to nursing homes
  • Personal expenses in nursing home Medicaid claims



“People who are committing fraud spend all day, every day thinking about it. They come up with new ideas, they come up with ideas about how to hide their tracks. That’s their job, it’s what they do,” says Jack Stick, deputy inspector general for enforcement for the State of Texas’s Office of the Inspector General (OIG). “But people whose job it is to fight fraud do it during a regular work day. So we’ve got to think faster than they do, think better than they do, and leverage technology.”

Related Stories

Oklahoma IT consolidation sets stage for analytics use cases.
Read the story »

Analytics, IT firms promote business benefits of federal data bill.
Read the story »

Philadelphia chief data officer seeks economic growth via open data.
Read the story »

National Hurricane Center predictive models improve forecasts.
Read the story »

The technology the Texas OIG has chosen to leverage in its battle against Medicaid fraudsters is LYNXeon, a data analysis and visualization software platform designed to collect, analyze and visualize data for investigative analytics and pattern detection.

LYNXeon is made by 21CT, an Austin, Texas-based company that began in 1999 as an incubator for the U.S. Department of Defense and intelligence communities. Customers of 21CT include the branches of the U.S. armed forces, the Department of Homeland Security, and several Fortune 1000 companies.

Stick says the Texas OIG, which currently has 90 investigators tracking down Medicaid fraud in a state with more than 26 million residents and one of the highest poverty rates in the country, began loading its billions of lines of data into LYNXeon last January.

“By March we were ready to go live with 10 percent or 11 percent of the total available data,” Stick says. “They identified $20 million in potential overpayments based on just that limited amount of data.”

Once LYNXeon flags possible Medicaid fraud, Texas OIG “goes in and lays hands on the cases to do the actual investigation,” Stick says.

The power of LYNXeon, which was launched in 2004, resides in its ability to turn huge amounts of data into visualizations that allow users to see patterns they otherwise may have overlooked. For an understaffed agency charged with enforcing compliance with a massive federal program, data visualization is an invaluable tool.

LYNXeon runs data through four stages, according to Kyle Flaherty vice president of marketing for 21CT.

“It starts with machine learning models and algorithm scoring,” he says. “This spits out a bunch of leads for investigators.”

From there LYNXeon filters data through targeted queries designed to get information about specific behaviors, such as unusual billing patterns or purchases.

Stages three and four involve visualizations, he says.

“We use link analysis to determine context between people, places and things,” Flaherty says. “By visualizing the connection paths, LYNXeon allows investigators to expand and pivot off this information and get to the root of how they’re perpetrating fraud.”

Finally, pattern and social networking analytics “can really arm a good investigator to find something he never could before,” he says.

Stick says this is especially true when you use a wide range of data.

“What we’re doing with LYNXeon is adding in Medicaid data, Medicare data, Dun & Bradstreet data,” he says. “So if we find that a provider bills for a lot of procedures, but Dun & Bradstreet  tells us that they never turn the electricity on, that’s a pretty good indicator that there’s waste, fraud and abuse going on.

“We also can look at the Medicaid payment data and put it into context,” Stick says. “We can compare that provider to his or her peers, we can look to see if they’re in a building that’s physically large enough to house what they’re doing.”

LYNXeon also gives Texas OIG the ability to investigate recipients of Medicaid and other benefit programs.

“We can see if your EBT (Electronic Benefit Transfer) card is active in Dallas one day, but you’re receiving Medicaid services in Houston on the same day,” Stick says. “And we can track retailers that are fraudulently buying electronic benefits for pennies on the dollar and then redeeming them for full value.”

Nearly six months since going live, Stick says LYNXeon has identified more than $180 million in potential Medicaid overpayments for Texas OIG to investigate.

“If only a fraction of what we’ve identified through LYNXeon proves to be waste, fraud and abuse, and we recover that money or at least avoid spending that money in the future, we will already have paid for LYNXeon,” says Stick. “It’s by far the best money I have ever spent in government.”

Contributing Editor Christopher Nerney (cnerney@nerney.net) is a freelance writer in upstate New York. Follow him on Twitter: @ChrisNerney.

Tags: , ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>