As the opioid crisis has grown, so too has the number of opioid use disorder treatment programs. But in this widely unregulated treatment system, some “sober homes” and other programs prove to be scams, taking on patients to milk insurance without providing real treatment.
Insurers may not be able to look across national data to find the telltale signs of this kind of fraud, but School of Public Health researchers can, with help from a top former federal healthcare regulator, a Washington, DC-based law and consulting firm, and some undercover “mystery shoppers.”
Together with support from the Laura and John Arnold Foundation, part of Arnold Ventures, four professors from the Department of Health Law, Policy & Management are working on a two-year project to identify insurance claims patterns that point to substance use disorder treatment scams; they will then build a tool for insurers and regulators to understand just how big the problem is, and how to stop it. Law and consulting firm Faegre Baker Daniels and insurers will then test the model by having mystery shoppers contact the potentially fraudulent programs, posing as people interested in treatment.
“This problem affects everybody,” says Melissa Garrido, research associate professor of health law, policy & management, and the lead researcher on the study. It harms people seeking treatment first of all, she says, but insurers, who are not able to identify every instance of this kind of fraud, or even tell how much of it is happening, will have to make up their losses somehow. “Everybody’s rates go up,” she says. “The fraud wasn’t targeted at the general public, but the general public suffers from it.”
No one knows all of the tricks involved in these scams, or how many fraudulent providers there are, but Garrido says there are a few known patterns. “You don’t need to run a urinalysis once a day or multiple times a day,” she says. (Opioids remain detectable in urine for about a week.) “Unbundling” lab tests can also be a red flag, where the patient—and thus the insurer—is charged for separate tests from a single blood draw, when a single bulk lab test would usually suffice.
“We’re looking for outliers,” says Garrido, whose expertise is in developing best practices for using large, healthcare-claim observational data sets, and who is also associate director of the Partnered Evidence-Based Policy Resource Center (PEPReC) at the Boston VA Healthcare System. “We can’t really assign intent to behavior, but we can see that someone is in the top 10 percent of people who are ordering frequent urine tests and the top 25 percent of people who are doing all these unbundled tests, so maybe this is someone who warrants closer examination.”
Garrido and her colleagues are looking at insurance data covering 50 million people, including coverage bought on Affordable Care Act exchanges. “Most insurance companies are able to look at potentially fraudulent activity within their own system, but when they kick a provider out, that doesn’t stop that provider from going and doing the exact same thing with another insurance company,” she says. “By using data that allows us to look across systems, we might be able to see other patterns where providers just barely escaped detection at one insurance company and moved on to another—a broader pattern of behavior.”
But before tackling the data set, the team is conducting a combination of literature and media reviews to find out what is already known about this kind of fraud. In this phase, they are also interviewing key stakeholders—people working in regulation, law enforcement, and opioid use disorder treatment—in Florida, California, Texas, New York, and Pennsylvania; all are states with areas of high per-capita opioid use disorder rates and large numbers of opioid treatment programs. The interview portion of the study is being led by David Jones, assistant professor of health law, policy & management. Steven Pizer, associate professor of health law, policy & management and PEPReC’s chief economist, will then work with Garrido on the claims analysis and developing the algorithm.
After the interviews, data analysis, and mystery shopping, the goal of the project will be to get the resulting tool out to as many entities and individuals as possible, Garrido says, so that it can inform new regulations and laws. Health economist Austin Frakt, associate professor of health law, policy & management, director of PEPReC, and a writer who regularly contributes to the New York Times’s Upshot column, will work with the team to disseminate the research findings to a broader audience.
Ultimately, Garrido says, the hope is that this project will help create a balance, where insurers won’t have to make it too hard for people with opioid use disorder to receive care—and it really will be care.
“These are patients who are at risk for really awful outcomes, including fatal overdose,” says Garrido. “They think they’re in a legitimate program, but they’re not receiving treatment—they’re just being taken advantage of.”