Fraud Prevention in the National Healthcare

0
36

By Alex Tate

According to CMS statistics, 2016 is the first year where U.S. healthcare expenditures are expected to surpass $10,000/person.  Health spending growth is expected to outpace GDP growth by 1.3% per year, expanding from being 17.5% of GDP in 2014 to 20.1% of GDP by 2025 (more than 1 out of every 5 dollars!).  This takes into account the bump likely to occur due to the Affordable Care Act.

With so much money at stake, how likely is it that by “following the money,” you will find medical fraud, because (as Willie Sutton was rumored to have said) that’s where the money is?  With so much money being handed out by insurers, it must be a very tempting target.  So let us now ask the question: Can anomalous behavior be detected from a data warehouse through proper use of business intelligence analytics?  The answer, of course, is a resounding yes.   The National Health Care Anti-Fraud Association (NHCAA) estimates that healthcare fraud losses nationwide amount to tens of billions of dollars annually.  And large part of the fraud solution includes having a data warehouse testing claims data, using Business Intelligence to identify fraud.

On June 18, 2015, it was announced that 240 individuals had been arrested in a 17-city operation by the FBI, Department of Health and Human Services, Department of Justice (DOJ), and local police, representing around $712M in Medicare fraud false billings, a new record in terms of arrests and dollars lost.  Since 2007, the Medicare Fraud Strike Force has charged over 2,300 people for over $7B in Medicare fraud.

How easy is it to identify medical fraud?  Several things complicate the process.  Obviously, none of the fraudulent claims are outright labeled as such.  Fake claims by a single physician, nurse, or practice may be spread across multiple insurers, making the reimbursement increases less notable.  But before we can strategize on how to detect fraud, we must identify different types and address them separately.

  • Medical identity theft: Stolen patient identifiers are used to bill services that are never provided.  Since there’s no real contact with the insured, the medical provider would have no way of knowing about provider or plan changes, or extreme patient changes (name change, address change, gender reassignment, death, etc.) that could raise a red flag for the insurer.
  • Kickbacks: This involves giving patients incentive to being active participants in the insurance deception.  Business intelligence may be able to detect doctors with, say, a large number of whiplash victims most of whom take unusually aggressive and lengthy treatment plans, with office visits that may seem too regular datewise.
  • Billing for Services or Items Not Furnished: Ever look at a bill and find a piece that should not be there?   These are just outright lies, charged items that never happened and don’t belong.  These might be billed to victims of medical identity theft or kickback patients.  This also includes duplicate billing, where a legitimate charge has details changed then rebilled, as a way of double-dipping the actual expense with an additional fake occurrence, like an extra therapy session.  If all of an office’s insurers talked to each other, Business Intelligence might see how overly-productive that office is.  A single insurer may still be able to determine a short list of probable fraud.
  • Billing for Unnecessary Services or Items: The patient is given a medical supply they did not need and maybe did not even ask for.  Perhaps this is a medical device, an unneeded office visit, or unnecessary lab tests that provides a nice reimbursement.  A major increase in these items by a particular physician might set off a red flag with the insurer, especially if the treatment plan is non-standard for the diagnosed condition, or if a rare condition becomes very prominent at this office.
  • Upcoding: This involves a bait-and-switch where a minor service or item is exaggerated as if it were a more costly service or item.  A new premium medical device might be replaced by a lesser-quality knock-off device, or an office visit might be billed as a therapy session.  Business intelligence may pick on the change in which this office has been treating patients.
  • Unbundling: Items normally billed together (like a panel of blood tests on a blood sample) are billed a la carte.  Imagine buying a new car, vs. buying all of the parts for that same new car.  Since an attempt to charge this way would tend to go through the same insurance carrier (unless there’s secondary coverage), the insurer should be able to catch a lot of these.

Proper investigation is still needed in each of these cases; it is entirely possible that word of mouth increases a doctor’s treatment of patients with higher-end needs.  Probability does not guarantee wrongdoing.

Author Bio:

Alex Tate is a health IT fanatic who is passionate about technology and its revolutionary impact on the healthcare industry. He adds value to the healthcare community by providing answers to problems faced by the providers. He is always hunting hot topics and opportunities that will open new dimensions in the field of Health IT.