Robin's Fraud Control Toolbox

Robin's Fraud Control Toolbox

Healthcare Fraud Control Articles

There is no single tool that will identify all fraud. Here are some of the tools you’ll need in your toolbox. Add one or two tools at a time to build a full defense against fraud:

  1. Basic edits and audits
  2. Computer Network Security
  3. Rules Based Post-payment Analysis
  4. Random Audits
  5. Outside Data Sources & Social Network Analysis
  6. Geographic Information Systems (GIS)
  7. Predictive Modeling
  8. Advanced Prepayment Edits
  9. Biometrics
  1. Basic edits and audits act as a simple first defense to the easiest types of fraud. They usually include hundreds or thousands of logical checks to make sure that the information submitted on a claim makes sense. This is like putting a lock on your front door. It deters people who aren’t too determined to commit fraud, but it doesn’t do much to stop the real criminals. Most claims payment systems include these kinds of edits. In addition, there are system add-ons available that apply edits specifically designed to reduce fraud. If the system you use is easy to modify, you’ll have more opportunity to limit fraud losses. I’ve seen organizations lose millions of dollars, because it took several months to implement new edits.
  2. Computer Network Security is required by HIPAA and necessary to keep your data safe. Network security ranges from simple measures like passwords to sophisticated procedures like having security experts monitor all systems, so they can immediately identify a hostile break-in. With information from your claims processing system, fraud perpetrators could commit fraud that would be very hard to identify with most tools. Here are just three examples:
    • David Black, a hacker, breaks into your network and changes the mailing address of a doctor named David Black. Now the hacker receives checks that should have gone to the doctor.
    • An organized crime ring hires a hacker to steal 1,000 patient ids from your system. The crime ring uses 10 provider ids to create phony bills for these patients.
    • Charlie copies 10,000 claims from your system, creates 20 new providers on your system, and runs a program that converts your historic claims into a new set of claims for his 20 phony providers. Charlie can attack several health plans in an evening, even though he’s on vacation in Aruba.
  3. Rules Based Post-payment Analysis allows you to look for known fraud patterns, so you can focus your investigations on the claims most likely to be fraudulent. With these systems and follow-up investigation, you should be able to get a $6-$12 return for every dollar you spend. Rules based analysis is a great leap forward for fraud detection. However, they miss a lot of fraud, because they require us to define the scam before we can find the perpetrators. If we don’t know about a scam, we can’t use rules to search for it. Rules based post-payment systems produce reports that identify providers, patients and claims that fit specific criteria. Each report identifies a particular kind of fraud. Here are some examples of rules-based reports:
    • Opticians billing for more than 2 pairs of glasses per year for a large number of their patients.
    • Patients receiving narcotics from more than 2 pharmacies or more than 2 prescribing providers in a 12-month period.
    • Podiatrists whose patients all have mental health diagnoses.
    • Providers who order chest x-rays for all patients with respiratory distress.
  4. Random Audits will not have the biggest return on investment, but they will act as a deterrent and allow you to learn more about fraud in your network. Use what you learn through random audits to enhance your other tools. For example, if you find that a certain procedure code is being billed incorrectly, you can add an edit to your system to flag this procedure before payment.
  5. Outside Data Sources & Social Network Analysis can allow you to uncover hidden links between providers and patients. For example, public records data can show you that a physician is referring all his patients to a lab owned by his brother-in-law. If he’s referring Medicaid or Medicare patients, he might be guilty of breaking self-referral or anti-kickback regulations. Data visualization tools can be used to identify fraud rings, for example, many providers billing the same services for the same set of patients.
  6. Geographic Information Systems (GIS) can be used to identify suspect billing patterns that may be the result of fraud. Here are a few examples:
    • Chiropractors, dentists and outpatient surgery centers sometimes use runners to recruit patients. The runners are paid a fee for bringing in patients. GIS analysis can identify providers who serve pockets of patients who live near each other, but not near the provider.
    • Physicians who prescribe drugs to many patients who live far from his office may be prescribing without providing an exam. This would be especially suspect if the drugs prescribed were controlled substances or had a high street value.
    • GIS systems can calculate mileage for ambulance trips to verify whether billing was accurate.
    • You’d expect a large percent of a physician’s prescriptions to be filled by pharmacies near his office. If 90% of a physician’s prescriptions are filled at a pharmacy that is not near his office, you might suspect fraud.
  7. Predictive Modeling uses past patterns to predict future patterns. The same people tend to commit crime again and again, and each new fraud is often a variation of an older scam. Predictive modeling and other data mining techniques look for hidden patterns that are fraud indicators. For example, somebody might use several different provider numbers to spread fraudulent payments into small enough amounts that they are not noticed or pursued. Data mining tools can identify a modus operandi for this provider, then group all his claims together, regardless of provider number.
  8. Advanced Prepayment Edits compare new claims to past claims to identify fraud patterns in real-time. These can be used to identify sudden changes, such as a surge in billings. Fraud perpetrators often bill a huge amount, then take the money and run before the health plan can catch them. Advanced prepayment edits can catch the scam before the payments are made.
  9. Biometrics can be used in a number of ways to prevent fraud before it is billed. Signature verification tools can be used to verify recipient identity, prove recipient presence at a location and time.