Fraud, waste, and abuse are serious problems in healthcare, diverting badly needed resources away from critical patient care. For example, it’s estimated that the federal share of improper Medicaid program payments totaled $29.12 billion in 2015.
To combat this epidemic, Cognosante is using cutting-edge health IT tools that are helping our customers identify and combat fraud, waste, and abuse in Medicaid and other healthcare programs. We’re helping states effectively oversee their Medicaid dollars to promote both program and fiscal integrity, using analytical tools to maintain program integrity and minimize fraud, waste, and abuse vulnerabilities.
The role of maps
One of the innovative tools we use to help clients is geographic information system (GIS) software. This is an information system that’s geography-based. You may have seen GIS in action on your local county’s website helping track property taxes. At Cognosante, we use it to create a map of any procedure, diagnosis, or prescription that we want to examine. What we’re looking for are spatial patterns, things that deviate from the norm.
This geography-based approach to studying healthcare trends was pioneered long before there were computers. In the most famous case, 19th century English physician John Snow, father of epidemiology, determined the cause of a local epidemic of cholera by carefully drawing up a detailed map that pinpointed all the affected households. The map made it clear that the farther a house was from the Broad Street water pump, the less likely it was that household members died from cholera. The answer was hidden all along in the raw data, but it took a map to make the connection clear. The pump handle was removed, and the epidemic ended.
Original map by John Snow showing the clusters of cholera cases in the London epidemic of 1854, drawn and lithographed by Charles Cheffins. (Published by C.F. Cheffins, Lith, Southhampton Buildings, London, England, 1854 in Snow, John. On the Mode of Communication of Cholera, 2nd Ed, John Churchill, New Burlington Street, London, England, 1855.)
Now that we have the assistance of computers and big data, we can do the same sort of work that John Snow did, but on a much higher level. We can focus on any procedure, diagnosis, or prescription about which we record copious amounts of data.
In some cases, we identify potential fraud, waste, and abuse problem areas by using a mapping technique known as hotspotting in combination with statistical modeling. This is a data-driven mapping process that lets us identify extreme patterns in a defined region of the healthcare system while controlling for known causes of variation.
Identifying trouble spots
It is a common practice to use road network data to analyze ambulance rides to hospitals. Sometimes, the mileage claims don’t match up with the distance between the patient and the provider. So, if an ambulance company claims that it drove 100 miles to deliver a patient to a provider, we can map the two locations and see if someone’s padding the books. We can use the same tools to analyze home healthcare claims. If a provider claims to have visited 20 patients within a 100-mile radius in an hour, that’s an obvious problem. And we can do this across thousands of claims with limited data (e.g., patient addresses missing), helping identify patterns and concerns that can then be addressed, ultimately resulting in cost savings.
Really, there’s no limit to the number of things we can analyze using these sophisticated mapping tools and algorithms. You can map the prevalence of prescriptions for certain drugs or surgical procedures like C-sections, or medical conditions such as low birth weight, and uncover meaningful patterns that can lead directly to better health care. At Cognosante, we regularly use sophisticated auditing algorithms and GIS tools for Medicaid program improvement. These tools can help businesses, government agencies, and healthcare organizations pinpoint improvements, which ultimately leads to better health care for patients.
In many cases, the answers that our clients are looking for are buried so deeply in the data that they’re extremely difficult to see. That’s why it helps to have a really great map.