For the past several years, standard insurance companies have been implementing innovative new technologies, including advanced analytics and big data, that can help detect and prevent fraud. But can these same ideas be used in Workers’ Compensation Fraud?
The answer, in short, is yes – but it’s a rather complicated process and may take some time to integrate. Here are the steps insurers are beginning to take in order to utilize analytics to prevent fraud.
Step One: Integrate all the data. The first step is probably the most daunting for insurers. In order to run large analytics, workers’ comp insurers must pull together all the data available to them from all of their different sources. This includes internal sources and third-party databases, as well as non-traditional data sources such as social media or other public content. This might include claims data, medical records, medical billing and bill audit data, pharmacy data, medical provider information, policy application data, underwriting data, loss history data, and internal and industry watch lists of known bad providers.
Step Two: Scrub the data. With all this information, there’s bound to be some data that is duplicated or incomplete. It is very important to cross-reference the data, combining any fields that represent the same sample or person in order to represent them factually and completely.
Step Three: Apply big data analytics approaches and techniques. This is the part where advanced technical processes come into play. Examples of analytic techniques used in fraud detection are heuristic rules, anomaly detection, predictive models, text mining and link analysis.
Step Four: Combine all the results. In order to get the kind of data that will offer specific fraud investigation opportunities, analytics professionals suggest a “holistic” approach – that is, combining multiple data techniques and cross-referencing those techniques in order to narrow down the results into tangible leads. Special Investigation Units for fraud usually rely on a few consistent sources for red flags, but with these data sets, they can instead get a fuller picture of fraud across geographic and categorical lines.
So when will we see more big data providing leads for workers’ comp fraud investigations? It’s not certain. Not only will giant amounts of data have to be collected and sorted, but they’ll have to be done so quickly – one of the biggest hurdles insurers face is state-mandated time limits on claims functions.
However, the fact remains: savvy analytics techniques aren’t just a technological fad, but a shift in the way we are able to harness and use data. Hopefully, the data needed will be in place to capture workers’ comp scammers before they commit fraud.
Sources:
http://www.propertycasualty360.com/2016/04/12/overcoming-workers-comp-fraud-with-detection-techn
http://www.insurancetech.com/big-data-for-fraud-detection/a/d-id/1314553?
 |
Charity Murphy is a freelance writer for Insured Solutions and Improve comedienne based in Louisville, Kentucky. She lives with her husband Ben and their two dogs, Sprocket and Ms. Brisby. |