At the 2018 CLM /Business Insurance Workers’ Compensation Conference, a panel discussed how technology is impacting risk management and claims handling. The panel included:
- Christopher Griffith – CIO, Safety National
- James Bentham – CEO, JB Knowledge
- Skip Brechtel – CIO and EVP, CCMSI
- Stan Smith – Practice Leader, Milliman, Inc
Predictive analytics is the practice of extracting information from existing data sets in order to determine patterns and assign a probability for a future outcome. The goal of predictive analytics is to help users make better decisions.
The key to predictive analytics is the quality of the data that you capture. On the workers’ compensation side this means having transaction level data about the medical treatment received along with all information, the diagnosis codes, and all the information about the injured worker’s vital statistics (Age, weight, blood pressure, comorbidities, etc).
Predictive analytics is good today for answering a question. For example, “will this claim exceed $75,000 total incurred?” If you give the model the right information, it can answer that question on claims with an 80-90% probability.
One of the big problems is that there is no national standard for what claims data should be collected. Different carriers and TPAs collect similar information, but there is variation.
Chatbots & Artificial Intelligence
We all interact with Chatbots. Amazon Alexia or Apple Siri are examples of this. When you call a customer service number and they recognize you by your phone number that is a chatbot. This is a great way to gather information and interact with customers through a series of simple questions.
Artificial intelligence allows computer to perform tasks that normally require human intelligence. Full artificial intelligence does not currently exist and we are at least 20-30 years away from this being a reality. But there are limited forms of artificial intelligence being used now.
Predictive analytics can help us make decisions while artificial intelligence can make decisions for us. This is already used on the personal auto side to automatically adjust physical damage claims.
A form of artificial intelligence can be used to review adjuster claim notes and medical records for additional information. That “data mined” information can be used to enhance the predictive models. An example of what this can do is tell you there is a 90% probably the claim will end up with back surgery. With that knowledge, the adjuster can make a decision on whether to authorize surgery now or insist on continued conservative treatment.
Robotics & RPA
Robotics is the branch of technology that deals with the design, construction, operation and application of robots.
Robotic process automation (RPA) is an emerging form of clerical process automation technology based on the notion of software robots or artificial intelligence (AI) workers. This allows you to automate some simple repetitive tasks. For example, carriers are now using this to do data regulatory reporting. The RPA will pull the required information from the policy, put it in the required format, then report it to the regulatory agency. You train the system to perform information gathering tasks that are manual and repetitive and get the RPA to perform these tasks.
Robotic process automation is also part of any claim auto-adjudication process. The robot gathers the information, makes a decision that the claim meets the guidelines for auto-adjudication, then the claim is processed automatically.
Telematics & Wearables
Telematics are already in widespread use in many industries with fleet vehicles. These can track everything happening in a vehicle including what the driver is seeing and doing. If there is an accident, you can go back and look to see what happened in the time before, during, and after an accident.
Wearables are devices that interact with your biology to record things like your steps, heart rate, temperature, location, etc. You are seeing some use of these in the construction industry to monitor worker locations for safety purposes and to monitor if they are being exposed to excessive heat or are over-exerting. You are seeing more of this in the group health space with things like glucose monitors.