By Patrick O’Neill, Founder and President, Redhand Advisors
Sophisticated technology that can reduce costs in claims management is top of mind — and top of budget — for efficient companies across the globe.
Our 2023 RMIS Report saw a significant jump in respondents who indicated that their companies are “definitely” planning to adopt new risk technologies within the next three years. These findings are echoed in the financial projections too: the global Insurance Claims Management Solution market size is expected to grow at a CAGR of 6.83% over the next four years, reaching almost $23 billion by 2027.
With so many eyes on smart tech solutions for claims management, Redhand Advisors hosted a recent webinar on the topic with guest speaker Principal Michael Paczolt, FCAS, MAAA, at Milliman Nodal. Michael joined me to explore the real-life benefits of artificial intelligence (AI) and natural language processing (NLP) in claims management.
Benefits of AI and NLP in claims management
When it comes it property and casualty insurance, the top 10% of claims typically account for an astounding 80% or more of total costs. Such a skewed distribution can wreak havoc on even the most solid companies’ internal systems, and until now there’s been little insight into which new claims will eventually become the costliest.
But there is a way forward: incorporating AI and NLP into your company’s risk management information system (RMIS) is the key to leveling the claims management field so your business can play smarter and reduce costs. Together AI and NLP can provide predictive, actionable feedback that leads to significant cost savings on a broad scale by:
- Identifying likely high-cost claims early on so there’s time for effective intervention
- Eliminating subjective decision-making so similar claims are treated the same way
- Providing an objective and clear understanding of claims data to drive overarching improvements
Because AI and NLP offer so much bang for their buck, the companies taking advantage of these solutions range from self-insureds to third-party administrators, and actual users include the adjuster, the claim supervisor, and the risk manager.
High tech claims management in action
Milliman’s Nodal Platform is a predictive modeling and decision support solution for managing claims. Two of the platform’s programs, Nodal Claims Triage and Nodal Medical Benchmarking, predict high-cost claims and identify wasteful medical spending. These automated solutions bring otherwise elusive data to light in a clear and actionable way, giving organizations timely feedback that helps to improve productivity and lower costs.
Using AI and NLP to extract details
Unstructured data such as the claim adjuster notes are full of rich information that the structured data is too limited to illustrate. Standard code lists fail to capture many comorbidities, prevent users from selecting multiple inputs when needed, and often don’t even accurately reflect or apply to the businesses who are using them.
Because of this, relying on structured data alone makes it almost impossible to predict the trajectory of a claim. Nodal Claims Triage uses AI and NLP to glean crucial details in the unstructured data to paint a fuller picture so risk teams are equipped with the information they need to make a real impact on those costly claims before they become so.
Here’s an example: A standard claim was coded as an elbow contusion. There have been no transactions yet, and no selected comorbidities. Nothing here jumps out to indicate this claim is headed for high cost. But, using via AI and NLP, vital information extracted from the doctor’s notes indicated that beyond an elbow contusion, there has been discussion of a dislocation and fracture involving the shoulder, wrist, and elbow. The treatment plan includes hospitalization, MRI, and surgery, and there is a comorbidity of diabetes. Suddenly we’re looking at a likely trajectory of much higher cost than just an elbow contusion, and the company has been alerted to this likelihood before any money has even been paid on the claim.
AI and NLP can provide these predictive solutions not only for workers’ compensation, but for auto liability and general liability as well, and produces the following benefits for all:
- Leverages both structured and unstructured (qualitative) data to offer a complete picture of the claim
- Employs machine learning to generate predictive insights delivered to users through dynamic reporting
- Utilizes feature engineering to draw conclusions from rich data interpretations
- Delivers predictions about which claims are likely to be high cost within the first two weeks of the claim
- Delivers predictions around the likelihood of other variables, including attorney involvement, litigation, or even cycle time
Using AI and NLP to benchmark claims
Workers’ compensation (WC) spends 60%-100% more than group health for similar conditions due to over-utilization and high pricing.
Previously, insurers have only been able to compare WC claims to other WC claims, with no transparency and real-time insights into what group health is paying for the same procedures and treatments.
Nodal Medical Benchmarking allows clients to compare costs apples to apples, and to almost instantly understand what the market cost is for the injury at hand. This new solution automatically benchmarks the cost of medical care in WC claims against outcomes in Milliman’s proprietary, robust group health database.
As an example, shoulder injuries are, on average, the most expensive WC injuries. When WC claims are compared to similar group health claims, the WC utilization is 42% higher than group health. When you can benchmark for the prescribed treatment before surgery, the day of surgery, and post-surgery, risk teams are empowered with the data they need to act.
Equipped with this powerful information, companies can work smarter to:
- Reduce medical spending on claims across the board
- Reduce the cost of medical care and cycle times
- Identify opportunities to reduce the cost of individual claims
- Better match clients or claims with the right provider, treatment plan, and at the right price
To see how engaging AI and NLP in your RMIS can reduce costs and improve efficiency within your organization, schedule an inquiry call with Redhand Advisors.