How AI Is Turning Workers’ Comp Claim Data Into Tools You Can Use Right Now

Over one million workers' comp claims were auto-coded using AI, creating a new source of injury intelligence intended to target the causes of work injuries.
By: | July 17, 2019

Over one million workers’ compensation claims were auto-coded by researchers using artificial intelligence, or AI, creating a new resource of injury intelligence that insurers, brokers and employers can use to better understand injury trends and to get ahead of common work-related injuries.

The results of the study, which was conducted by the National Institute for Occupational Health and Safety and the Ohio Bureau of Workers’ Compensation, are already being used by industry researchers to prioritize prevention, using ergonomic or slips, trips and falls intervention studies, according to the Center for Disease Control and Prevention.

By the Numbers:

  • AI determined the cause of 1.2 million claims and placed them into one of three categories: ergonomic related; slips, trips and falls; or all other categories combined.
  • It took the revised computer program less than 3 hours to categorize all these claims.
  • Assuming an average manual coding rate of 2.2 claims per minute, it would have taken a human 4.5 years to categorize all of these claims. By speeding up this process, AI is enabling researchers, insurers, brokers and employers to use far more recent data when looking into specific injuries and their causes.

How Does This Help?

The CDC noted that having another source of surveillance data will allow researchers to analyze data from multiple sources in order to find common patterns of risk among specific industries. Other state bureaus and insurers may also be able to use similar technology to auto-code claims and understand common causes of injury.

The data from the study has also been visualized in this dashboard, which makes it easy for other states, researchers, and insurers to access the data.

The results of the study are already being used by NIOSH to prioritize NIOSH-funded research to a greater degree of precision with the hopes of maximizing impact and finding more specific ways to prevent workplace injury.

Dr. Alysha R. Meyers, one of the authors of the study said “the data collected through this system can also provide valuable information on how to prevent injury from occurring in the first place by identifying the cause.”

Meyers also praised the use of AI in analyzing worker’s compensation claims. “The use of auto-coding to explore these and other databases makes them infinitely more valuable as an occupational injury and illness surveillance tool,” she said.

Recommended Reading

From auto-coding claims to generating actionable data, AI is impacting the worker’s compensation industry as well as the broader property-casualty realm.

The Company ChronWell offers two platforms, Recovry and Recovry Central, that use AI to create and adjust profiles and care plans for injured workers. Their pilot program has shown a 38% reduction in attorney involvement, according to CEO Joe Rubinsztain.

Aon is also getting in on the AI game. This month, the company announced that they are expanding their alliance with CLARA analytics with the goal of using AI to reduce spending on workers’ compensation litigation. The technology they’re using looks at AI to help detect claims that are at high-risk for litigation.

Actionable data derived from AI is also being used to make insurance underwriting more accurate. 63% of insurance executives say the industry will be transformed by these new technologies.

The use of AI in both insurance and workers’ compensation doesn’t come without risks, however. The use of AI in businesses is creating new questions of liability and coverage: Who is at fault if an automated truck fails to break? How do insurers cover the loss of data in the event of a cyber attack?

Additionally, some worry that AI may discriminate against women and non-white people. &

Courtney DuChene is a freelance journalist based in Philadelphia. She can be reached at [email protected].

More from Risk & Insurance