Can the Auto Casualty Claims Process Assist Workers’ Compensation? : Risk & Insurance | #employeefraud | #recruitment | #corporatesecurity



Can advancements in artificial intelligence that are being implemented in the auto insurance realm find traction in workers’ compensation?

In today’s world, data is readily available. In an instant, your smartphone can tell you how many steps you’ve walked today, how many hours you’ve spent on Instagram and how many times you took an Uber last month.

In the insurance industry, data is also becoming increasingly more accessible due to the widespread availability of technology, helping claims organizations to better manage premiums, control medical spending and improve internal workflow efficiencies.

Specifically, in the auto insurance industry, companies have been using a variety of data sources, including telematics, smart phone apps and video, over the past few years to collect data that not only allows carriers to offer usage-based insurance to its customers, but also has been able to help claims organizations try to predict the severity of injuries based on an accident scenario.

Injury prediction technology, which is at the outset of being adopted in the auto casualty industry, also has the potential to be applied to workers’ compensation claims, which could provide improvements to the claims process for both claims organizations and injured employees.

How Does Injury Prediction Work in the Auto Casualty Claims Process?

In the auto insurance industry today, it is easier than ever for carriers to collect collision data and use that information to predict the severity and type of injuries that may result from the crash.

Shahin Hatamian, product management and strategy group leader, Mitchell Auto Casualty and Workers’ Compensation Solutions

Many cars are outfitted with telematics systems — and this number is constantly growing. For example, Statista predicts that by 2030, there will be 146 million connected cars in operation in the U.S. — and claimants frequently are providing detailed information and photos almost instantly after an accident due to the availability of carrier mobile apps.

Additionally, photo-based estimating systems using computer vision and artificial intelligence are commercially available in the market today, making it possible for carriers to leverage technology to detect vehicle damage from photos submitted at first notice of loss.

Once carriers are able to collect data about an accident scenario, predictive models can then take that information and make an estimation about the type and/or severity of the injury.

For example, the following data could be input into the model:

  • Statistics from the telematics system reporting that the claimant was driving at 40 miles per hour at the time of the crash, the driver seatbelt was fastened, and that the airbag system deployed.
  • Information from the photo-based estimating system that the left side panel of the vehicle was damaged.
  • Data submitted by the claimant in the first notice of loss app and retrieved via policy information that he is a 55-year-old male, and was the driver at the time of the accident.

As an output, the artificial intelligence may use the collision data collected from the different sources to help predict what type of injuries that claimant may have sustained and the potential severity.

Knowing that information, carriers can first appropriately triage the claim to the right adjuster within the organization. Adjusters can also be better prepared to handle the claim.

For example, if the model predicts mild injuries, the adjuster could put the claim on a fast-settlement track to try to settle the claim quickly and avoid it spiraling.

If the model predicted a more severe injury, the adjuster could better understand the predicted treatment timeline to help manage the claim moving forward.

Additionally, the injury prediction information can help adjusters to better judge if treatments submitted for reimbursement are actually related to the injuries sustained in the accident.

The auto insurance industry has already seen the benefits of having the capability to predict injury type and severity from auto accident data. Injury prediction analytics has helped insurance carriers get the claim to the right teams quickly, identify potential fraud and speed up claims processing.

Predicting Workers’ Compensation Injuries

Historically, the auto casualty market has been on the forefront of consumer-facing digital transformation, while the workers’ compensation market has typically led the way in technology adoption for better interaction with providers and for cost containment.

Typically, there is a fair amount of overlap between the industries sharing technologies and solutions, since the claims process for both lines of insurance are very similar at the core.

Norman Tyrrell, product management and strategic planning activities director, Mitchell Auto Casualty and Workers’ Compensation Solutions

Using data to help predict injuries is one example of auto insurance technology that could be applied to the workers’ compensation claims process and has the potential to provide a variety of benefits for the industry.

First, about 6% of all workers’ compensation claims result from motor vehicle accidents, according to an NCCI report, providing an obvious application of the previously mentioned predictive technologies for these types of injuries.

But beyond motor vehicle claims, there is potential to use this technology in many different types of workers’ compensation claims.

How Would This Technology Work in Workers’ Compensation?    

The first notice of loss after an auto accident parallels the first report of injury that takes place at the onset of a workers’ compensation claim.

At the first report of injury, workers’ compensation payors typically gather a fair amount of data about the accident, including the accident scenario, information about the claimant and more.

Workers’ compensation claims organizations could begin looking at ways to standardize this data to make it more usable for predictive modeling purposes, as well as, collecting additional information, such as photos of the accident scene, which they can then process using computer vision technology, and accident scene descriptions, equipment involved, physical environment and more.

Additionally, telemedicine mobile apps can be integrated, enabling nurses or other medical personnel to provide immediate triage information.

All of that data collected at the first report of injury can be used to provide actionable insights that could  be pre-populated into the adjuster’s system to save time and inform predictive models on the type and severity of the injury that may have occurred, just like in the auto casualty claims process.

In the workers’ compensation industry, a prediction of the injury types and severities could not only be used to better triage claims and help manage treatment timelines, but also help claims organizations better direct care, improve workplace safety and more.

Here is a list of a few of the potential benefits to the workers’ compensation industry:

  • Improve triage to get claims to the right teams quickly
  • Help injured employees get the right treatments from the start of the claim to improve return-to-work time
  • Assign case management resources to claims early in the process to help manage costs and improve treatments
  • Speed up claims processing, which provides benefits to both the payor and the injured employee
  • Help reduce fraud and prevent abuse of the workers’ compensation system
  • Potentially provide workplace safety improvements and recommendations, for example encouraging truck drivers to wear their seatbelts or providing improved safety gear for construction workers

The Future of Injury Prediction Technology

In order for claims organizations to effectively use injury prediction technology, it is important that artificial intelligence models produce accurate results and that those results can be effectively interpreted and utilized by an adjuster in the claim handling process.

In a study comparing adjuster performance to a predictive model at first notice of loss, Mitchell found that a predictive model was able to more accurately and more consistently predict both the existence and severity of an injury than a human adjuster.

As injury prediction models continue to improve and become more precise over time and more data becomes available from both auto casualty and workers’ compensation accidents, this technology could expand and provide additional benefits for both claims organizations and claimants. &

Shahin Hatamian leads the Product Management and Strategy group and oversees product direction, marketing, consulting services and strategic initiatives for Mitchell Auto Casualty and Workers’ Compensation Solutions. Norman Tyrrell directs the product management and strategic planning activities for Mitchell Auto Casualty and Workers’ Compensation Solutions.





Click here for the original source and author.

Leave a Reply

Your email address will not be published. Required fields are marked *

7 + 3 =