By: Eric Siegel, Founder, Predictive Analytics World
In anticipation of her upcoming conference presentation, Triaging Insurance Claims as They Come In the Door, at Predictive Analytics World for Financial in Las Vegas, June 3-7, 2018, we asked Christina 
Hoy, Vice President at Workplace Safety and Insurance Board (WSIB), a few questions about her work in predictive analytics.
Q: In your work with predictive analytics, what behavior or outcome do your models predict?
A: Our greatest goal at the Workplace Safety and Insurance Board is to support those who are injured at work to recover as completely and quickly as possible, and to make sure that level of recovery can be maintained. We need to predict what will best help each worker, depending on their situation, to do this. Sustainable recovery by injured and ill workers is critical not just to the workers and their families, but also for the success of our employers and the economic strength of our province.
Q: How does predictive analytics deliver value at your organization – what is one specific way in which it actively drives decisions or operations?
A: As we support injured and ill workers in their recovery, there are typically dozens of touchpoints along their journey with us. Each time, their case manager or another member of their support team must make a decision about what type of service or care will deliver the greatest value towards the long-term recovery of the worker. Predictive analytics is already guiding our support teams to make better decisions regarding worker care and services, and there is much more we are planning to do to tap into its power as well as to refine decision guidance through machine learning.
Q: Can you describe a quantitative result, such as the predictive lift of your model or the ROI of an analytics initiative?
A: In recent years, we have seen the percentage of workers who experience permanent impairments from their workplace injury or illness come down dramatically – from 13% of lost-time claims in 2009 to 6% in 2016. Permanent impairments can be devastating to the worker and involve additional costs to the workers’ compensation system. While not all of this reduction can be attributed to our use of predictive analytics, predictive analytics has been allowing us to identify “high risk” claims sooner, get these workers the services they need, and reduce the chance that a permanent impairment will occur.
Q: What surprising discovery or insight have you unearthed in your data?
A: Years ago we learned from our data that a worker who continues to require benefits three months after their injury is far more likely to need benefits through the long term. Getting workers safely back to work before this point, even with modified duties, greatly increases their chance of making a full recovery. This finding meant that we have sharpened our focus, including our use of predictive analytics, towards early decision making on the claim and getting the right support to each injured worker as soon as possible in the claim.
Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.
A: My talk will cover how we have been able to use predictive analytics to identify “high risk” workers’ compensation claims using only limited data – the information available at the time the claim is first registered with us. Overcoming this data limitation is allowing us to target these high risk claims with specialized services and resources days, or even weeks, earlier than before. With these claims especially, every day is crucial to the worker’s prospects for a full recovery.
———————
Don’t miss Christina’s conference presentation, Triaging Insurance Claims as They Come In the Door on Wednesday, June 6, 2018 from 3:30 to 4:10 pm at Predictive Analytics World for Financial in Las Vegas. Click here to register to attend.
By: Eric Siegel, Founder, Predictive Analytics World