Effective predictive modeling can flag the small percentage of claims that would typically be in the $10,000 to $15,000 range but end up costing six figures. By identifying them early in the claims process, claims handlers can use strategies that result in better outcomes.
That's the idea behind Liberty Mutual's Next Generation Predictive Model, according to company officials. They say about 20 percent of the 825,000 lost time claims they analyzed fell into the high-cost claims category.
The company also evaluated more than 140 million individual medical billing transactions. The carrier then ran more than 200,000 lost time claims through the updated model to validate its accuracy.
A spokesman says the updated model uses a combination of:
- Broader data. In addition to medical and billing information, the company captured psychological and social issues and comorbid medical conditions based partly on research from the Liberty Mutual Research Institute's Center for Disability Research.
- Sophisticated multivariate analysis. The model looks at the data holistically by taking the impact of all data points together and over time to identify likely changes in the claim and their potential impact.
- More frequent model runs. The predictive model looks at claims monthly to pick up changes in each claim's profile that can negatively impact the claim's development, such as emerging medical and nonmedical factors.
The motivation of injured workers can be pivotal in their recovery time. Events in the person's home life, as well as personal issues or employment-related challenges, such as fear of layoffs, can play a significant role in how quickly the claimant reaches maximum medical improvement, the company says.
Also important is jurisdictional issues. The representatives said costs are higher in some areas and disability periods sometimes vary by region.
The model also includes tools to address high escalating claims. A medical referral tool triggers referrals to specialized medical resources while a dashboard presents relevant claims information, including flags to indicate importance.
As an example, a company official related the recent case of a claimant who was identified as diabetic by the medical billing information. That led the claims handler to ask different questions of the treating physician, which uncovered additional risk factors. The insurer got a nurse, the treating physician, and other stakeholders involved in a discussion to help with the claim.
Treating the whole person, rather than just the injury, can lead to quicker recoveries and returns to work, the company said.
Read more at the WorkersComp Forum homepage.
September 29, 2011
Copyright 2011© LRP Publications