Two items in today’s news indicate the growing significance of data analytics in managing workers comp claims and costs. While there are differences in focus, with Liberty focusing on provider identification and Crawford on claims and case management, Liberty Mutual and Crawford are both using data analysis to build the capability to improve results.
Crawford has worked with e-Triage to develop and implement a predictive modeling capability to enable the company’s claims adjusters to identify those claims that are more likely to be problematic.
“”Crawford Claims Advantage provides an objective means to identify from the beginning the 20 percent of claims that account for 80 percent of all workers’ compensation costs,” said Bob Kulbick, senior vice president of Crawford’s Risk Management Services. “This new proprietary system will ensure consistency in the handling of workers’ compensation claims and will give us an unparalleled advantage in the marketplace. We expect to see significant reductions in both costs and claims duration for our corporate clients.”
“The opportunity to apply objective criteria to identify those cases that can benefit from case management is significant,” said Larry Mattingly, senior vice president of Crawford’s Healthcare Management Services. “Rather than subjective assignments, our case managers will intervene on the right cases at the right time. Phase two of our association with e-Triage will incorporate the evidence based research into our case management process, providing an objective means of measuring improvement as a result of the services provided.”
Liberty Mutual also announced the preliminary results of their partnership with Thomson Medstat, the well-known health data analysis firm. Liberty has been working with Medstat to develop a data warehouse and analytic capability to enable the company to better understand the delivery of medical care to their insureds.
According to Insurance Journal,
“…Liberty Mutual uncovered what it believes are the two keys to managing workers’ comp medical costs.
First, the statistical averages for treating specific injuries in any city or state – for example, how many office visits are needed to heal a torn rotator cuff in Denver, or what does it cost to set a compound arm fracture in Pennsylvania. Second, how individual caregivers and facilities in the area compare to that baseline.”
Interesting that the keys do not include the level of discount, network penetration, usage of case management or utilization review. These tools have been viewed as critical; Liberty’s statement seems to imply that finding the right docs is more important than the traditional approach.
I wholeheartedly agree.
The people at Liberty are some of the most knowledgeable in the WC industry, and their application of analytics to this business is consistent with what others (Hartford, Aetna) are also attempting. While there are inherent challenges in dealing with WC data due to relatively small sample sizes (total WC medical expense nationally is about 2% of all medical expenses), historically poor data quality (little emphasis on capturing data at the bill processor level), and significant external complicating factors (co-morbidities, psycho-social factors, job satisfaction) this is a step in the right direction.
What does this mean for you?
Depends on how much data you have…
Nov
17
Joe –
Very useful information. I am particularly interested in how Crawford is incorporating predictive modeling into their claims management – thanks for the insight.
I’m a Crawford client. I’m disappointed regarding their attempts to incorporate this into their claims management process. I’d also like to partner with other Crawford clients on WC benchmarking. I’d like to share my date with others interested in gaining a better understanding of WC costs. jeff.plass@kohler.com