Predicting, reducing high-cost claims can be done, with the right approach
According to Joe Anderson, Director of Analytic Services at Progressive Medical Inc., a pharmacy benefit management (PBM) firm headquartered just outside of Columbus in Westerville, Ohio, the dream is closer to reality than one might think. Simply stated, Progressive Medical believes that by harnessing data appropriately, claims professionals are positioned to make better decisions surrounding a claim. But what exactly is analytics? And why is technique so important to unravel the mystery?
According to Anderson, there are volumes of data in varying forms that can be used to predict if a claim will be high risk or high cost. This data can range from the factors traditionally used in utilization reviews, such as a history of multiple physicians or duplicate prescriptions, to non-traditional information such as physician prescribing patterns. Analytics involves taking all this information and applying techniques such as statistical modeling and hypothesis testing to extract valuable knowledge that can be used to predict future trends and behavior patterns.
For years, payors and PBMs alike have relied on only a few "triggers" to manage high-cost claims. But, according to Anderson, these triggers fail to identify and capture long-term pharmacy costs and they rarely integrate into an effective clinical program.
"Most triggers identify claims that have already escalated or have been so broadly defined that they will flag 10-20 low-risk claims before they identify a high-risk claim. This is not to say that the data being used is improper or incorrect, but there is an opportunity to go further with the analysis. That's why we believe that using triggers already tested by statistical models is a more ideal approach."
Progressive Medical uses data mining techniques to build a multivariate statistical model, which encompasses interactions among a claim's injury type, medication mix, type of doctor, and more.
"We quantify relationships between these variables early in a claim's lifetime and then summarize that information down to a single risk score that predicts future outcomes," Anderson said. "Essentially, by leveraging data early in the claim, and because our model correctly incorporates the influence of each factor, we can quickly and effectively determine claims with the highest cost potential."
Controlling pharmacy expenses, managing utilization and getting employees back to work quickly are not new goals for payors. In fact, these are some of the primary objectives for retaining the services of a PBM in the first place. But new challenges are arising that makes accomplishing these goals more difficult. One such challenge is the nation's rising dependency on prescription medications -- particularly narcotics -- which currently account for the largest portion of workers' compensation medication expenses. Another factor is the increasing ability for injured workers to obtain medications outside of the pharmacy network (i.e. through physician dispensaries), significantly reducing the opportunity to manage utilization.
"Our approach looks at why and when these factors occur," Anderson said. "As a result, we are able to more effectively identify and understand what types of claims are most expensive and then quantify a predicted value for new claims."
- Progressive Medical's Clinical Services
- Contact Us
- Progressive Medical's Workers' Compensation Resource Center
This correlation is critical to payors because it gives them the power to clinically intervene sooner on claims before they become expensive or troublesome. That being said, Anderson notes that all this analysis is only beneficial if you integrate it with clinical programs and expertise. "Forecasting is only half the solution," he said. "By incorporating the findings into our clinical programs, we are able to provide our clients with a powerful tool for reducing expenses and improving outcomes. This proprietary approach will ultimately lead to shorter claim duration and a reduction in fraud, misuse and abuse that enables payors to ensure better outcomes for their injured workers."
To fully understand the benefits of its proprietary approach, Progressive Medical is engaging in a series of internal pilots, along with a more extensive one that involves a large payor. The pilots use both traditional and non-traditional data and look at patterns occurring as early as one month after date of injury to a few months later. Once high risk claims are identified, Progressive Medical engages its clinical intervention team, whose efforts have already demonstrated medication cost savings of 28% per claim for clients. It is this opportunity, the ability to further integrate data analytics with clinical intervention practices, that is really exciting for the company.
"We generate the risk scores to predict high-cost claims, including the reasons that our model identified claims as 'high risk'" Anderson said. "We then provide the data to our clinical team to intervene on claims where it is most appropriate." The latest pilot digs deeper into the triggers of high-cost claims and other factors that may not be as easy to identify with traditional analysis. "Again, if you only look at individual triggers you are not fully understanding the drivers of claim costs," Anderson emphasized. "A deep, multivariate model is required to ensure nothing is missed."
At the end of the day, Progressive Medical's goal is to leverage expertise in both data analytics and clinical operations to change the way payors look at their data to ultimately shape better results, not because it's trendy to do so, but because it's the right thing to do.
(The above piece is part of our continuing Insights series designed to highlight key products and services to our readers. This paid-for Insights was written and edited by Risk & Insurance®
on behalf of our marketing partner. Additional Insights can be found on our Web site at www.riskandinsurance.com.)
August 22, 2012
Copyright 2012© LRP Publications