Mining Claims Data Helps Estimate Savings For Managed Care Programs

By

Industry experts agree that medical management helps control the cost of workers compensation claims. Today, the vast majority of employersboth self insured and fully fundedoffer managed care workers compensation programs, and most are well versed on the benefits of case management to reduce both injury costs and duration.

But, employers and payors arent necessarily satisfied with generalities about claims savings estimations and/or results. More and more they are looking for specifics about how much their managed care programs are saving them and ways to increase these savings even more.

Providing these estimates has been difficult in the past, because cases that benefit most from claims management are complexand usually the injury is more severe than in their unmanaged counterparts. Naturally, the costs associated with these complex claims are higher and consequently, employers and payors are likely to question their true savings.

A new approach to claims analysis and savings estimations that incorporates sophisticated data-mining tools provides a more accurate answer to this problem. And it gives risk managers a leg up on developing programs to drive down medical costs, reduce employee absences and improve return-to-work outcomes.

There are many attributes or variables that impact how a company measures the financial impact of workers compensation managed care. These range from the most common, such as disability duration, diagnosis nature and injury severity, to the less frequent, such as geography or job. Often, these unique factors may not be captured by the claims system.

In addition, there are factors such as infrequent but severe injuries or rare catastrophic claims that can account for a significant portion of the total cost of managed claim. If not filtered out, these factors can dramatically skew savings estimations.

Some organizations are turning to methodologies that incorporate data mining to provide more accurate and realistic savings estimations. In general terms, data mining involves transforming raw data into useful business intelligence to reveal trends and predict and explain outcomes.

Data mining software allows analysts to quickly sort through all claims data and spot a wide variety of claim trends and patterns not easily identified by traditional methods, and that provide a more realistic prediction of total claim costs and/or savings.

Additionally, when data mining incorporates predictive modeling to calculate future opportunities for savings, this yields an even more reliable picture of future claim costs.

Data mining tools have evolved considerably over the past two years and their level of sophistication and complexity has increased greatly in the most recent past. Older tools often evaluate fewer data points and are biased by the variables that the analyst chooses to model. Often these are not inclusive of all the data that is available and do not allow for adjustments, making them less reliable sources of information about claim trends and durations.

In many instances, older tools provide information that can lead to overly optimistic savings evaluations. The tools limitations, in turn, restrict what risk managers could do to actively plan for, manage and perhaps mitigate losses. By eliminating these barriers, newer data mining tools provide much more meaningful and reliable results for risk managers.

Data mining tools can flag potential future claim issues as well as analyze historical results, such as injury trends or lengthy claim durations in certain geographic areasfactors risk managers need to evaluate. It is up to the managers to create the solutions and execute strategies to mitigate or prevent additional losses.

A hypothetical example demonstrates the power and promise, of more effective data mining.