Three facts related to diabetes management should make most employers take notice:

1. Plan participants with diabetes typically are responsible for more than 10% of a health plan's claim costs in any given year.

2. Many plan participants with diabetes fail to comply with their prescribed treatment.

3. The costs and the quality-of-life hardships that can result from complications from diabetes can now be averted without high-cost intervention.

Just a few years ago, expensive technology or inflexible claim payers prevented employers from taking control of diabetes management among their workforce. Today, new electronic data management tools are making it possible to study large amounts of medical data on aggregate populations and individual patients.

Data mining is one new tool that can be particularly effective in dealing with diabetes. By shining a light on diabetes treatment, data mining enables plan sponsors to demand more transparency and begin to require physicians and hospitals to compete on outcomes and price instead of on references and reputation.

Employer case study

By applying The Segal Co.'s medical data mining, analysts found that in one employer's health plan, which had about 3,500 participants, 12% of the costs were due to diabetes.

The data mining analysis also found that only 15% of diabetics who participate in the employer's health plan had received an A1c hemoglobin test in the past year, compared to the national average of 44%. The test is essential for managing blood sugar levels. Some studies have shown that every percentage point drop in A1c cuts a patient's risk of eye, kidney or nerve-related complications by 40%.

These findings shocked the employer, which was working on programs to increase the involvement of participants with specific diseases, including diabetes, and hold vendors accountable for improving the health status and behavior of plan participants.

Nevertheless, improving A1c levels will yield major dividends in the form of fewer kidney dialyses, heart treatments and other expensive therapies. It will also improve the quality of life for many patients and help the employer avoid changing health plans due to rising costs.

Understanding data mining

Data mining uses analytical software to examine a plan's raw data. Medical data management companies take this data from health insurers, third-party administrators and pharmacy benefit managers and organize it into robust clinical utilization and financial data sets.

Sophisticated software allows analysts to sort, combine and contrast key data elements to help decision-makers and clinical managers take action to improve plan performance, engage participants in their own care, contain costs and improve patient outcomes. Skilled health care analysts can use data mining to help plans uncover problems and focus on areas for improvement.

Data mining can give plan sponsors the tools they need to best meet the needs of diabetics by allowing them to tailor treatments and modify lifestyles to control costs, improve medical outcomes and hold vendors accountable for cost-effective treatments. It can help them find ways to increase adherence to testing and drug therapy compliance and slow overutilization.

In addition, data mining can aid employers in recognizing the best ways to counsel, support and educate employees and encourage providers to fully support patients emotionally and physically.

Plan sponsors can use a data mining exercise to answer essential questions about their health plans:

Is the plan's benefit design (copays and cost sharing) steering participants to cost-effective therapies, treatments and medical providers?

Which major conditions and illnesses consume the greatest portion of claim costs?

Can plan rules be designed to steer participants to the most effective health care facilities and treatments?

Is the right mix of prescription drugs being administered?

Does the plan discourage proper preventive and diagnostic treatments as a result of patient cost-sharing levels that exceed their ability to pay?

Are diabetes management programs producing improved clinical and cost outcomes?

Are complications from diabetes fueling plan cost trends?

One of the best aspects of data mining is that it can allow doctors to identify patients with risk factors associated with the onset of diabetes before the disease fully develops. Patients who have been diagnosed with diabetes can be identified earlier, before complications set in, to make sure that they are receiving optimal treatment. Data mining can show doctors which treatments have the best outcomes.

Data mining also can help patients control their disease, live healthy lives and avoid the potential consequences of diabetes. Patients can be empowered by information about optimal treatments, and data mining can show who is and is not receiving them. Patients who need the most help can be identified. People can be steered away from poor providers, and outcomes can be improved.

A better future

Medical data management tools will point the way toward a better future for plan sponsors, providers and patients. Specifically, plan sponsors can:

Determine effective levels of participant out-of-pocket costs.

Accurately measure costs or savings from plan modifications.

Perform predictive modeling to target interventions to high-risk populations.

Intervene to help patients navigate their health care treatment options more effectively.

By investing in freestanding, independent data mining tools, plan sponsors can gain control over the true drivers of plan costs and use the insights generated by these tools to make effective changes without simply shifting costs to plan members. In doing so, they take a critical step in preserving the long-term viability of health coverage for workers and their families.

Edward A. Kaplan is senior vice president and national health care practice leader at The Segal Co.