Advances

A prescription for savings

By Susan Green

Prescribing brand-name rather than generic drugs for patients with diabetes cost Medicare Part D more than $1 billion in 2008. Compared to patients treated in the VA health system, patients enrolled in Medicare Part D were more likely to be prescribed brand-name drugs for four common classes of medications.

Every year, Medicare Part D spends about $60 billion on prescription drugs. According to Nancy Morden, an associate professor of community and family medicine at Geisel, at least some of that expense is the result of the unnecessary use of brand-name drugs rather than generics.

In 2010, when data from Medicare Part D (Medicare's prescription drug plan) was released by the private insurers administering the plan, researchers were for the first time able to analyze prescription patterns. Early studies showed that prescription spending varied widely from one area to another and that as much as 70 percent of spending differences in commonly used drug classes was the result of drug choice rather than the volume of drugs prescribed. That raised the question of what was causing the variation.

Morden thought that the answer might involve something more than just patient choice. She believed it could be the result of the behavior of the physicians prescribing the drugs and of how the Part D program functions.

In a recent analysis, Morden focused on four medication groups regularly prescribed to patients with diabetes and compared both the overall and regional rates of brand-name drug use in Medicare Part D to rates of brand-name drug use among similar patients served by the Department of Veterans Affairs Health Administration.

She found that brand-name drug use in Part D, which relies on more than 1,000 private insurers with distinct formularies and cost-sharing arrangements, is up to three times higher than it is among similar VA beneficiaries, costing Medicare more than $1.4 billion dollars annually.

Morden explains that in contrast to Medicare, the VA has a more limited list of drugs that are covered. "They don't allow beneficiaries to select what they want from a list of all the drugs on the market," she says. "We found that with a closed system, you have less brand-name use and more efficiency in prescribing."

The findings confirmed Morden's belief that the disparity in spending is tied to both the individual prescriber and to the system as a whole. Systems have the ability to narrow drug options and to direct prescribers toward less expensive drugs.

"Our findings draw on actual rates of generic drug use in an existing high-quality health system, such as the VA, and demonstrate what could be attainable in Medicare," Morden says. "These potential savings could be realized through policies that promote Part D plan efficiency and by encouraging physicians to consider costs and value in their prescribing."

The VA is a reasonable benchmark for the use of generic drugs in Part D because it performs as well as or better than commercial health plans on several measures of quality for diabetes and related conditions. "These findings point to opportunities for improving efficiency without harming quality of care or access to effective medicines," Morden says.

She also notes, however, that for drugs that the VA doesn't control, regional prescription patterns among VA patients and Medicare Part D patients were similar. "This correlation led us to believe that there's something about regional culture, prescribers, and patients that leads people to use more expensive products," she says. It's not yet clear what determines these differences in regional prescription patterns, but Morden and her colleagues at the Dartmouth Institute for Health Policy and Clinical Practice are examining these variations in prescribing behavior and trying to figure out how to curb unnecessary spending.

"We have to start holding people accountable for their prescribing practices," Morden says. "Ultimately, it is the prescriber's responsibility to prescribe well, to consider the safety, cost, and quality of the drug. Systems aside, there is no excuse for bad prescribing."