Medicare prescribers use twice as many brand-name drugs as VA

Medicare beneficiaries with diabetes take many more brand-name drugs than similar patients treated in the Veterans Affairs (VA) system, adding as much as $1 billion per year to health care costs, a recent study found.

Researchers conducted a retrospective cohort study using 2008 prescription data on more than 1 million Medicare Part D beneficiaries with diabetes and more than 500,000 similar veterans. Results were published online June 11 by Annals of Internal Medicine.

In all the studied categories of drugs, Medicare patients were two to three times more likely to be taking a brand-name drug. Specifically, 35.3% of Medicare patients took brand-name oral hypoglycemics versus 12.7% of veterans. For statins, the rates were 50.7% of Medicare patients versus 18.2% of VA patients. For angiotensin-converting enzyme inhibitors and angiotensin-receptor blockers, the rates were 42.5% versus 20.8%, and for insulin analogues, the rates were 75.1% versus 27.0%.

The researchers calculated that reducing brand-name drug usage among these Medicare patients to levels seen in the VA system would have saved $1.4 billion in 2008. They theorized that the difference may be attributable to the VA's national formulary, which encourages therapeutic substitution, or the interchange of a generic drug in the same class for a brand-name drug with no exact generic equivalent (e.g., generic simvastatin for Lipitor). The VA also has a national electronic medical record with electronic prescribing, limits on pharmaceutical representative visits, and salaried physicians.

Medicare Part D plans, on the other hand, have incentives to allow greater use of brand-name drugs because they compete to enroll patients. As an example policy difference, only about 10% of studied Part D patients faced step therapy requirements to take atorvastatin or valsartan, while the VA requires step therapy and prior authorization for both those drugs. Even the VA, however, showed substantial geographic variation in brand-name drug use, which the study authors attributed to local adjudication of prior authorizations and local physician practice patterns.

The VA scores as well as or better than Medicare on quality measures for diabetes care, the authors noted, and the system's lower rates of brand-name usage should be attainable by Medicare. “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,” they concluded.