Drug costs, not volume, causes regional differences in Medicare drug spending

The cost of medications through Medicare's subsidized prescription drug program varies from region to region across the United States largely due to the use of more expensive brand-name drugs and not because of the amount of drugs prescribed, according to a study led by researchers from the University of Pittsburgh Graduate School of Public Health (GSPH). The authors said that more efficient prescribing practices could have saved the Medicare program and its beneficiaries $4.5 billion.

The study results, which appeared in the Feb. 9 issue of The , suggest increased use of lower-cost could substantially reduce drug spending and beneficiary out-of-pocket without compromising quality of care or health. In addition, regional costs per prescription closely parallel the use of brand-named drugs.

The research suggests the benefit design, which promotes cost-sharing and utilization management, may be an important tool for boosting the use of generic drugs in the program and saving money, particularly in high-cost regions, said Julie M. Donohue, Ph.D., associate professor of health policy and management and lead author of the study.

"Promoting the use of generics could greatly lower out-of-pocket costs for patients and save Medicare money. Lower costs could potentially lead to improved adherence to medication regimens, which in turn would lead to overall improvements in health," Dr. Donohue said.

Studies have shown that there are differences in Medicare drug spending across the United States, but until now the reasons behind those differences have not been well understood.

Dr. Donohue's team examined 2008 for 4.7 million beneficiaries. In addition to studying overall medication use, they looked at three drug categories widely prescribed to the elderly: blood pressure medications, cholesterol-lowering statins and newer antidepressants. The data was analyzed across hospital referral regions and adjusted for demographic, socioeconomic and health status differences.

They found that mean adjusted per capita pharmaceutical spending ranged from $2,413 in the lowest hospital referral region to $3,008 in the highest. More than 75 percent of that difference was due to the cost per prescription ($53 versus $63). However, the data indicated differences in the role of volume versus cost depending on the drug class studied. For example, the cost per prescription was the most important factor for medications used to treat hypertension and high cholesterol while the differences in volume were more important in the costs of antidepressants.

"The preference to use antidepressants among the elderly may vary by region, which may account for some of the difference the research found in this drug class compared to the other two studied," Dr. Donohue said.

Related Stories

Study addresses impact of Medicare Part D on medical spending

Jul 01, 2009

After enrolling in Medicare Part D, seniors who previously had limited or no drug coverage spent more on prescriptions and less on other medical care services such as hospitalizations and visits to the doctor's office, according ...

Recommended for you

Study reveals state of crisis in Canadian foster care system

11 hours ago

A new study of foster care in Canada led by a researcher at Western University reveals a shrinking number of foster care providers are available across the country to care for a growing number of children with increasingly ...

Researchers prove the benefits of persimmons for diet

12 hours ago

Alba Mir and Ana Domingo, researchers from the Department of Analytical Chemistry of the University of Valencia, under the supervision of professors Miguel de la Guardia and Maria Luisa Cervera, from the same department, ...

Hand blenders used for cooking can emit persistent chemicals

12 hours ago

Eight out of twelve tested models of hand blenders are leaking chlorinated paraffins when used according to the suppliers' instructions. This is revealed in a report from Stockholm University where researchers analyzed a ...

User comments