Most of the wide geographic variation in Medicare spending for prescription drugs is due to the cost of the drugs selected rather than to the volume of drugs used in each region, according to a report in the Feb. 9 issue of the New England Journal of Medicine.
The choice of brand-name drugs rather than generics accounts for approximately 76% of this variation. If the geographic areas with the highest drug spending had used generics as often as those with the lowest drug spending in 2008, Medicare and its beneficiaries would have saved $4.5 billion that year, said Julie M. Donohue, Ph.D., of the University of Pittsburgh and her associates.
A considerable variation in prescription-drug spending was reported in 2010, but the reasons for that variation were not identified. Some speculated that costs were higher in regions with the most patients or the most seriously ill patients than regions with fewer patients or less seriously ill patients.
Dr. Donohue and her colleagues analyzed data on 4,666,866 Part D beneficiaries in 2008. They assessed spending for three classes of drugs commonly used by the elderly for which there are no over-the-counter substitutes and which together account for nearly one-fourth of all drug spending annually: ACE inhibitors and ARBs, statins, and newer antidepressants (SSRIs and SNRIs).
After adjustment for demographic, socioeconomic, and health-status differences among the 306 geographic regions studied, per capita prescription-drug spending varied by 25% from the area with the lowest drug spending ($2,413 per patient) to that with the highest ($3,008 per patient).
Overall, most (76%) of the regional variation was ascribed to the cost per prescription filled, and the remaining 24% was ascribed to differences in volume. However this pattern differed among the three drug categories studied, the investigators said (N. Engl. J. Med. 2012;366:530-8).
For ACE inhibitors and ARBs, 90% of the variation in costs was due to the use of brand-name ARBs, which had no generic equivalents during the study period. For statins, 56% of the variation was due to the use of brand-name rather than generic drugs.
But for antidepressants, only about 36% of the variation was due to the use of brand-name rather than generic drugs. Most of the variation instead was due to wide differences in the volume of SSRIs and SNRIs used in various areas of the country. The highest-use areas filled nearly 30% more prescriptions per capita for these antidepressants than did the lowest-use areas.
Some areas of the country consistently used a high percentage of brand-name drugs, while others consistently used a high percentage of generics. For example, in Miami, patients almost always were prescribed brand-name ACE inhibitors, ARBs, statins, SSRIs, and SNRIs, while patients in Rochester, Minn., rarely were prescribed brand-name drugs in any of these categories.
If the regions in the highest quintile of prescription-drug spending had used the same number of generics as those in the lowest quintile of spending, "we estimate that overall Part D drug spending would have been 10% lower in 2008." This translates into a nationwide savings of $4.5 billion just for Medicare and its beneficiaries, Dr. Donohue and her associates noted.
This study was supported by the National Institute on Aging, the Agency for Healthcare Research and Quality, the National Institute of Nursing Research, the National Institute of Mental Health, the Robert Wood Johnson Foundation, and the Veterans Affairs Health Services Research and Development Service. No relevant financial conflicts of interest were reported.