Radiology is less susceptible than many other specialties to the lure of healthcare vendors who wine and dine physicians to drive sales of their products. Decisions around imaging equipment acquisitions are seldom made unilaterally, and now a new study confirms the lack of an association between increased imaging utilization and monies, gifts or perks given to rads for purposes other than supporting research.
The Journal of the American College of Radiology published the study online Jan. 9.
Nima Kokabi, MD, of Yale and colleagues identified all non-research-related industry payments made to radiologists in 2014 as posted in CMS’s Open Payments database.
The researchers used federal Area Health Resources Files to identify actual and population-weighted numbers of radiologists by state, calculating average Medicare imaging spending per beneficiary in each state using a 5 percent random beneficiary sample from 2014.
They further calculated average frequency and dollar amounts of non-research non-royalty payments to radiologists at the state level and evaluated, also at the state level, amounts of non-research payments to radiologists versus per-beneficiary Medicare imaging spending.
Among their key findings: noteworthy geographic variation.
Just under 13 percent of radiologists in the District of Columbia received such payments versus 0 percent in Vermont, for example.
More to the point of the project, the authors report finding no correlation between average per-beneficiary Medicare imaging spending and the proportion of non-research-funded radiologists in each state (r = 0.06).
Additionally, they found no correlation between average per-beneficiary Medicare imaging spending and the average non-research transfer amount to radiologists in each state (r = -0.08).
Also of interest in the findings:
- Overall, 2,008 radiologists (1,670 diagnostic, 338 interventional) received non-research non-royalty payments from industry in 2014, representing 5.2 percent of all 38,857 radiologists nationwide.
- A total of 4,975 individual transfers translated to 2.5 ± 1.3 discrete payments per receiving radiologist with a mean of $432 ± $1,976 (median $26; range $1-$34,050).
- Food and beverage expenses constituted the vast majority of disclosed transfers (4,111; 83 percent), followed by travel and lodging (444; 9 percent), consulting fees (279; 6 percent) and educational expenses (51; 1 percent).
“In 2014, only a small minority of United States radiologists received non-research payments from industry,” Kokabi et al. conclude. “At the state level, medical imaging utilization does not seem to be influenced by such financial relationships.”
The authors list several limitations in their study design. Among these was their inability to identify specific radiologists who may fall into “safe harbors” allowing the ordering of imaging for Medicare beneficiaries in nonhospital settings. This inability, they note, owed to the nonstandardized way in which Open Payments classifies specialties.
“The fact that we found no associations at all, however, suggests that the frequency of referrals by this narrow group of providers is minuscule at best,” they write.
The authors offer as a take-home point their finding that non-research industry payments to radiologists are “overall infrequent and relatively small.”