- Doctors who received general payments from oncology drugmakers were twice as likely to prescribe a cancer medicine made by that company, according to a new study published in JAMA Internal Medicine.
- Researchers used data from the Centers for Medicare & Medicaid Services Open Payments website from 2013 and 2014 to gauge whether general payments such as gifts or speaker fees influenced prescribing.
- Lead study author Aaron Mitchell said they also found prescriptions increased as the amount of money physicians received went up. Payments for research, by contrast, were not associated with statistically significant differences in prescribing.
Though many studies have linked payments from drugmakers to increased prescribing, the JAMA Internal Medicine study by researchers at the Vanderbilt University and the University of North Carolina contends to be the first to find an association between industry payments and prescribing for cancer drugs specifically. The researchers also say they are the first to examine how type of compensation — money paid for research versus general payments, for example — influence prescribing patterns.
Data for the analysis were only collected from Medicare, the authors told BioPharma Dive, because the older patient group is at the highest risk for most cancer types. In addition, CMS has the most cleanly aggregated data across all patients and physicians.
The researchers restricted their investigation to cancer treatments for metastatic renal cell cancer (including sorafenib, sunitinib malate, and pazopanib hydrochloride) and for chronic myeloid leukemia (dasatinib, imatinib mesylate, and nilotinib hydrochloride monohydrate).
Investigators chose these specific cancer types because there are several similar medications to choose from within each indication.
"We wanted to look at cancer types where the different treatments would all be considered as 'standard of care' based on FDA approval and [National Comprehensive Cancer Network] recommendation," said Mitchell. "RCC and CML (and the associated drug sets) best met these criteria."
Some were skeptical of the potential takeaways.
"It's not so straightforward to say that dollar-values listed by companies in the interest of newer transparency initiatives means that those dollars are payoffs, though now that the public has access to these data, it's become more common" for people to conclude there is a cause and effect relationship at play, said Ben Locwin, a consultant to the pharmaceutical industry.
A major limitation of the study is the fact that records in the Open Payments database can be inaccurate. Payment reporting is the responsibility of drugmakers, which the research team acknowledged could be problematic.
For example, if a physician were to eat food a manufacturer purchased but left in a conference room, it could be possible the physician did not interact with a pharma rep but still may be reported as receiving a payment. Mitchell admitted this type of scenario could potentially distort the perceived correlation.
"To the extent that this actually occurs, it would be expected to skew our results towards the null, as some percent of physicians would be listed as receiving payments but would then (presumably) not have any change in prescribing, if they did not actually come in contact with the reps," Mitchell explained. "I would not think this effect would change the overall takeaway of our findings, given that we did find a statistically significant association for general payments."
But, Mitchell also conceded he is not aware of any regulation that specifies whether all physicians in a practice have to be listed as receiving a payment, or if it only applies to those who actually had a confirmed interaction with a rep. Mitchell said it depends on how diligent reps are about reporting. "It certainly would be possible for a physician to be listed there for something that was given to other practice members," he said.