A recent study in the journal JAMA Internal Medicine shows that drug companies are bribing doctors with food to prescribe their expensive, brand-name medicines.
At least that’s the way it’s been reported in the media. Dig deeper into the study, however, and it becomes clear that doctors aren’t selling out for sandwiches. There’s no proof that industry-sponsored meals drive prescribing behavior. In fact, the relationships between pharmaceutical sales representatives and physicians may actually benefit patients.
The study reported that physicians who received even one industry-sponsored meal were more likely to prescribe the brand-name medicine that was being promoted. Receipt of additional meals was associated with greater increases in relative brand-name prescribing rates. And receipt of costlier meals was associated with increased target drug prescribing for three of the four drugs studied.
These findings were based on flawed methodology. The study examined physician prescribing over the entire year of 2013 but only looked at meals provided in the last five months of 2013. As the authors conceded, “The findings reflect an association and not necessarily causality.”
Indeed, it is hard to fathom how lunch in October could have persuaded a physician to prescribe a particular drug the previous February.
The overwhelming majority of the studied physicians did not receive any industry-sponsored meals promoting the studied drugs. Among those that did, nearly all had a single meal. The notion that the pharmaceutical industry is buying off physicians en masse is hyperbole.
So are reports of industry largesse. The average values of the sponsored meals for the four study drugs ranged from $12-$18 — hardly lunch at the Four Seasons. When the authors arbitrarily set the threshold for expensive meals at $20 they presumably found costlier meals were associated with increased prescribing.
But this claim was based on “predicted probabilities” for selected physicians who are most likely to prescribe rather than actual probabilities for all physicians. And close examination of the study graphs only showed significant differences in prescribing based on meal cost for one of the four drugs, not three out of four as claimed by the authors.
This new study is not the first to connect drug industry gifts and physician prescribing behavior — an editorial accompanying the study cited two recent publications. But the link was as tenuous in those studies as in this new one.
The first was a 2011 Massachusetts study which found a positive association between industry payments and higher rates of prescribing brand-name statins. But it was only significant for physicians who received at least $2,000 from industry and the effect was miniscule — for every $1,000 in payments received, the brand-name statin prescribing rate increased 0.1 percent.
In the second study, ProPublica looked at 2014 contemporaneous payment and prescribing data and found that physicians who received any kind of industry payment, even just meals, had higher rates of brand-name prescribing. But the study didn’t link payments for specific drugs to increased prescriptions for those drugs. Nor did the study link payments by particular companies to increased prescriptions of those companies’ drugs. As ProPublica’s authors acknowledged, “We have not established…a causal link. …There is certainly a plausible reverse causation here — that drug companies give money to physicians who already are heavy brand-name prescribers.”
Physicians who eschew interactions with pharmaceutical representatives could be under-prescribing brand-name drugs and harming their patients. Industry-sponsored meals are merely an entry point for drug representatives to present product information to physicians. If that presentation is truthful there is no harm. The new study’s authors admitted, “If events where industry-sponsored meals are provided affect prescribing by informing physicians about new evidence and clinical guidelines, then the receipt of sponsored meals may benefit patient care.”
There are many reasons brand-name drugs are prescribed. Well-fed physicians are not among them.
Joel Zinberg, M.D.
American Enterprise Institute
The opinion page features letters to the editor. To submit a letter, email firstname.lastname@example.org. Letters are not the views of the Queen Creeek Independent or Independent Newsmedia Inc USA.