Researchers correlate brand-name prescribing with industry-funded lunching

A group of academic medical experts, digging into Physicians’ Sunshine Act data from 2013, finds a statistical correlation between money spent for lunches and the volume of prescribing of those physicians; as with similar studies, however, the authors note that “The findings represent an association, not a cause-and-effect relationship.” Whereupon the editor of JAMA Internal Medicine, where the study appeared last week, editorialized that “Is it necessary to prove a causal relationship between industry payments to physicians and the prescribing of brand name medications?  … [I]f drug and device manufacturers were to stop sending money to physicians for promotional speaking, meals and other activities without clear medical justifications and invest more in independent bona fide research on safety, effectiveness and affordability, our patients and the health care system would be better off.”

Would that this were so. The JAMA Internal Medicine article was quickly criticized by, among others, Bill Coyle, a principal at ZS Associates, the sales/marketing consulting firm that works closely with pharma manufacturers. Coyle points out that the study was based on four relatively seldom prescribed products (Benicar, Bystolic, Prestiq and, perhaps the best known of the lot, Crestor—which had around 9% of the statin market back then but has since gone generic). These drugs corresponded with a relatively small number of physicians receiving meals for them—only 6% of all physicians receiving meals. Most questionable is whether a slew of confounding factors--rep visits, peer-to-peer meetings and samples to help patients, for example—occur in tandem with the meals. “Without factoring in these other activities, a correlation between meals and prescriptions is grossly and incorrectly exaggerated,” says Coyle.

The JAMA study makes many of the same points as the one published by ProPublica, which has been maintaining its “Dollars for Docs” reports since even before the Sunshine Act. Coyle points out that missing from these analyses is the question of educational value from physician-rep interactions; even ProPublica notes that speaking engagements have significantly dropped since the public reporting began (and to believe that all physician speaking engagements have zero educational value is to impugn physicians themselves).

Meanwhile, the bigger healthcare issues dwarf the free-lunch-for-docs question. Billions are spent on patient copay coupons (a bugaboo of the pharmacy benefit managers), and billions more given up by pharma companies formulary negotiations with those same PBMs. Pharmacy dispensing costs (or rebates obtained by healthcare systems managing that dispensing) are additional billions. When it comes to drug costs, there are many places to look.