Researchers surveyed a national sample of medical students and residents in 2011 about their dealings with drug companies. In 2024, they went back to the same people, now practicing physicians, and asked again. Free drug samples had nearly doubled, conversations with pharmaceutical representatives had risen by half, and a quarter of respondents could not say whether their own institution had a policy governing industry contact. The findings appear in Clinical Pharmacology & Therapeutics, published online on August 9, and were reported by STAT this month.
What the study measured matters as much as what it found. Researchers asked physicians what they received and where they get drug information. They did not examine anyone's prescriptions. So the honest reading is that industry contact remains routine and has shifted in shape, not that any individual doctor's judgment was bought.
For patients, the practical question is simpler than the ethics debate: when a clinician reaches for a particular brand-name drug, is there a cheaper or longer-established option that works as well? That is a reasonable thing to ask in the exam room.
What Changed Between 2011 and 2024
The study, led by Helen Mooney with senior author Aaron Kesselheim at the Program on Regulation, Therapeutics, and Law at Brigham and Women's Hospital, tracked individual responses across both surveys. Some categories barely moved. Food or beverages in the workplace stayed flat at about 30 percent, and branded items such as pens and notepads held at about 14 percent.
Other categories moved a lot. Free drug samples in the prior six months rose from 11.4 percent to 21.6 percent. Meals away from the office or hospital rose from 12.8 percent to 17.9 percent. Conversations with a pharmaceutical representative about a product rose from 30.4 percent to 45.6 percent.
Two categories fell. Attending a company-sponsored lecture dropped from 25.0 percent to 14.9 percent, and observing a colleague interact with a representative dropped from 44.4 percent to 31.2 percent. The picture is not simply more contact. It is contact that has moved out of lecture halls and into one-on-one exchanges.
Where Doctors Now Learn About Drugs
The biggest shifts in how these physicians learn about medications were toward mainstream evidence. Reliance on clinical guidelines jumped from 20.2 percent to 55.0 percent of respondents saying they used them often, and peer-reviewed articles rose from 21.2 percent to 41.8 percent. Use of textbooks, lecture materials, Wikipedia, and Google all fell.
Alongside that, use of industry material also grew. Respondents saying they sometimes referred to information provided by pharmaceutical representatives rose from 8.5 percent to 21.9 percent, and use of drug company materials rose from 8.1 percent to 13.8 percent. The authors note this matters because promotional materials have been shown to favor the product being advertised.
On policy, 64.8 percent said their institution had a policy regulating industry interactions, 8.9 percent said it did not, and 26.3 percent did not know. Most respondents supported tighter limits: 84 percent said receiving food and gifts from drug companies was not important to them, and 38.5 percent said health system leaders should never be paid to sit on drugmaker boards. Physicians who had received gifts in the prior six months were less likely to favor tighter limits and more likely to work where no policy existed.
The Limits of a 33 Percent Response Rate
The response rate is the study's central weakness, and the authors say so. The 2011 survey drew on a random national sample and included more than 2,300 trainees. Researchers had usable 2011 contact details for 1,130 of them; 901 were reachable in 2024, and 294 responded, a 33 percent rate. After exclusions, 283 were analyzed. The published paper notes the responding group included more women and more white physicians than the US physician workforce, and fewer physicians in private practice.
Everything here is self-reported and subject to social desirability bias. The authors deliberately did not cross-check responses against federal payment records, because participants were not told in advance that their answers would be compared that way. Respondents received a $50 gift card for participating.
The study also cannot separate two explanations for the change. Some of it may reflect broader shifts over 13 years, and some may simply reflect what happens when trainees become attending physicians with prescribing authority. The authors say both are plausible.
Context from other work is worth keeping separate from this study's findings. Earlier research has linked industry-sponsored meals to Medicare prescribing patterns, and a separate analysis found higher industry payments among men, later-career physicians and specialists. Free samples cut both ways: they have been associated with preferential brand-name prescribing and higher long-term costs, but they also let a clinician hand medication to a patient who cannot fill a prescription that week.
Disclosures belong in a story about conflicts of interest. The work was funded by Arnold Ventures. Kesselheim reported consulting for Alosa Health, a nonprofit that produces non-industry drug education, and serving as an expert witness for a payer in a case against Teva over the marketing of glatiramer and for attorneys general and payers in a generic price-fixing case. Mooney divested holdings in several drugmakers before joining the research program in 2023 and has received fees from Alosa Health.
Three Checks Patients Can Run Today
The federal Open Payments database, run by the Centers for Medicare and Medicaid Services, lets anyone search a physician by name and see payments and transfers of value reported by drug and device manufacturers, including meals, travel and speaking fees. A listed payment is context, not evidence of improper care.
Patients can also ask whether a generic or older, well-established alternative exists for a newly prescribed drug, and what the out-of-pocket difference would be. Pharmacists are often better placed than prescribers to answer the cost half of that question.
Anyone handed samples should ask what the medication will cost once the samples run out and whether their plan covers it. That prevents the common sequence in which a patient starts a drug free and then finds it unaffordable. Nobody should stop a prescribed medication based on a payment record. Concerns belong in a conversation with the clinician or a second opinion.
The most actionable finding is institutional rather than individual. If a quarter of physicians cannot say whether their employer has a conflict-of-interest policy, existing policies are not being communicated, and health systems can fix that without new legislation. Federal reporting under the Physician Payments Sunshine Act remains in force.
Key Questions Answered
What did the study find? Among 283 physicians resurveyed 13 years after taking part as trainees, free drug samples rose from 11.4 percent to 21.6 percent and conversations with drug representatives rose from 30.4 percent to 45.6 percent, while sponsored lectures fell.
Does it prove gifts changed anyone's prescribing? No. The survey measured what physicians received and which information sources they used. It did not examine prescription records, so it cannot establish an effect on prescribing for any individual.
Did doctors move away from independent evidence? No. The largest increases were in the use of clinical guidelines and peer-reviewed articles. Use of industry-supplied material grew alongside that, not instead of it.
How reliable is the sample? Modest. Of 901 reachable participants, 294 responded, and 283 were analyzed, a 33 percent response rate. The group skewed toward women, white physicians, and non-private practice compared with US physicians overall.
Are free samples bad for patients? It is mixed. Samples have been linked to preferential brand-name prescribing and higher long-term costs, but they also help uninsured patients start treatment they could not otherwise afford.
Can patients look up their own doctor? Yes. The CMS Open Payments database lists manufacturer-reported payments by physician name. A listed payment is context, not evidence of improper care.
