(iStock)

My first “pharma lunch” was unimpressive: cold lasagna in the back office of the clinic where, as a medical student, I was shadowing a doctor. As I scarfed down pasta in between patients, I asked a secretary where it had come from. “Another pharmaceutical rep,” he replied. He couldn’t remember the name of the drug being promoted. But he knew that it was the second time that week that an industry rep had brought Italian food.

The neurologist I was shadowing knew the name of the drug but said, “I’d never prescribe it. It doesn’t work as well as the older one.” She was beating the system: Free lunch with no strings attached.

If this was true, she was an outlier among physicians. Two studies published recently in JAMA Internal Medicine shed light on the influence of lunches sponsored by drug firms.

The first looked at Massachusetts physicians who prescribed statins for Medicare patients. The doctors who received industry payments (usually company-sponsored meals valued at less than $200) were slightly more likely to prescribe more-expensive brand-name cholesterol medications over generics.

The second study was perhaps more eye-opening. Physicians who had received even one ­company-sponsored meal promoting a brand-name drug were twice as likely as other doctors to prescribe more-expensive drugs over generics. The average cost of a meal in this study was under $20. That’s a small price to pay to convince a physician to prescribe a drug that costs $5 a pill. The study suggested that contact with an industry representative — not the actual value of the lunch — was what influenced physician behavior.

These studies question the effectiveness of a decade-long effort to extricate industry meals from medical practice. Until the 2000s, pharmaceutical companies marketed drugs directly to doctors, with no rules against it.

According to a report from the ZS Associates, a sales and marketing firm, the U.S. drug industry nearly doubled the number of sales representatives (from 40,000 to 80,000) between 1996 and 2001 despite a declining number of new drugs entering the market. Meals were the most common service provided. A 2007 study in the New England Journal of Medicine found that 94 percent of all physicians had some relationship with drug manufacturers; 83 percent of those relationships involved receiving food or drinks in the workplace.

Industry gifts came in all sizes. While growing up as the son of a physician, I often found pens and stress balls promoting drugs I could barely pronounce. Other physicians received sports tickets, golf clubs and even paid vacations in exchange for speaking services.

But things began to change in 2002 when the Food and Drug Administration forced the withdrawal of the painkiller Vioxx after researchers found it was associated with cardiovascular problems and death. Vioxx had become a hugely successful drug on the basis of strong company promotion and large-scale prescription by doctors. In 2003, the Office of Inspector General of the Department of Health and Human Services released general guidelines for pharmaceutical marketing practices, including meals to physicians — practices that the HHS office determined had led to the overprescribing of Vioxx and other drugs. PhRMA, the largest trade association representing pharmaceutical manufacturers, then came up with voluntary guidelines for interacting with health professionals.

National efforts to ban ­industry-sponsored meals suddenly became all the rage: A group of health-care providers started No Free Lunch, an organization that asked doctors and medical students to pledge not to accept gifts of any sort from pharmaceutical representatives. (Somewhat ironically, the group provided pens, buttons and even coffee mugs to doctors holding informational lunches about drug-company influence.) In 2009, the Institute of Medicine issued a report calling for all doctors to “forgo gifts of any amount from medical companies,” including drug manufacturers.

These attitudes made their way into medical schools and doctor’s offices. Responding to student demands, several medical schools, including Harvard (where I studied), Stanford, the University of Pennsylvania and Yale enacted policies requiring faculty members to disclose conflicts of interest before presentations and banning all free industry meals. Suddenly, all of my lectures were preceded by the “Conflicts of Interest slide.” Disclosure became so common that many of my friends were no longer interested when a lecturer had a potential conflict.

Despite all this, the pharma lunch lives on. Medicare’s Open Payments website, which lists all health-care providers’ financial transactions with the makers of drugs and medical devices, released data in July showing that these industries made nearly $24 million in food and beverage payments to physicians in 2015 — a figure unchanged from 2014.

But the real question isn’t whether industry-sponsored meals are still too common or even whether they influence prescribing habits. What matters is whether pharma lunches cause poor care or unnecessary health spending.

You would be hard-pressed to convince doctors that they are overly influenced by the meals they accept. After all, in the statin study mentioned earlier, every $1,000 in payments to physicians increased brand-name prescribing by only 0.1 percent. And these studies show associations, not causal relationships.

It is, of course, possible that doctors who accept pharma lunches are already inclined to prescribe more brand-name drugs. And there is still no convincing data that drug-company meals cause significant increases in costs or decreases in quality of care.

Pharma lunches — which, in addition to free food, include information about the drugs being promoted — may have value. In an era of constantly evolving medical knowledge, pharmaceutical companies play a leading role in educating doctors. And as the Institute of Medicine said in its 2009 report, medical training without pharmaceutical influence “may involve higher costs for physicians and require cost- ­cutting steps by education providers.”

Industry-sponsored meals ­often provide valuable information that doctors otherwise might have very little exposure to. As Mike Ybarra, a physician who is senior director of alliance development at PhRMA, told me, “If a doctor doesn’t get exposed to new therapies, they are less likely to be aware of these as available ­patient-treatment options.”

Ybarra further noted that brand-name drugs may have specific indications for certain patient populations. Recent studies “don’t differentiate between good and bad prescribing,” he said.

As a newly minted physician, I can testify that it is less than ideal that the current choice for doctors’ education is between ­industry-sponsored free lunches and nothing at all. Other entities should provide educational programs — with or without meals — to physicians. Pharmacists not employed by industry may be most qualified to provide lectures on new medications.

But every potential source of education will come with a likely conflict of interest. Pharmaceutical companies want to sell their product. Hospitals want to cut costs. Researchers want to promote their discovery. I am only beginning to realize these complex incentives. Perhaps the best I can do is to recognize the potential biases of these sources — and curb any tendency to respond to them. Then I might not feel so guilty about eating my free lunch.

Parikh is a resident in internal medicine at Brigham and Women’s Hospital and a clinical fellow in medicine at Harvard Medical School. Follow Parikh on Twitter.