For every dollar the pharmaceutical industry spends on research and development, it spends two on marketing. When I worked as a pharmaceutical sales representative in the neuroscience division of Eli Lilly, I was the living embodiment of this investment. I took doctors out to so many fancy Manhattan restaurants that the maitre d’s greeted me by name. The company hosted them at catered “speaking programs” and gave away tickets to baseball games and Broadway musicals. We even sent doctors and their families to sponsored academic conferences at tony resorts in Florida and California. During the day, if doctors didn’t have time to see me, I chatted up their receptionists, plying them with food and gifts (stress balls, umbrellas, clocks) and asking, breezily, which medications their bosses preferred prescribing, and why.

Nowadays, it’s not all fine wine and caviar. Since I left the industry in 2000, pharmaceutical marketing has changed — at least on the outside. After years of bad press, and with mounting fears of a regulatory crackdown, drugmakers adopted a voluntary code of conduct in 2009 that, among other restrictions, permits only “modest, occasional meals” in “appropriate circumstances,” facilitating “the exchange of medical and scientific information.” Though this curbed the most ridiculous excesses — no more “unrestricted grants” and way fewer free pens — research shows that even a $20 lunch can sway prescribing behavior. The fundamentals of the business remain the same: A good sales rep builds credibility with doctors by using scientific-seeming language to push the product and makes them feel subtly obligated to write more scrips.

Over the years, as I’ve followed the revelations about opioid-makers and how they fueled an epidemic of addiction, I’ve recognized the aggressive marketing tactics that I was trained to use. Now I see the effects first-hand as an emergency room doctor: A week doesn’t go by without my having to resuscitate someone who has overdosed. When my patient’s loved ones ask how this could have happened, or start to blame themselves, I think with anger about how, for years, drug companies promoted the idea that addiction was not a common effect of the medication but rather a personal failing of the addict.

A JAMA Network Open study of more than 2,200 counties found that the more marketing interactions doctors had with opioid-makers, the more painkiller prescriptions were written and the more deaths there were from overdoses. The researchers also found that opioid-makers including Insys Therapeutics, Teva Pharmaceuticals and Janssen Pharmaceuticals were liberal with those free meals, as well as with consulting fees, “education” programs and free travel — all of which are linked to increased opioid prescriptions. But food and payments aren’t the only ways pharma reps get products in front of medical providers. In my experience, sometimes the “gifts” were less literal: They included simple friendship.

When I was a sales rep, I was typically the one visitor who came to doctors’ offices without a complaint in need of fixing. I was the guy who broke up their usual drudgery with a friendly chat, asking them about their lives. With some doctors, just being genial was enough. With others, I brought sheaves of medical journal articles that supported the efficacy and safety of my drugs. Depending on their personalities, I would present these findings soberly or briskly or even somewhat incompetently. With a certain kind of doctor, playing dumb and having them explain the article to me, deferring to their intellect and expertise, proved much more effective.

Some of the doctors in my territory seemed to think of themselves as impervious to my charms. They acted as if they just nodded along, they could get my free samples and rush me out the door, no harm done. They didn’t realize how easily I could find out if they’d lied about how many scrips they would write. I’d be back the next week, guilt-tripping them about why they had underdelivered. That’s because, in addition to the extensive personal information we gathered — psychological profiles, their kids’ birth dates, who was having marital problems — sales reps had access to extensive databases about doctors’ prescribing habits. Private health information companies compiled this treasure trove of data using records bought from pharmacies. Such databases, which have only grown bigger and more sophisticated in recent years, were essential to opioid sales: Drug companies could easily target the most frequent prescribers in a Zip code, zeroing in on the doctors who saw many chronic-pain patients at their practices — or who were just freer with their prescription pads.

I made use of this data by following up with the reluctant prescribers, nonchalantly mentioning that the supply of samples had dried up, or just guilt-tripping them. Even though doctors don’t technically owe sales reps anything, the latter tactic was surprisingly effective: Little favors have a subtle way of making the recipient feel obligated to reciprocate in some way. But simply getting in the door was a start: A concept I introduced in a casual pitch would later be echoed from the podium, during a formal address by a sponsored speaker. Repeat a marketing message enough times, through enough mouthpieces, and it comes to seem like established scientific consensus.

I helped recruit those speakers from among my high-volume prescribers. I’d look out for physicians who were Ivy League-educated and spoke well on their local lecture circuits — and who spoke well of our medications, parrying any critiques. If they were good, we would pay them thousands of dollars to lecture at national conferences, interlacing their clinical language with our company line. Some physicians stood to make hundreds of thousands a year through speaking engagements and “consulting fees.” These programs are key to establishing a drug’s legitimacy when it enters the market or when it begins to have competitors. Similarly, Purdue Pharma’s launch plan for OxyContin called for the company to use its “speakers bureau” of several thousand doctors to give promotional talks at hospital programs.

Sales reps have a vital role in their company’s synergistic marketing apparatus, which aims to permeate the medical establishment and control its discourse. Recently, former Purdue Pharma reps have come forward, describing how they played down addiction risks during sales calls. If patients reported withdrawal symptoms like nausea or the shakes, the reps were trained to call this “pseudo-addiction.” A doctor who was skeptical of this concept could find studies to back it up — many of them conducted by J. David Haddox, Purdue’s scientific adviser. His research claimed that these symptoms resembled drug-seeking behavior but were actually caused by unrelieved pain. This pain, of course, could be treated by increasing the opioid dosage; these patients should not be diagnosed with addiction. Though “pseudo-addiction” became a popular term in medical literature, there has been no empirical evidence backing up the concept.

After my experience selling Prozac and Zyprexa, I know how easy it is to cherry-pick studies and present them as straightforward, neutral information: A review of industry-sponsored research on SSRI anti-depressants in the 1990s revealed that the published literature exaggerated how many clinical trials had positive results. Over the past 30 years, opioid-makers built their case for the long-term safety and non-addictiveness of their products on a thin scientific foundation. Based on an innocuous five-sentence letter to the New England Journal of Medicine, in which a Boston doctor observed that, according to his hospital’s case files, narcotic use rarely led to major addiction, Purdue Pharma claimed that less than 1 percent of patients taking opioids became addicted. That number was cited in lectures by pain specialists, and it became a mantra in videos designed for doctors to show their patients. But recent studies estimate that 21 to 29 percent of patients prescribed opioids end up misusing them, and the rate of addiction actually ranges between 8 and 12 percent.

An unspoken bedrock principle guided my actions when I was in the pharmaceutical industry: It was not enough to grow our market share — we had to grow the market, too. This dynamic played out during the early 2000s, when the American Pain Society and the Joint Commission started calling pain the “fifth vital sign.” Such organizations, whose boards included doctors who received consulting fees and honoraria from opioid-makers, circulated teaching materials designed by drug companies. Medical students and doctors didn’t just learn how to assess and pay attention to patients’ pain — they also internalized the idea that prescribing opioids was a professional, even an ethical, obligation. Exaggerating the clinical significance of pain drastically expanded the market for opioids, bringing them to populations with a high risk for addiction, like adolescents. From 2005 to 2015, nearly 15 percent of teens and young adults who went to the emergency room received an opioid prescription, according to a study in the journal Pediatrics. The prescribing rate was 38 percent for ankle fractures; for dental issues, the rate was 60 percent.

Physicians and sales reps are locked in a double delusion. When I was a drug rep, I really believed my pitch for our products — and I believed that by exerting influence over doctors, I helped patients access medicine they needed. As a doctor, I now have colleagues — colleagues with sharp, clinically trained minds and only the best of intentions — who think they write prescriptions on a wholly rational basis.

They don’t know what I know: that people are paid six figures and armed with fat expense accounts to make them feel confident that they’re acting without bias. In the case of opioids, this delusion has exacted a terrible human cost.

As told to Post editor Sophia Nguyen.

Read more from Outlook:

Follow our updates on Facebook and Twitter.