What One Forbidden Question Revealed About Physician Decisions and How It Contributed to the Fastest Launch in Industry History

Vertex was preparing to launch Incivek, the company's first product and the first protease inhibitor for the treatment of Hepatitis C. It was a breakthrough. In the clinical trials it delivered materially better outcomes than the standard of care.

I was in Boston, interviewing a gastroenterologist. Fifteen people from Vertex were watching from an observation room behind a one-way mirror.

There was a catch. The protease inhibitor still had to be co-administered with interferon, and interferon was a miserable product for patients to take. The side effects were significant and unrelenting, and managing patients through them was a real burden on the physicians treating them. Incivek's clinical advantage was only accessible to patients who were willing to endure the interferon.

That was the launch challenge, and it was what the gastroenterologist across from me was describing. From the early questions in the interview, I could tell he was very smart and very pragmatic and he was complaining. Treating Hep C patients was a real burden to the practice. The side effects were exhausting to manage. His patients were unhappy and so was he.

I listened. Then I set the discussion guide aside.

"I'm listening to you complain about treating these Hep C patients," I said. "I run a business. You run a business. If they're such a pain in the ass to treat, why do you do it?"

The door to the interviewing room opened. A note came through telling me to stop the interview and come to the back room.

I did. The concern of the Vertex observers was immediate and reasonable — the question had nothing to do with the drug, per se. It wasn't on the discussion guide. It wasn't why we were there. I pushed back, quietly. I asked them to let me continue. Cautiously, they agreed.

I went back into the room. The physician laughed. "They didn't like that question, did they?"

Then he told me the truth.

“I treat Hep C patients because most general practitioners have one gastroenterologist they refer to. If I stop treating the Hep C cases, the GPs might send their scope patients elsewhere too. My own partners don’t want me taking on too many of them — relative to the time they consume, Hep C patients are a money loser for the practice. So I compromise.  I treat them, but I space the appointments out. If you called in January, I will see you in July.”

That was the answer. It made total sense from a business point of view and balanced clinical benefit against practice economics. It was invisible from anything the clinical script would have surfaced.

At the conclusion of the project, I told Vertex what I thought it meant. If we didn't take these patients off the physician's burden, the launch was going to stall. The doctors wanted the drug. The system around them couldn't absorb the patients fast enough to prescribe it at scale.

Vertex worked out an arrangement with a third-party group of nurses to manage patients' side effects. Physicians could prescribe Incivek without absorbing the operational cost of managing the interferon-driven side-effect burden. The bottleneck moved. Patients who had been spaced out into next year could be treated now. The launch that followed was the fastest in industry history at the time.

I've thought about that Boston interview a lot over the years, because it's a compact illustration of something I only fully understood later. The question I asked wasn't a hunch. It was a deliberate reframe. I stopped talking to him as a clinician and started talking to him as a businessman — and once I did, he answered as one. Everything he said after that point was information no clinical interview would have surfaced, because clinical interviews don't ask.

The Boston interview wasn't an isolated case. Over years of similar conversations — in oncology, cardiology, urology, primary care, across companies I worked for and consulted to — I found the same three questions surfacing again and again beneath whatever the discussion guide said we were there to discuss. Physicians weren't evaluating new products the way clinical trials evaluated them. They were running a quieter calculation, mostly without articulating it. Once I saw the shape of that calculation often enough, I organized it into a model I call RAMPx, for Realistic Assessment of Market Potential. It's the healthcare-specific cousin of the Diffusion of Innovations theory that Everett Rogers introduced sixty years ago. The three questions didn't arrive fully formed — the third one, in particular, took years to see clearly — but here's the shape they eventually settled into.

1. How much better is the new product than the standard of care? This is the dimension every launch team wants to talk about, and it's the one that gets measured most rigorously in the clinical trials. It matters. But by itself it doesn't predict much.

2. How easy is the product to incorporate into a physician's practice — or a patient's lifestyle? This is where most launches fail. Ease of adoption isn't just logistical. It's economic. It's operational. It's about who absorbs the cost of change. The Boston interview is a case in point: Incivek's ease of adoption was constrained not by the drug itself but by the interferon it required, by the partnership economics of the practice absorbing the patients, and by the referral system the gastroenterologist was quietly protecting. All three constraints were invisible from any clinical data set. All three shaped how fast the drug could actually reach patients.

Incivek wasn’t the only example. Years earlier, I'd interviewed a urologist about the impending launch of Proscar, a drug that reduced prostate size and might also reduce the number of patients requiring a surgical procedure that his practice depended on. He held up his index finger and waived it in my face, told me “this finger is insured by Lloyd's of London", and said the Proscar would reduce his cash flow. He wanted it to exist. He wasn't happy about it, even as I told him in the long run it could increase his volume. That's ease of adoption from the physician's economic perspective — and no clinical trial captures it.

3. How much risk is the physician willing to take to be an early adopter? Another physician once told me a new product was better than what he was using. He said he'd wait to prescribe it anyway. “I don’t want to sit on a witness stand in a malpractice trial explaining to a jury why I tried the new drug rather than the one most commonly used. I’d rather wait.” Early adoption is a personal wager the manufacturer isn't paying for. Understand that and you understand why some clinically superior products are adopted slowly after approval.

The interesting thing about the third dimension is that it isn't fully independent. Risk tolerance is a function of the first two: how much better the product is, divided by how easy it is to try. A big improvement that's easy to try is a small risk. A modest improvement that requires reorganizing your practice is a big risk. Physicians and patients are constantly running this calculation, mostly without knowing they're doing it, and the answer determines how fast a new product actually gets adopted.

Which brings me to the phrase I've come to organize my whole thinking around. Approval is not adoption. An approved drug is a drug the FDA has decided is safe and effective. An adopted drug is a drug that physicians actually prescribe, patients actually take, and systems actually accommodate. Those are two very different things. Companies routinely treat approval as the finish line and are then surprised when the launch curve underperforms. The launch curve is being determined by the three questions above, and most of the answers were knowable long before the approval letter arrived.

The RAMPx framework didn't come from a whiteboard. It came from a gastroenterologist in Boston explaining why he made his patients wait six months, a urologist waving his insured finger, and a physician who was more worried about a jury than about being right. Every one of them was answering one of the three questions. Every one of them was telling me something that would have been invisible if I'd stayed on the discussion guide. And in hundreds of interviews conducted with physicians and patients, I hear the same themes over and over again.

My new book coming out in September takes its title from the phrase: Approval Is Not Adoption. It's the longer treatment of the model and what to do with it. This piece is the elevator version.

The one-line version is even shorter. If you want to know whether a new product will succeed, don't just ask whether it works. Ask whether the people who have to prescribe it, use it, and pay for it can afford to say yes.