Jodi* had been diagnosed with ovarian cancer several years earlier, had received adjuvant carboplatin and paclitaxel therapy, relapsed three years later, and since then, had been on several forms of therapy—most recently receiving weekly paclitaxel. She was tolerating treatment well, but a CT scan done to re-evaluate her extent of disease (EOD) showed that the disease had progressed.

She took the news of her disease progression without flinching; she had long made the mental switch of seeing her cancer as a “chronic disease,” rather than one that was “life-threatening.”

“Well, here we are again,” she said. “What’s up next?”

With that, I laid out some options, both standardly available and ones on clinical trial. We talked about each one briefly, and then I put forth my recommendation.

“This time, I want to try bevacizumab. I had become accustomed to a certain pattern where she would leave it up to me to decide on next steps, so without thinking about it, my next statement to her was a question: “When would you like to get started?” I expected her to retrieve her calendar and say, “Fine, let’s start on this day.” However, her next question broke the pattern I thought we had established.

“That’s fine, but I want to know why you recommended this,” she stated.

“What do you mean, why?” I echoed, as I looked at her slightly taken aback.

“I want to understand why you are recommending this particular agent. The side effects of this drug are scary—even if you don’t think so. Friends of mine who’ve taken it have gotten hypertensive, and I would really prefer not starting a new drug for a new condition caused by this medication,” she said.

At that moment, I feared I had done something I have strived not to do—I was treating her paternalistically. Indeed, I questioned whether I had been imposing my views all along, superseding her wishes, all in the name of “doctor knows best.”

Paternalism as a concept runs counter to modern medical practice, which has emphasized patient autonomy and the individual right to self-determination. Still, I have come to understand that not all patients want this role. Hitz and colleagues surveyed over 400 patients with advanced cancer, all of whom were initiating a new treatment regimen.[1] Their goal was to seek determinants of satisfaction with their treatment decisions (SWTD). What I found most surprising is that decision-making preferences were evenly split: 45% preferred a shared-decision making format while 44% preferred the doctor to direct it.

So, here I was—outlining options and then mapping a way forward for her without much of her own input—a model of doctor-directed decision-making. Although it had been the approach we took in the past, she showed me that preferences are dynamic, and that I always need to be prepared to justify my own recommendations. After all, I was not treating cancer; I was treating a person living with this cancer, at risk for toxicity related to my recommendation—both physical and financial. At the end of the day, all I had to do was write an order for it. She, on the other hand, would be truly living with the impact of “my” decision.

Her question of “why” was not only fair, it was entirely justified. We owe it to our own patients to explain our decisions, help them navigate the medical choices in a way that makes sense for them, and when we do recommend a treatment program, it is our duty to be able to justify it. While it is ideal that our decisions be based on evidence, preferably gained from large randomized trials, I know all too well that so much of what we do in oncology is not backed by that level of data. But you know what? I don’t think our patients require that degree of evidence. In my experience, they simply want to know that our decisions were thoughtfully considered—with them in mind.

So with that, I did the best I could. I looked at her and began, “Well, I think this is the best choice for you because . . .”

*Name and descriptors changed to protect patient identity.