We're Asking the Wrong Questions About Informed Consent

In July 2023, I learned the power of three simple words: I withdraw consent.

By that point, I had spent over a decade working in clinical research. I had guided thousands of participants through informed consent forms that all included some version of the same line: you may withdraw consent at any time, for any reason. Of course, that’s not what we hope people will do. In research, retention often matters more than recruitment.

But that summer day, sitting in a hospital room as clinicians debated how to move Rona from one study to another, they were balancing protocol constraints, data integrity, and eligibility criteria. I realized something they hadn’t yet said out loud.

“Okay,” I said. “So I withdraw consent.”

The room went silent. Providers exchanged glances. In a single sentence, I had cleared a path that policy and process had made complicated. We signed a new informed consent form a few days later.

That moment is why improving informed consent for families is a hill I will die on.

Too often, the conversation about consent focuses on simplification: shorter documents, lower reading levels, cleaner formatting. Those are worthwhile goals, but they miss the deeper issue. In many cases, especially in rare disease, a clinical trial is not one option among many. It is the only option. When that’s true, “consent” becomes complicated. The existence of a trial as the sole path forward introduces a kind of structural coercion that bioethics has long acknowledged (1, 2).

In those moments, informed consent is less about consent, because the alternative may be no treatment at all, and more about the informed.

If we take that seriously, then improving consent isn’t about making forms shorter. It is about creating real opportunities for understanding. Right now, we are working hard to optimize the mechanism by tweaking forms, adjusting readability scores, and digitizing signatures, without questioning the framework itself.

What if the framework is the problem?

What if, instead of asking how to improve the form, we asked what role the form should even play?

Because that form is not primarily for families. It is for regulators, institutions, and compliance systems. The conversation is for participants. Consent is not a document. It is a process grounded in clarity, context, and trust (3, 4).

I signed essentially the same set of consent forms four times. The first time, I asked technical questions for over an hour. The answers were detailed and supported by printed data on long-term risks and outcomes. At that point, we believed there was a 90% chance Rona would survive.

The fourth time, we believed there was a 90% chance she would not.

I asked one question: “Is this going to work? Because it feels like she’s going to die.”

The answer: “It’s her best chance.”

And then, gently: “Shhh, mama. We hold out hope until the very end.”

That is what informed consent looks like at the bedside. It is not a checklist, a reading level, or a signature. It is a human conversation held in uncertainty, shaped by fear, hope, and the weight of impossible decisions.

And no chatbot, no matter how advanced, could have replaced that moment.

If we want to truly improve informed consent, we need to stop centering the form and start centering the human experience of understanding.

Selected references:

1. Appelbaum, P. S., Lidz, C. W., & Klitzman, R. (2009). Voluntariness of consent to research: a conceptual model. Hastings Center Report.

2. Largent, E. A., Grady, C., Miller, F. G., & Wertheimer, A. (2012). Misconceptions about coercion and undue influence. IRB: Ethics & Human Research.

3. Faden, R. R., & Beauchamp, T. L. (1986). A History and Theory of Informed Consent.

4. National Institutes of Health (NIH). (2016). Clear Communication in Informed Consent.cs & Human Research.