Ethical Choice Architecture in Clinical Trials.

Every form, sequence, reminder, and handoff shapes behavior. The ethical question is whether that design improves agency or quietly replaces it.

Choice architecture is always present.

There is no neutral experience. The order of information, the wording of buttons, the availability of a human contact, and the effort required to decline all influence what people do. Ignoring these choices does not remove their effect. It leaves them ungoverned.

Behavioral design becomes valuable when it reduces avoidable work while keeping the decision visible and voluntary.

Research consent has hard boundaries.

The Belmont principle of respect for persons and FDA informed-consent requirements establish that participation must be voluntary and informed. Designs that preselect participation, obscure alternatives, exaggerate benefit, or make refusal harder than acceptance are incompatible with that standard.

Use design to support action around the decision.

Ethical applications can reduce friction without determining the answer:

• Let patients choose communication channels and change them later.
• Offer reminders for appointments the patient has already chosen.
• Save progress so information can be reviewed over time.
• Provide a clear next step for questions or human support.
• Make declining or withdrawing as understandable as continuing.
• Use small voluntary steps to prepare for participation after consent.

Every intervention needs an autonomy test.

Before using a behavioral intervention, teams should ask whether it is transparent, reversible, proportionate, easy to decline, and aligned with the patient's interests. They should also examine whether the design affects vulnerable groups differently.

A useful metric is not only whether the intervention increases progression. It is whether patients understand the choice, retain control, and report that the process respected them.

Sources.

1. HHS OHRP, The Belmont Report
2. FDA, Informed Consent Guidance
3. NIH, Guiding Principles for Ethical Research
4. Stacey et al., Decisional needs systematic review

About the author

Andy Proctor

Head of Behavioral Science

Andy is a social and health psychologist and mixed-methods researcher whose work examines social connection, health behavior, psychophysiology, and human interaction with AI. At Jumo, he translates cognitive, emotional, social, and practical friction into experiences and interventions designed to support durable participation.