01 · The Question
How distressed is too distressed to continue an interview or survey?
A participant begins crying during an interview about bereavement. Another becomes anxious while completing questions about childhood trauma. Someone asks for a moment, regains composure, and says they want to continue. Another participant repeatedly says they are fine while becoming increasingly agitated and unable to answer coherently.
Researchers need something more useful than either “stop whenever someone becomes emotional” or “continue as long as the participant agrees.” Both rules are too crude.
Research ethics requires attention to autonomy and protection from harm at the same time. The Belmont Report describes respect for persons as recognizing autonomous choice while also requiring protection when autonomy is diminished, and beneficence as an obligation to minimize possible harms. The practical question is therefore whether the participant can continue safely, voluntarily, and meaningfully under the approved study procedures.
03 · What You Need to Know
The stopping decision depends on more than whether a participant looks upset
Emotion is not automatically evidence that participation has become unsafe
Sensitive interviews can produce sadness, tears, anger, embarrassment, or discomfort. These reactions deserve attention, but they do not all represent the same level of risk.
Automatically terminating an interview whenever a participant becomes emotional can itself undermine autonomy. The Belmont Report emphasizes that respect for persons includes giving weight to autonomous people's considered choices. At the same time, it recognizes that protection becomes increasingly important when a person's capacity for self-determination is compromised.
The researcher therefore needs to distinguish an emotional response from distress that is impairing the participant's ability to continue safely and voluntarily.
Emotional response
The participant may cry, become sad, pause, or express strong feelings while remaining oriented, communicative, able to make choices, and willing to proceed.
Concerning distress
The reaction is escalating, overwhelming, impairing meaningful engagement or decision-making, creating a safety concern, or meeting a protocol-defined stopping criterion.
Pausing and stopping are different decisions
A pause is often the appropriate first response when distress becomes noticeable. It interrupts data collection without prematurely deciding whether the session must end permanently.
During that pause, the participant can recover, ask questions, request a break, skip the topic, decide to stop, or indicate that they would like to continue. The researcher can also assess whether the participant's condition requires support or escalation.
Notice
Recognize meaningful changes in emotion, behavior, communication, or physical state.
Pause
Stop asking research questions rather than continuing while assessing the participant.
Check
Determine what the participant needs and whether there is an immediate safety concern.
Restore choice
Remind the participant that the available options may include taking a break, skipping a question, rescheduling, or stopping.
Decide
Resume only when continuation remains safe, voluntary, meaningful, and consistent with the approved protocol.
Stop if distress is escalating rather than resolving
A participant who becomes upset but settles after a pause presents a different situation from someone whose distress continues to intensify.
Concerning changes might include increasing agitation, inability to regain composure, progressively impaired communication, panic-like symptoms, disorientation, or another pattern indicating that continuing the research is worsening the situation. The relevant signs should be defined prospectively where foreseeable rather than improvised by individual interviewers.
The researcher's role is not to diagnose the participant from across the interview table. It is to recognize when the study's safety threshold has been reached and follow the approved response procedure.
Stop if the participant cannot meaningfully engage with the decision to continue
Voluntary participation requires more than obtaining consent once at the beginning. The Belmont Report treats voluntariness as a central component of informed consent and states that participants should be offered the opportunity to withdraw from research.
If distress has become so substantial that the participant cannot understand questions, communicate choices coherently, or meaningfully evaluate whether they want to proceed, repeatedly asking “Do you want to continue?” does not solve the problem.
A verbal “yes” has ethical force when it reflects a meaningful voluntary choice. It should not be treated as a magic word that removes every responsibility from the researcher.
Stop when the participant asks to stop or withdraw
This is the clearest threshold. A competent participant's request to stop the interview, survey, or research procedure should be respected according to the consent arrangements and applicable protocol.
The Belmont Report grounds informed consent in the participant's opportunity to choose what will or will not happen to them and identifies the opportunity to withdraw as part of the information ordinarily provided to research participants.
Researchers should also avoid creating friction around withdrawal. Repeatedly asking the participant to reconsider, emphasizing the value of completing the interview, or expressing disappointment can turn a nominal right to stop into a socially difficult choice.
Stop or escalate when an immediate safety concern emerges
Some distress episodes raise concerns beyond discomfort. A participant may disclose an immediate threat, develop acute physical symptoms, become unable to remain safe, or meet another emergency criterion specified in the study's procedures.
At that point, collecting the next answer is not the priority. Follow the approved safety, supervisory, referral, clinical, or emergency procedure appropriate to the situation.
If the participant becomes physically ill or injured, the study should follow its procedures for responding to illness or injury during research rather than treating the incident solely as interview distress.
Stop when privacy or the research environment becomes unsafe
Distress is sometimes inseparable from context. A participant discussing violence or another sensitive experience may become unsafe if another household member enters the room, if an abusive partner can overhear a remote interview, or if confidentiality can no longer reasonably be protected.
In such circumstances, the stopping decision may be driven by the environment rather than the participant's emotional presentation alone. Safety protocols for sensitive research should anticipate this possibility.
The protocol should define foreseeable stopping and escalation criteria
Researchers conducting studies likely to evoke distress should not leave every decision to interviewer instinct. Ethical review should be supported by a practical distress protocol describing foreseeable indicators, available participant choices, stopping criteria, escalation procedures, referral options, documentation, and staff responsibilities.
This follows the broader principle of beneficence in the Belmont Report: research should be designed to reduce risks to those necessary to achieve the research objective.
Preparation is particularly important when studies ask about trauma or highly sensitive experiences. Appropriate support arrangements should be established before sensitive data collection begins, not assembled for the first time after a participant becomes overwhelmed.
Unexpected distress can reveal that the study's original risk assessment was incomplete
OHRP recognizes psychological harms as adverse events and provides an example in which survey questions trigger intense sadness and depressed mood by evoking memories of childhood abuse. In that example, the reaction was unexpected, research-related, and indicated greater psychological risk than previously recognized, so it constituted an unanticipated problem.
An isolated distress episode may require individual follow-up. Repeated episodes can point toward a broader problem with question wording, sequencing, consent information, interviewer training, or the study's assumptions about risk.
The response should therefore include not only what happens to the participant in front of you, but also whether the incident should be documented and whether similar participants may face the same risk.
Watch Out
Do not use completion of the interview as the default outcome against which every alternative must be justified. Once significant distress appears, continuation itself should be justified by the participant's condition, meaningful voluntary choice, the approved protocol, and the absence of a safety reason to stop.
04 · A Practical Example
Two participants cry, but the appropriate decision is different
Hypothetical Example
The visible behavior looks similar at first
Two participants are separately interviewed about bereavement. Both begin crying while describing someone they lost.
Participant A
The interviewer pauses. The participant says the memory is painful but that they want a moment. After a break, they are composed, communicate clearly, understand their options, and independently ask to continue.
Assessment
Subject to the approved protocol, resuming may be reasonable because the emotional reaction has stabilized and the participant remains able to make a meaningful voluntary choice.
Participant B
The interviewer pauses, but the participant's distress intensifies. They struggle to communicate, cannot focus on the interviewer's questions, and repeatedly alternate between saying they should continue and saying they cannot think.
Assessment
Continuing merely because the participant has uttered “yes” would be difficult to justify. The interview should be stopped and the study's support or escalation procedure followed.
The difference is not the presence of tears. It is the participant's condition, trajectory of distress, ability to exercise meaningful choice, and the safety framework governing the research.
07 · A Quick Checklist
Before resuming an interview after participant distress
Before continuing, check:
Has data collection actually paused long enough to assess the participant rather than merely asking if they are okay between questions?
Is the distress stable or resolving rather than escalating?
Can the participant communicate and understand their options meaningfully?
Has the participant been reminded that they may skip questions, pause, reschedule, or stop as permitted by the study?
Is the decision to continue free from pressure to finish, please the researcher, or protect compensation?
Is the research environment still private and safe?
Does the approved protocol permit continuation under these circumstances?
Does the incident require documentation, referral, escalation, or safety reporting even if the participant later recovers?
