How Should You Explain Risks Without Exaggerating or Minimizing Them?

How do you describe risk without making the study sound safer or scarier than it is?

The risk section of an ethics application can create an awkward incentive. Researchers want to demonstrate that participants are well protected, so they may be tempted to emphasize that the study is "safe," "minimal risk," or has "no foreseeable risks." Others move in the opposite direction and list every imaginable bad outcome, however remote, because omission feels dangerous.

Neither approach necessarily helps an ethics committee or a prospective participant understand the study.

A useful risk description identifies the harms and burdens that can reasonably arise from the research, explains their nature and context, acknowledges uncertainty where it matters, and shows how the study has been designed to reduce or manage them. The goal is accurate characterization, not persuasion.

Describe foreseeable risks proportionately and connect them to safeguards

Risk is more informative when you separate what could happen from how concerning it is

Research ethics frameworks generally require more than acknowledging that "risk exists." They ask whether foreseeable risks and burdens have been identified, minimized, and justified in relation to potential benefits and the value of the knowledge expected from the research.

For medical research involving human participants, the World Medical Association's 2024 Declaration of Helsinki requires careful assessment of predictable risks and burdens in comparison with foreseeable benefits and states that measures to minimize risks and burdens must be implemented. Under the U.S. Common Rule, covered research must satisfy criteria including minimization of risks and a determination that risks are reasonable in relation to anticipated benefits, if any, and the importance of the knowledge reasonably expected to result.

These standards apply within their respective scopes. Your own committee may use different terminology, categories, or thresholds, so its current guidance should determine how formal risk classifications are reported.

Start with the procedure, then identify what could result from it

Risk descriptions are strongest when they remain connected to the actual research procedure.

If you collect blood, consider the foreseeable physical consequences of that procedure. If you interview people about traumatic experiences, consider emotional discomfort or distress. If you collect identifiable information about illegal behavior, consider confidentiality and possible legal or social consequences. If employees discuss their supervisors, consider what unauthorized disclosure could mean in that context.

This procedure-to-consequence approach helps prevent two common problems: overlooking nonphysical risks and inventing risks unrelated to the study.

Risk and harm are related but not identical

It helps to distinguish a possible harmful outcome from the risk associated with it.

For example, unauthorized disclosure of identifiable information is an event. Loss of employment, stigma, embarrassment, discrimination, or legal consequences could be harms resulting from that disclosure, depending on the information and context. The risk assessment considers how plausible and consequential those outcomes are under the proposed research procedures.

This is why statements such as "there is a confidentiality risk" are only a starting point. Reviewers need to understand what disclosure could involve and why it matters for these participants.

Consider severity and likelihood separately

A relatively likely inconvenience may be minor. A very unlikely disclosure may have severe consequences. Treating both merely as "risks" loses information.

Where reliable numerical estimates exist, they may be useful. Where they do not, avoid manufacturing precision. Qualitative descriptions can be more honest than assigning unsupported percentages or categories.

Also consider duration and reversibility when relevant. Temporary mild discomfort and a lasting reputational consequence may have similar likelihoods while carrying very different ethical significance.

Do not forget burdens that are not usually called harms

Research can impose burdens even when no conventional injury occurs. Time, inconvenience, fatigue, repeated appointments, travel, uncomfortable questions, dietary restrictions, disruption of routine, or lengthy assessments may matter to participants.

The Declaration of Helsinki explicitly discusses both risks and burdens. That distinction is useful well beyond medical research because participant experience cannot always be captured by asking only, "Could this cause harm?"

Risk is contextual

The same procedure can carry different consequences for different populations or settings.

Asking about political opinions, sexual behavior, workplace misconduct, immigration status, health conditions, or illegal activity may have very different implications depending on local law, social norms, employment relationships, community size, identifiability, and who could obtain the information.

A quotation that seems harmless in a large anonymous population may identify a participant in a small specialist community. An interview with employees may create different concerns when recruitment occurs through their supervisor.

Risk assessment therefore requires more than classifying the research technique. It requires understanding the context in which the technique is used.

Do not equate nonphysical research with no risk

Surveys, interviews, focus groups, observations, records research, and online studies may involve no physical intervention while still creating emotional, social, economic, legal, reputational, privacy, or confidentiality concerns.

WHO's protocol guidance explicitly notes that even administering a research questionnaire can have adverse effects on individuals.

That does not mean every questionnaire is dangerous. It means the researcher should assess what this questionnaire asks, who answers it, what information is recorded, and what could happen if participation or responses became known.

Do not exaggerate remote possibilities merely to be comprehensive

Ethical transparency does not require presenting every logically imaginable outcome as a meaningful research risk.

If an event is extremely remote, indirect, or unrelated to participation, giving it the same prominence as a reasonably foreseeable harm can distort the participant's understanding. It may also obscure the risks that actually deserve attention.

Under the U.S. Common Rule, for example, the IRB's risk-benefit assessment distinguishes risks and benefits resulting from the research from those of therapies participants would receive even without participating. The regulations also exclude possible long-range effects of applying the knowledge gained from the research from that particular risk-benefit assessment.

Your applicable framework may draw boundaries differently, but the general writing lesson remains useful: identify the risks of the research rather than compiling every adverse event imaginable in the surrounding world.

Explain safeguards as mechanisms, not reassuring adjectives

"The researchers will ensure confidentiality" sounds protective but does not explain how confidentiality is protected.

A stronger account identifies the mechanism: collecting only necessary identifiers, separating contact information from study data, limiting access, using approved storage arrangements, restricting what appears in reports, training research personnel, or using another study-specific protection.

The same applies to other risks. If participants may become distressed, explain what they can do and what the research team will do. If a procedure creates physical risk, describe the relevant monitoring or response arrangements. If authority relationships could create pressure, explain how recruitment and consent are structured to reduce it.

Be careful with the phrase "minimal risk"

"Minimal risk" can be a defined regulatory term rather than an everyday synonym for "not very risky." Its definition and implications depend on the jurisdiction and review framework.

Under the U.S. Common Rule, for example, minimal risk has a specific regulatory definition tied to the probability and magnitude of harm or discomfort ordinarily encountered in daily life or during routine physical or psychological examinations or tests.

Do not assume that definition applies everywhere, and do not use the label as a substitute for describing the actual risks. If your committee asks you to classify the study, use its definitions and criteria.

Risk statements should agree across the submission

The application might describe emotional discomfort as a foreseeable risk while the participant information sheet says "there are no risks associated with participation." That contradiction matters.

Participants should receive an understandable account of reasonably foreseeable risks relevant to their decision, subject to the consent requirements that apply to the study. The protocol, application, consent materials, recruitment claims, and safety procedures should therefore tell a coherent story.

Replace a vague risk statement with a proportionate one

This description does not make the study sound harmless or hazardous. It tells the reviewer what the concerns are, why they arise, and what the research team does about them.

Common ways researchers misrepresent risk

Build each risk statement from procedure to residual risk

A practical way to draft the risk section is to examine each important study procedure individually rather than beginning with a conclusion such as "low risk."

Then compare the risk section with the procedures, consent materials, data plan, and supporting documents. If a risk appears in one place but disappears elsewhere, determine which description accurately represents the study.

Finally, keep risks separate from benefits. A risk does not become smaller because the project sounds valuable, and a potential benefit should not be inflated to compensate rhetorically for an uncomfortable risk section. The appropriate way to describe potential research benefits requires the same restraint.

Check whether your risk description is proportionate

Questions about describing risks in an ethics application

Describe risk to inform judgment, not to sell the study

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.