03 · What You Need to Know
Research Risk Extends Far Beyond Physical Harm
Human-Participant Research Is Broader Than Clinical Research
Research ethics is sometimes mentally filed under medicine: clinical trials, experimental drugs, surgical procedures, blood samples, and hospital research. Those are important applications, but they do not define the outer boundary of human-participant research.
WHO describes research involving human participants broadly enough to include social-science, biomedical, behavioral, and epidemiological activities involving systematic collection or analysis of data for generating new knowledge when humans are involved through manipulation, intervention, observation, other interaction, or identifiable records or biological materials.
A sociologist interviewing workers, an education researcher surveying students, and a psychologist observing behavior may therefore encounter research-ethics requirements even though none is conducting a medical procedure.
Physical Risk Is Only One Type of Risk
WHO guidance expressly states that research risks are not limited to physical harms. It identifies psychological, social, legal, and economic ramifications as potential research risks.
| Type of Risk |
Possible Non-Physical Example |
| Psychological |
Distress from recalling trauma, abuse, bereavement, discrimination, or another difficult experience |
| Social |
Stigma, damaged relationships, discrimination, or exposure of membership in a sensitive group |
| Informational |
Unauthorized disclosure of personal, educational, employment, financial, or health information |
| Legal |
Disclosure of information about unlawful or potentially sanctionable behavior |
| Economic |
Loss of employment, income, insurance opportunities, or other financial consequences following disclosure |
| Educational |
Consequences for academic standing or advancement if sensitive student information is disclosed |
These risks are not present in every non-interventional study. The point is that the absence of bodily harm does not establish the absence of meaningful harm.
A Questionnaire Can Create Psychological Risk
A questionnaire can look innocuous because participants sit safely in front of a screen and click answers. But content matters.
WHO guidance gives the example of questionnaires concerning traumatic or especially stressful events as potentially creating psychological risk.
A study asking survivors to recount violence, employees to describe harassment, patients to revisit a difficult diagnosis, or students to disclose severe anxiety can generate distress without any physical intervention.
Researchers may need to consider whether questions are necessary, how participants are warned about sensitive content, whether questions can be skipped, whether participation can be stopped, and whether appropriate support or referral information is warranted.
Disclosure Can Be More Harmful Than the Research Procedure
Sometimes answering a question creates little difficulty. The real risk lies in what happens if the answer becomes known.
Suppose employees are interviewed about corruption within their organization. The interview itself may involve no physical risk and little emotional distress. A confidentiality breach, however, could affect employment, professional relationships, reputation, or legal exposure.
WHO guidance specifically recognizes discrimination and stigmatization resulting from inadvertent disclosure of confidential research information as social, legal, and economic risks.
Identifiable Data Can Create Risk Without Any Participant Contact
You do not even need to speak with participants for research to create informational risk.
Analyzing identifiable medical records, academic records, employment files, or other private information can expose individuals to harm if information is accessed improperly, inadequately protected, linked with other datasets, or disclosed.
This is why research using existing data may still require an ethics determination even when researchers never meet the people represented in the dataset.
Observation Can Raise Privacy Questions
Observation is physically non-invasive, but the setting matters. Watching behavior openly occurring in a public square is ethically different from recording behavior in a setting where people reasonably expect privacy.
Even observation of public behavior can require careful consideration of identifiability and potential harm. U.S. HHS regulations, for example, include observation of public behavior within an exemption category only when specified conditions are met.
Recording faces, voices, license plates, usernames, or distinctive behavior can also transform apparently anonymous observation into identifiable research data.
Behavioral Research Can Involve an Intervention Without Touching Anyone
Another source of confusion is the word intervention. Researchers sometimes interpret it as meaning a physical or medical procedure. Regulatory definitions can be broader.
Under the U.S. Common Rule, intervention includes both physical procedures by which information or biospecimens are gathered and manipulations of a participant or their environment performed for research purposes.
The revised Common Rule even contains a specific exemption category for certain benign behavioral interventions involving adults, provided its conditions are satisfied. Examples can include particular brief behavioral tasks or manipulations that are harmless, painless, not physically invasive, and not likely to have significant adverse lasting effects.
So “nobody was physically touched” does not necessarily mean “there was no research intervention.”
Deception and Withholding Information Can Raise Ethical Questions
Some behavioral and social research cannot answer its question if participants know the full hypothesis or purpose in advance. Researchers may therefore use incomplete disclosure or, in some designs, deception.
That can affect informed consent and participant autonomy even when the activity presents no physical danger. Ethics review may consider whether deception is scientifically necessary, whether risks remain acceptable, whether participants should be debriefed, and whether consent requirements can appropriately be altered under the governing framework.
Power Relationships Can Create Risk Without Any Intervention
A study can be physically harmless yet difficult to refuse.
Students recruited by their professor, employees recruited by their manager, patients approached by their treating clinician, or junior researchers recruited by a supervisor may worry that declining participation could have consequences.
The researcher should therefore consider voluntariness, recruitment procedures, incentives, and dependent relationships separately from physical risk.
An anonymous survey, for example, can protect responses while still being recruited in a coercive manner.
Research Can Affect Groups as Well as Individuals
Not every harm occurs because a named participant is exposed. Findings can stigmatize communities or population groups.
WHO guidance notes that ethnic or population groups may experience discrimination or stigmatization when research identifies them as having elevated rates of a disease or other characteristic. CIOMS likewise emphasizes that risk assessment should consider risks to groups and populations as well as individual participants.
This matters particularly when researchers study small communities, marginalized populations, stigmatized behavior, or characteristics that can readily be associated with a recognizable group.
Minimal Risk Does Not Mean No Ethics Review
Many non-interventional studies present no more than minimal risk. That can make them eligible for exemption or a streamlined review pathway under the applicable framework. It does not create a universal “non-interventional research” exemption.
The U.S. Common Rule, for example, establishes specific exemptions for certain surveys, interviews, observations, educational research, and benign behavioral interventions. Those exemptions contain conditions concerning methods, identifiability, disclosure risk, participant population, and sometimes limited IRB review.
Likewise, the Philippine national framework permits specified low-risk activities to be considered for exemption and provides expedited review for eligible research. The applicable criteria, rather than the absence of a needle or medication, determine the pathway.
Non-Interventional Does Not Mean Exempt
This distinction deserves to be explicit.
Non-interventional or non-invasive
Describes something about what the researcher does to or with participants.
Exempt from review
Describes a formal status under an applicable ethics framework when defined criteria are satisfied.
The first is a feature of the study. The second is a regulatory or institutional classification. One does not automatically produce the other.
Ethics Review Is About Rights and Welfare, Not Just Injury Prevention
Reducing research ethics to “make sure nobody gets physically hurt” misses much of what ethics committees actually examine.
WHO describes research ethics as protecting participants' dignity, rights, and welfare and identifies beneficence, justice, and autonomy as central to ethical review.
Accordingly, review may consider informed participation, fair selection, privacy, confidentiality, vulnerability, voluntariness, scientific and social value, and appropriate treatment of participants alongside physical safety.
Watch Out
Do not describe a study as “no risk” simply because it contains no physical procedure. Identify the realistic risks created by the questions, information, relationships, setting, data handling, and publication of findings.
