01 · The Question
When Can You Actually Call a Study “Minimal Risk”?
A researcher plans an online survey and describes it as minimal risk because there are no physical procedures. Another researcher labels an interview minimal risk because participants can stop whenever they want. A third assumes that anything more serious than minimal risk must be dangerous.
None of those shortcuts captures what the term actually means.
“Minimal risk” is not simply an informal way of saying that a study seems safe. In several influential research ethics and regulatory frameworks, it is a defined threshold based on the probability and magnitude of possible harm or discomfort compared with an external reference point. That comparison can affect how a study is reviewed and, in some frameworks, whether particular consent or review provisions can apply.
03 · What You Need to Know
What the Minimal-Risk Threshold Is Actually Measuring
Minimal Risk Is a Comparative Standard
The concept makes little sense without a comparator. You are not simply asking, “Is there some possibility of harm?” You are asking whether the probability and magnitude of anticipated harm or discomfort exceed a reference level established by the applicable framework.
Under the U.S. Common Rule, minimal risk means that the probability and magnitude of anticipated harm or discomfort are not greater, in themselves, than those ordinarily encountered in daily life or during routine physical or psychological examinations or tests. The Philippine National Ethical Guidelines for Research Involving Human Participants 2022 use essentially the same definition.
Canada's TCPS 2 uses related but not identical language. It defines minimal-risk research as research in which the probability and magnitude of possible harms are no greater than those participants encounter in aspects of their everyday lives that relate to the research.
That variation matters. Researchers should therefore use the definition applicable to their institution, jurisdiction, funder, or regulatory regime rather than treating one formulation as universal.
Both Probability and Magnitude Matter
Minimal risk is not determined only by asking whether an adverse event is likely. The seriousness of the possible harm also matters.
Probability
How likely is the possible harm or discomfort to occur?
Magnitude
How serious, intense, prolonged, or consequential could it be?
A study does not necessarily satisfy a minimal-risk threshold merely because a serious outcome is unlikely. Conversely, a relatively common experience may remain within the threshold if its consequences are sufficiently minor and comparable to the relevant baseline. How to assess very unlikely but potentially serious harms therefore requires attention to both dimensions.
Minimal Risk Does Not Mean Zero Risk
The word “minimal” is easily misread as “none.” That is not what these definitions say.
A study may involve foreseeable discomfort or possible harm and still fall within a minimal-risk category if the probability and magnitude remain within the applicable comparison standard. The threshold exists precisely because ordinary life and routine examinations are not completely free of risk.
This is also why ethical research does not require eliminating every conceivable risk. The question is whether risks have been appropriately minimized and whether what remains is ethically justified under the applicable framework.
Minimal Risk Is Not the Same as “I Think This Is Safe”
A researcher's personal impression is not the standard. Neither is familiarity with the method.
Surveys, interviews, observations, behavioral experiments, secondary data analyses, and routine clinical procedures can differ substantially in their consequences depending on what information is collected, who participates, how identifiable the data are, and what could happen if confidentiality fails.
For example, asking adults which type of music they prefer may present very different consequences from asking identifiable employees whether their supervisor has committed misconduct. Both use questionnaires. The data-collection method alone does not determine the level of risk.
Non-Invasive Does Not Automatically Mean Minimal Risk
A study can create substantial non-physical consequences without ever touching a participant. Research may involve psychological, social, legal, economic, or reputational risks in addition to physical risks.
Consider an interview about illegal activity, an identifiable survey about workplace harassment, or research collecting highly sensitive information about stigmatized behavior. The absence of needles, drugs, or medical devices tells you little about what disclosure or participation could do to the participant.
Risk assessment should therefore follow plausible pathways to harm rather than classify studies by method alone.
Minimal Risk Is Not the Same as Minimal Burden
A study can involve little risk of harm while demanding considerable time or effort. A participant might complete a lengthy but innocuous questionnaire battery with little meaningful possibility of harm, yet find the process exhausting and burdensome.
The reverse is also possible. A brief disclosure of highly sensitive information may take only minutes but create meaningful risk if the information becomes identifiable.
Keeping risk, burden, and inconvenience distinct prevents “minimal risk” from being misinterpreted as “minimal everything.”
Minimal Risk Can Affect the Review Pathway
The classification has practical consequences, although those consequences differ across systems.
Under the U.S. Common Rule, certain categories of research involving no more than minimal risk may be eligible for expedited review, subject to the applicable requirements. Minimal risk also appears in provisions governing some waivers or alterations of informed consent.
The Philippine National Ethical Guidelines likewise use the minimal-risk threshold in several provisions, including certain circumstances involving waiver of informed consent or waiver of signed informed consent.
Canada's TCPS 2 adopts a proportionate approach in which minimal-risk research is generally eligible for delegated REB review, while above-minimal-risk research receives full-board review.
Watch Out
Do not assume that calling your project minimal risk automatically qualifies it for expedited, delegated, exempt, or otherwise reduced review. Review categories and eligibility criteria depend on the applicable regulatory and institutional framework, and the authorized ethics body makes the relevant determination.
Minimal Risk Does Not Mean Exempt From Ethics
A particularly persistent misunderstanding is that minimal-risk research somehow sits outside research ethics. It does not.
For example, TCPS 2 explicitly states that minimal-risk research within its scope still requires REB review, although it is generally eligible for delegated review. The policy emphasizes that reduced scrutiny does not mean reduced adherence to ethical principles.
Likewise, under the U.S. Common Rule, “exempt” research is governed by specified regulatory categories. Minimal risk and exemption are not simply interchangeable labels.
Minimal Risk Does Not Mean Automatically Acceptable
Risk level is only one dimension of ethical research. A minimal-risk study may still have inadequate consent, poor confidentiality protections, unfair participant selection, insufficient scientific or social value, inappropriate recruitment, or a weak design.
Nor should the label replace a proper assessment of whether the research risk is ethically acceptable. A very small research risk may still be difficult to justify if the study is incapable of producing useful knowledge.
The Hard Part Is Often the Baseline
Definitions based on “daily life” sound straightforward until researchers ask whose daily life counts.
Should the risk faced by a person living with a serious illness be compared with that person's unusually risky daily experience? Could researchers recruit people living in dangerous circumstances and classify additional research risks as minimal because their ordinary lives are already dangerous?
Ethical frameworks have approached this problem differently. Canada's TCPS 2, for example, explicitly warns that research should not exacerbate the vulnerability of people who already live with relatively high levels of risk.
That question deserves separate treatment because determining the correct baseline for minimal risk can materially change the classification.
04 · A Practical Example
Why “It’s Just an Online Survey” Is Not a Minimal-Risk Assessment
Hypothetical Example
Two Surveys With Very Different Risk Profiles
A university researcher proposes two 10-minute online surveys. Survey A asks adults about preferences for different classroom layouts. Survey B asks identifiable employees whether they have witnessed corruption by their immediate supervisors.
Look beyond the method
Both studies use short online questionnaires. That similarity does not establish that they have equivalent risk.
Identify plausible harms
Survey A appears to offer few plausible pathways to consequential harm. Survey B could expose participants to workplace conflict, retaliation, reputational damage, or economic consequences if identities and responses became known.
Assess probability and magnitude
The researcher evaluates not only how likely disclosure or other harm might be but also what the consequences could be if it occurred.
Apply the governing standard
The resulting risks are compared with the minimal-risk definition required by the applicable ethics and regulatory framework rather than with the researcher's intuitive judgment that online surveys are generally safe.
The example does not establish automatically that either survey belongs in a particular review category. That determination depends on the complete protocol and applicable framework. It does show why method labels are poor substitutes for risk assessment.
07 · A Quick Checklist
Before Calling Your Study Minimal Risk, Check These
Before describing research as minimal risk, check:
What definition of minimal risk applies under my institution, jurisdiction, funder, or regulatory framework?
Have I identified all plausible research-attributable harms rather than considering physical harm alone?
Have I considered both the probability and magnitude of each meaningful harm or discomfort?
Have I assessed the actual study procedures and information collected rather than relying on labels such as survey, interview, or observation?
Have reasonable safeguards already been incorporated into the study design?
Can I explain what baseline or comparator I am using and why it is appropriate?
Have I kept participant burden and inconvenience distinct from the probability of harm?
Have I checked the institution's current review requirements instead of assuming minimal risk means exempt or expedited review?
