03 · What You Need to Know
How a Research Risk-Benefit Assessment Works
Start With a Study That Can Actually Answer Its Question
Risk-benefit assessment begins earlier than the ethics form. It begins with research design.
The Belmont Report describes risk-benefit assessment as an opportunity to examine whether proposed research is properly designed. Under the U.S. Common Rule, risks must be minimized through procedures consistent with sound research design that do not unnecessarily expose participants to risk. CIOMS likewise connects the ethical acceptability of research risk with the study's social and scientific value.
This matters because participant exposure cannot be justified by knowledge the study is unlikely to produce. A poorly designed study may have modest procedures and still present an unfavorable ethical proposition if participants bear those procedures for little credible scientific value.
Identify the Risks Created by the Research
Map the protocol procedure by procedure rather than writing a generic paragraph about risk. Consider interventions, interviews, questionnaires, observations, randomization, withholding or delaying interventions where applicable, specimen collection, data linkage, identifiable information, recruitment procedures, and follow-up.
For each one, ask what could plausibly happen because the person participates. Relevant types of research risk may include physical, psychological, social, legal, economic, and reputational harms.
Also distinguish research-attributable risks from risks participants face independently of the study. Otherwise, the assessment may either blame the study for background hazards it did not create or overlook the additional risk that participation actually adds.
Do Not Collapse Probability and Magnitude Into One Vague Label
The Belmont Report specifically distinguishes the probability of harm from its magnitude. Calling a risk “small,” “moderate,” or “high” without explaining these dimensions can conceal rather than clarify the assessment.
Probability of harm
How likely is the adverse outcome to occur?
Magnitude of harm
How serious, prolonged, reversible, or consequential could the outcome be?
A relatively likely episode of temporary discomfort raises a different ethical problem from a remote possibility of permanent injury. Both deserve analysis. This is particularly important when evaluating very unlikely but very serious harms.
Assess Burdens as Well as Risks
Not everything participants endure is a risk of future harm. Long visits, repeated questionnaires, travel, dietary restrictions, fatigue, uncomfortable procedures, or intensive monitoring may impose burdens even when the probability of serious harm is low.
The 2024 Declaration of Helsinki explicitly requires assessment and minimization of predictable risks and burdens.
Keeping risk, burden, and inconvenience conceptually distinct can therefore produce a clearer assessment without making burdens ethically invisible.
Minimize Risk Before Trying to Justify It
This ordering is fundamental. Do not begin by asking whether a potentially avoidable risk is worth taking. First ask whether the protocol needs to create that risk at all.
The Belmont Report states that risks should be reduced to those necessary to achieve the research objective and that alternative procedures should be considered. The U.S. Common Rule similarly requires risks to be minimized through sound research design and, where appropriate, by using procedures already being performed for diagnostic or treatment purposes. The Declaration of Helsinki requires measures to minimize risks and burdens.
Remove unnecessary procedures
Do not expose participants to a procedure simply because the resulting data might be interesting.
Use safer alternatives where scientifically adequate
Change the method, frequency, setting, data collected, or procedure when the research question can still be answered adequately.
Add safeguards
Use appropriate monitoring, confidentiality protections, eligibility criteria, stopping rules, trained personnel, referral arrangements, or other protections suited to the risk.
Assess the residual risk
Only after reasonable minimization should the remaining risk be weighed ethically.
Identify Benefits Without Inflating Them
Potential benefits need the same intellectual discipline as risks. Ask what positive outcomes are reasonably expected, who may receive them, and how credible those expectations are.
The Belmont Report uses “benefit” for something of positive value related to health or welfare and makes a subtle distinction: risk expresses a probability of harm, while benefit itself is not a probability term. A rigorous assessment therefore considers the probability and magnitude of possible harms alongside anticipated benefits and the likelihood that those benefits will occur.
Direct Participant Benefit and Knowledge Value Are Not the Same Thing
A participant may have a prospect of direct benefit from a research intervention. Alternatively, the participant may receive no personal benefit while the study produces knowledge that could benefit future patients, communities, policy, practice, or science.
Those are ethically different forms of value and should not be merged into a vague sentence claiming that “participants will benefit from contributing to knowledge.”
Whether research needs to benefit participants directly depends on the study and applicable ethical framework. Research without direct participant benefit can sometimes be ethical, but the absence of such benefit changes what can legitimately justify participants' exposure.
Payment Is Not a Research Benefit in the Risk-Benefit Calculation
Compensation, reimbursement, and incentives should not be used to make a risky protocol appear more favorable. Under the U.S. Common Rule, when an IRB evaluates whether risks are reasonable in relation to anticipated benefits, it does not consider possible long-range effects of applying knowledge as though they were direct research benefits to participants. More broadly, payment is ethically evaluated separately from the substantive benefits produced by the research.
This is why compensation cannot simply make an otherwise unacceptable research risk ethical.
Examine Who Bears the Risk and Who Receives the Benefit
A favorable assessment cannot be reduced to total benefits minus total harms. Distribution matters.
The 2024 Declaration of Helsinki asks researchers to consider how research benefits, risks, and burdens are distributed. The Belmont Report similarly notes that risks and benefits may affect participants, their families, and society, while giving special weight to risks and benefits affecting the immediate research participant.
A protocol in which one group bears substantial research risk while another group receives nearly all the expected benefit raises questions that cannot be solved merely by claiming that aggregate social benefit is large.
Do Not Turn the Assessment Into Fake Arithmetic
The phrase “risk-benefit ratio” is common, but it can be misleading if interpreted literally. There is rarely a common unit in which psychological distress, a one-percent probability of infection, possible clinical improvement, and future scientific knowledge can all be converted and divided.
The Belmont Report explicitly describes balancing risks and benefits as metaphorical and notes that quantitative techniques will only rarely be available for protocol assessment. What matters is systematic, non-arbitrary analysis based on the best information available.
Watch Out
A numerical risk matrix can help organize information, but a score such as “risk = 6, benefit = 8, therefore acceptable” does not by itself establish ethical justification. Numbers can structure judgment without replacing it.
Evaluate Procedures Individually as Well as the Study Overall
A study-level judgment can conceal unnecessary procedures. CIOMS recommends attention to individual research interventions and procedures because a global assessment may miss an intervention that creates risk without producing direct benefit or sufficiently important information.
Suppose a trial is valuable overall but includes an additional invasive procedure used only to collect exploratory data of little scientific importance. The overall value of the trial does not automatically justify that procedure. Each meaningful exposure should have a defensible role.
Ask Whether the Remaining Risks Are Reasonable
After risks have been minimized and benefits characterized realistically, the central ethical judgment remains.
Under the U.S. Common Rule, an IRB must determine that risks are reasonable in relation to anticipated benefits to participants, if any, and the importance of the knowledge reasonably expected to result. The Declaration of Helsinki states that medical research may proceed only when the importance of its objective outweighs the risks and burdens to participants.
This is not permission to justify any level of individual risk by invoking sufficiently important science. The acceptable relationship among participant risk, direct benefit, and social or scientific value depends on the nature of the research and the ethical rules governing it. The limits of using societal benefits to justify individual participant risks therefore require separate scrutiny.
The Assessment Continues After the Study Begins
A favorable assessment at approval is based on information available at that time. New adverse events, unexpected distress, emerging efficacy data, new external evidence, confidentiality incidents, or changes in available treatment can alter the equation.
The Declaration of Helsinki requires risks and burdens to be continuously monitored, assessed, and documented. When risks and burdens outweigh potential benefits, researchers must assess whether the research should continue, be modified, or be stopped.
A risk-benefit assessment is therefore a continuing ethical judgment, not a paragraph written once for the ethics application.
