An IRB, or Institutional Review Board, is an independent committee that reviews research involving human participants to make sure it’s ethical and safe. Before a university researcher can survey students about their mental health, before a pharmaceutical company can test a new drug on volunteers, and before a hospital can run a clinical trial, an IRB must review and approve the plan. The board has the authority to approve a study, require changes, or reject it entirely.
What an IRB Actually Does
An IRB’s core job is to protect the rights and welfare of people who volunteer for research. It does this in two phases: reviewing the study before it begins and monitoring it periodically while it’s underway. The board examines every detail of how participants will be recruited, what they’ll be asked to do, what risks they face, and whether those risks are justified by the potential benefits of the research.
To approve a study, the IRB must confirm that it meets several specific criteria. Risks to participants must be minimized through sound research design. Whatever risks remain must be reasonable compared to the expected benefits, whether those benefits go to the individual participant or to society through new knowledge. The selection of participants must be fair, meaning researchers can’t target vulnerable groups simply because they’re easy to recruit. The study must include a proper informed consent process. And if participants are especially vulnerable to pressure or coercion (such as prisoners or children), additional safeguards are required.
Who Sits on the Board
An IRB must have at least five members with diverse backgrounds. Federal regulations require specific types of expertise on every board: at least one scientist, at least one person whose background is outside the sciences, and at least one member who has no affiliation with the institution running the research. That unaffiliated member might be a community volunteer, a former patient, or anyone whose only connection to the institution is their seat on the board. The mix is intentional. A panel of scientists alone might undervalue participant concerns, while a panel without scientific expertise couldn’t evaluate whether a study’s design is sound.
The regulations also require that the board not be composed entirely of one gender and that members with a conflict of interest in a particular study step out of the vote, though they can still answer questions the board has. When a study involves prisoners, a separate requirement kicks in: the board must include a prisoner representative.
The Three Ethical Principles Behind IRB Review
IRB review is built on three principles laid out in the Belmont Report, a foundational 1979 document that still shapes research ethics today.
Respect for persons means that individuals are treated as autonomous agents who can decide for themselves whether to participate. It also means that people with diminished autonomy, such as children or people with cognitive impairments, deserve extra protection. In practice, this principle drives the requirement for informed consent: participants must enter research voluntarily and with adequate information about what they’re agreeing to.
Beneficence requires researchers to minimize harm and maximize potential benefits. This isn’t just about avoiding obvious dangers. It obligates the research team and the institution to think carefully, in advance, about every way a study could cause physical, psychological, or social harm, and to design the study so those risks are as low as possible.
Justice addresses who bears the burdens of research and who reaps the benefits. Historically, certain groups (prisoners, racial minorities, people in institutional care) were disproportionately enrolled in risky studies while the resulting medical advances benefited wealthier populations. The justice principle requires that participant selection be driven by scientific relevance, not convenience or the vulnerability of a particular group.
Three Levels of Review
Not every study gets the same level of scrutiny. Federal regulations define three tiers based on how much risk the research poses to participants.
- Exempt review applies to research that falls into specific low-risk categories, such as anonymous surveys or analysis of existing data. Exempt studies don’t require continuing review or formal adverse event reporting, though they still need an initial determination that they qualify for the exemption.
- Expedited review is for studies that pose no more than minimal risk, meaning the likelihood and severity of harm are no greater than what someone would encounter in daily life or during a routine medical exam. These studies can be reviewed by the IRB chair or a designated experienced member rather than the full board, as long as the research fits within one of nine federally defined categories.
- Full board review is required for any study that doesn’t qualify for exemption or expedited review. This means the complete IRB committee meets, discusses the protocol, and votes. Studies involving more than minimal risk, such as clinical trials testing new drugs or procedures with meaningful side effects, go through this process.
What Informed Consent Requires
One of the IRB’s most important functions is ensuring that participants give truly informed consent. Federal regulations specify a checklist of elements that must appear in every consent document. Participants must be told that the study is research (not standard care), what its purpose is, how long their involvement will last, and exactly what procedures they’ll undergo. The document must describe foreseeable risks, any expected benefits, and alternative options available to them.
Participants must also learn how their confidentiality will be protected, who to contact if they have questions or experience a research-related injury, and whether compensation or medical treatment is available if something goes wrong. Critically, the consent form must state that participation is entirely voluntary, that refusing to participate carries no penalty, and that the person can withdraw at any time without losing any benefits they’re otherwise entitled to.
A 2018 revision to the federal regulations added a new requirement: consent forms must now begin with a concise summary of the key information a person would need to decide whether to participate, including the study’s purpose, its risks and benefits, and available alternatives. This was a direct response to the problem of consent documents that ran dozens of pages and buried essential information in legal language.
The Regulatory Framework
Two sets of federal regulations govern IRBs in the United States. Research funded or conducted by federal agencies falls under a regulation known as the Common Rule, formally codified as 45 CFR Part 46. Sixteen federal departments and agencies follow this rule. It includes a base set of protections (Subpart A) plus additional safeguards for specific vulnerable populations: pregnant women and fetuses (Subpart B), prisoners (Subpart C), and children (Subpart D).
Research regulated by the FDA, including clinical trials for new drugs, medical devices, and biologics, falls under a parallel set of rules at 21 CFR Part 56. Any clinical investigation that requires FDA submission must be reviewed and approved by an IRB before it can begin. The two regulatory frameworks overlap substantially but aren’t identical, so studies that are both federally funded and FDA-regulated must satisfy both sets of requirements.
Ongoing Monitoring and Adverse Events
IRB oversight doesn’t end at approval. Researchers must report problems that arise during a study, particularly “unanticipated problems” that meet three criteria: the event is unexpected given what the study protocol described, it’s possibly related to participation in the research, and it suggests that participants face greater risk than previously recognized. These problems can involve physical harm, but they can also be psychological, economic, or social in nature.
Federal guidance recommends that serious adverse events be reported to the IRB within one week of the researcher learning about them. Other unanticipated problems should be reported within two weeks. The IRB can then take action: requiring changes to the protocol, suspending enrollment, or halting the study entirely if participant safety demands it.