What is an Institutional Review Board?

An Institutional Review Board, or IRB, is a committee with legal power to approve, change, or shut down a clinical trial. No trial can enroll a single person until an IRB says yes.

The board reads the protocol, which is the written plan describing exactly what will be done to participants and why. Federal rules require the IRB to check that the study is well designed, legal, and ethical, that it does not carry risks that could have been avoided, and that it has a real safety plan.

What the board is required to weigh

The rules are specific. Under the federal regulation at 45 CFR 46.111, an IRB may approve research only if risks to participants are minimized, and only if those risks are reasonable next to the expected benefits and the value of what will be learned.

The board also has to judge whether the selection of participants is equitable, meaning the burden of research is not dumped on one group. The regulation names people who need extra care here: children, prisoners, people with impaired decision-making capacity, and people who are economically or educationally disadvantaged.

Then it checks consent. The board reviews the consent form itself, confirms that consent will be sought properly, and confirms it will be documented. It looks at whether the study has a plan for monitoring the safety data as the trial runs, and whether your private information will be protected.

Who has to be in the room

Federal rules set a floor of five members with varying backgrounds. The mix is not optional.

At least one member's main expertise must be scientific. At least one member's main expertise must be nonscientific. And at least one member must have no connection to the institution running the trial, and must not be in the immediate family of anyone who does. That outside seat exists so the review is not entirely people with something to gain from the trial going forward.

Beyond that floor, boards often add doctors, nurses, social workers, chaplains, patient advocates, and community members. A member with a conflicting interest in a study is barred from taking part in that study's review, except to answer the board's questions.

The power to stop a trial

An IRB can approve a study, require changes before approving it, or refuse it. If it refuses, it has to say why in writing and give the researcher a chance to respond.

Approval does not last forever. For studies reviewed by the full board, continuing review must happen at intervals matched to the risk, and no less than once per year. If the risk picture changes, the board can amend the protocol or halt the trial outright.

Many phase 3 cancer trials add a second watchdog, a data and safety monitoring board. Unlike the IRB, a DSMB looks at the results as they accumulate, including how well each treatment is working. Either body can end a trial early, whether because of unexpected severe side effects or because one arm is clearly better and it would be wrong to keep enrolling people in the other.

What this means for you

If you are considering a trial, you can ask which IRB approved it and when it was last reviewed. That is a fair question and the study team should answer it.

An IRB approval is not a promise that the treatment will work, and it is not a promise you will not be harmed. It is a promise that a group including at least one outsider read the plan, judged the risks worth the potential knowledge, and keeps checking. Your own protection also rests on informed consent and on your right to leave the trial at any point without losing your regular care.

Sources

https://www.cancer.gov/research/participate/clinical-trials/safety

https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46/subpart-A/section-46.107

https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46/subpart-A/section-46.109

https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46/subpart-A/section-46.111