Exempt Research and Why You Do Not Decide It Yourself

Part of Human Subjects Protection for Clinical Research Staff: What Counts as Human Subjects Research.

About 5 minutes. Free, in your language, no account needed.

Why this matters

Many people hear "exempt" as "not regulated" or "no one needs to look." That is not what it means. This lesson explains what exempt means under the Common Rule, why the list is narrower than it sounds, and why the label is not yours to attach.

Exemption is a US Common Rule concept, taught here as the worked example. Under ICH E6(R3), the international baseline, clinical trials are subject to independent IRB/IEC review. Outside the United States, follow your national law and ethics committee on what needs review.

What exempt means

Under 45 CFR 46.104(a), exempt research is exempt from the policy's requirements, "except that such activities must comply with the requirements of this section." So exempt research is still human subjects research. It is not a free pass: it is a defined set of categories that must meet specific conditions.

The categories at a glance

The 2018 Common Rule lists eight exemption categories (46.104(d)(1) to (8)). Studies still under the pre-2018 Requirements were determined under the older categories, which differ. In summary form:

You do not need to memorize these. Each category has conditions, and a project does not become exempt because it sounds low-risk.

Special populations narrow the list

The exemptions do not apply the same way to everyone (46.104(b)):

So a survey that fits an exemption for adults may not for children. Always say who the participants will be when you ask.

FDA is different

FDA's exemptions from IRB review (21 CFR 56.104) are narrow:

FDA has no general exempt categories like the Common Rule's. Its exemptions are limited to the three above; the taste and food quality exemption overlaps Common Rule category (d)(6).

Screening and eligibility without consent

An IRB may approve research in which an investigator obtains information or biospecimens only to screen, recruit or determine eligibility, without consent. That can be through communication with the prospective subject, or by accessing records or stored biospecimens (45 CFR 46.116(g)). The IRB makes that call. Staff do not get to skip consent because an activity "is only screening."

Why you do not decide it yourself

Deciding that a project is exempt, or that it is not research, is a determination. A good habit, and CRLN practice advice rather than a regulatory statement: do not label your own study exempt or not research. Ask the IRB or the institutional office that handles the determination, and follow your institution's procedure.

In practice

A graduate student asks a coordinator for help with an anonymous survey of clinic patients about wait times. The student says, "It is exempt, so no paperwork." The coordinator does not argue about whether that is right. She says, "I cannot tell you it is exempt, and neither can you. Let's send the description to the IRB office and ask them." The survey waits for the office's answer, and everyone knows where it stands.

What this looks like when it goes wrong

A study team calls its project exempt because it sounds harmless: a questionnaire, some interviews, no needles. Midway, they add recordings of sessions with minors and a request for identifiable details. No one asks again. The facts that supported the label changed, but the label stayed, and the team has data collected under a label no one confirmed.

Key takeaways