Can You Publish Research Without IRB Approval? What Journals Actually Require

You finished your study. The data looks good. You write up the manuscript, pick a journal, start the submission process, and then you hit a form field asking for your IRB approval or ethics committee documentation. And you don't have one.

This comes up a decent amount, especially with teams that ran secondary data analyses, usability studies, or quality improvement projects where IRB review didn't seem necessary at the time. So: can you still publish?

The short answer

Some research genuinely doesn't require IRB approval under federal regulations. If your work doesn't meet the definition of human subjects research under the Common Rule, say you analyzed a fully de-identified public dataset with no way to link records back to individuals, then no IRB review was legally required. We covered the legal vs. practical distinction in a prior post.

But legal requirements and journal requirements are two different things.

What major journals actually require

The ICMJE (International Committee of Medical Journal Editors) recommendations are the baseline that most biomedical journals follow. The ICMJE says that research involving human participants should have been approved by an appropriate ethics committee, and that authors need to include a statement about that approval in their manuscript.

Individual journals layer on top of that. JAMA requires that all studies involving human participants include documentation of IRB or ethics committee approval, or a statement explaining why it wasn't needed. BMJ group journals ask for ethics approval details during submission and will follow up if they're missing. Nature group journals require authors to confirm that research was conducted in accordance with relevant ethics guidelines, with specific approval documented.

Surveys of journal policies bear this out. Research has shown that roughly 80% of journals require either IRB approval documentation or a formal ethics statement as part of the submission process. For clinical and biomedical journals, that number is higher. Even journals that don't explicitly mandate IRB approval will typically ask authors to address it.

What counts as acceptable documentation

Journals don't all require full IRB approval. What they want is evidence that someone independent looked at the ethical dimensions of your work. That usually means one of the following:

• A full IRB approval letter, which is the standard for clinical trials and higher-risk studies.
• An exempt determination letter from an IRB confirming your study fits one of the exempt categories under 45 CFR 46.104(d). We wrote a detailed guide to exempt determinations if you're not sure whether yours qualifies.
• A "not human subjects research" (NHSR) determination, which is a formal letter from an IRB stating that your activity doesn't meet the regulatory definition of human subjects research. We wrote a guide to NHSR determinations if you want to evaluate applicability to your study.
• Waiver of consent documentation for studies where the IRB approved a waiver of informed consent under 45 CFR 46.116(f).

Any of these will satisfy the journal's requirement. The common thread is that an independent body reviewed what you did.

What happens without documentation

Without any IRB documentation, the most common outcome is that the journal holds your manuscript and asks you to go get ethics documentation before they'll proceed. Some journals skip the ask entirely and desk-reject. And even if you make it to peer review, a reviewer who notices missing ethics documentation is going to flag it. That's not a conversation you want to be having in a revision letter.

Can you get IRB approval after the fact?

Generally, no. IRBs review research prospectively, before data collection begins. The whole point is to evaluate risks and protections before participants are exposed to them. Once the study is over, there's nothing left to protect against. Most IRBs won't issue retroactive approval for completed research, and it's not because they're being difficult. The regulatory framework just isn't built that way.

There are narrow exceptions. Some IRBs can issue a retrospective NHSR determination or exempt determination for completed work if the study clearly falls outside the scope of the regulations or fits an exempt category. In that case, the IRB is confirming what was already true, that the research didn't require prospective oversight, rather than blessing something after the fact. But this is case-by-case and not something you should count on.

Get the determination before you start if you can.

Don't forget clinical trial registration

IRB documentation isn't the only thing journals check. If your study meets the ICMJE's definition of a clinical trial (broadly: any study that prospectively assigns participants to an intervention to evaluate health outcomes), most major journals also require that it was registered on ClinicalTrials.gov or an equivalent registry before enrollment began.

ICMJE-member journals have enforced this since 2005, and more journals adopt the policy every year. Missing registration is another path to automatic rejection, and just like IRB approval, you can't do it retroactively in a way that satisfies most editors.

Why bother if you think you don't need it?

Say you're running a secondary analysis of public data, or an anonymous survey, or an internal QI project. You're pretty sure none of that requires IRB review. Maybe you're right. But if the results turn out to be interesting and you want to publish, you'll wish you had a determination letter.

Projects that start as internal analyses produce publishable findings more often than people expect. And the judgment calls involved, whether a dataset is "truly de-identified," whether a QI project is "really generalizable," are grayer than they look from the inside. An independent reviewer is better positioned to make those calls than the team that designed the study.

The process for an exempt determination or NHSR review is also lighter than most people assume. It doesn't require the same documentation as a full protocol review. If the study is genuinely low-risk, it can be quick and inexpensive.

We covered the regulatory framework for determining which rules apply in another post if you want to think through whether your study falls under HHS regulations, FDA regulations, or both.

Getting this handled

Tempus IRB issues exempt determinations, NHSR determinations, and full approvals, all at flat rates with fast turnaround. If you're not sure what your study needs, or if you're mid-manuscript and realizing you should have gotten a determination letter six months ago, email us. We deal with this situation regularly and can help you figure out your options.

This post is for informational purposes only and does not constitute legal advice.