Kim Serpico wants you to love your IRB.

Institutional Review Boards (IRBs) might be seen as the bane of existence for a researcher trying to get a study up and running. Serpico is trying to change that perception, one study at a time.

A research ethicist who sits at the helm of an operation assigned with no less a task than keeping human research subjects safe, Serpico is associate director of IRB operations for the Harvard Longwood campus (HLC). Her team reviews every research study that involves human participants in any way, large or small, from Harvard Medical School (HMS), Harvard T.H. Chan School of Public Health, and Harvard School of Dental Medicine.

Serpico also co-chairs Harvard Catalyst’s subcommittee on Emerging Technologies, Ethics and Research (E-Tech), where thorny issues such as how to use AI to research advantage–safely–are tackled. She thrives on demystifying regulatory approval for investigators and helping researchers form collaborative partnerships with their IRBs, to everyone’s advantage.

What do you do and why?

I’m a research ethicist. My team is responsible for the oversight of all human research undertaken or conducted by HLC affiliates, students, faculty, staff, administrators, postdocs, fellows, and everybody in between. That could be as complicated as clinical trials or a relatively simple secondary data analysis study. We see it all.

We make sure research meets regulatory criteria as well as ethical standards for the treatment and protection of human subjects. We’re also making sure that there’s scientific and social need for the work.

As to why, all of us in the research world have a responsibility to make sure people trust science, trust researchers, and trust our institution with their data. Participants play a very important role in contributing to scientific and social development and innovation across the world. Without their trust, none of this happens. It’s our job to make ourselves trustworthy.

“Participants play a very important role in contributing to scientific and social development and innovation across the world. Without their trust, none of this happens.”

Researchers seem to universally dread the IRB. Is it really that bad?

It’s a common misperception that if something bad happens in your research, it’s your fault and you should be afraid of the IRB. That’s not true. The IRB endeavors to get the problems resolved and researchers back out in the field.

We try to reshape the idea of IRB as these shadowy figures in the background making decisions with no human context. We bust through those perceptions one project at a time, through our relationships with investigators.

I see our role as a partnership. I tell researchers: I’m on your team. We all have the same goal. We want you in the field spending your funding as soon as possible. But let’s make sure that we have everything above board and set to go. We really try to make it painless and understandable for researchers.

But yeah, it’s a process. It is labor-intensive and onerous, and I understand why researchers hold those perceptions. It can be overwhelming to think you’ve put together a really beautiful study and the IRB gives you tons of feedback on things you have to change. That can be a little bit of a gut punch. We get that. And we say: How can we help you with these things? That’s the education component.

What do early-career investigators need to know about IRB?

Our IRB review is really a curated, individualized service. Customers receive an education. They have a dedicated, focused review specialist. We treat every case individually and every PI according to their level of knowledge. Our in-depth knowledge of the research community we serve allows us to meet investigators where they are, whether they’re a first-timer or a lifelong researcher.

We’re going to receive with open arms anybody who comes to us with questions. We love when investigators are proactive, especially early-career and new-to-research folks, because we really have a unique opportunity to help shape their research practices. That’s a big responsibility for us.

It works best as a collaboration where we all know our mutual goal, rather than researcher and IRB being on different planets. The researchers help us understand the science and we help them understand ethics and regulations.

We always ask investigators for feedback on what we could have done better. Many have helped us improve our practices, our online IRB application system, and the toolkit of templates, forms, and resources on our website. We’re always looking for that constant improvement that we can only get through feedback from researchers.

You’ve described your department as a kind of “one-stop shop” for regulatory and ethical aspects of human research. How does that help advance science?

“The researchers help us understand the science and we help them understand ethics and regulations.”

The one-stop shop model helps us tailor our services to the needs of individual researchers. Each member of our team is assigned to particular departments from across the HLC, so we become really well-versed in the kinds of research that come from those departments. We know who the researchers are, the typical study populations they’re looking at, where they do their research, whether it’s domestic or international, and so on.

That continuity helps ensure that our department-assigned review specialists really understand researchers’ work inside and out. Investigators don’t have to re-educate reviewers every time they submit a modification because the IRB review specialist already knows it, probably as well as the researchers do. We are able to hit the ground running.

Switching to the Harvard Catalyst subcommittee on emerging technologies that you co-chair, what are the big issues this committee is facing?

Our Harvard Catalyst committee is really special because it comprises a blend of IRB professionals and IT professionals working together to figure out the ethics of emerging technologies. How do we know that all of this new tech in the world is going to be safe? How will we keep our participants’ data from being used by third parties? Is it okay to include participant data in AI or large language models? Do participants have to consent to put their data in those big models?

Those are really big, tricky questions right now that we are all trying to figure out within our little buckets of expertise. We’re frontline. We’re on the ground. We’re seeing firsthand what researchers are attempting to do or asking us if it’s okay to do. It’s a really cool time because researchers are coming up with the most ingenious ways to study these technologies in the most ethical ways.

We’re trying to tackle the breadth of academic and scholarly uses of AI that we’re seeing across the HLC schools. How can we lend our expertise to those researchers, whether it’s regulatory, ethical or technological? We’re not asking researchers to be experts in AI technology right now, because they aren’t.

Rather, we’re trying to empower people to ask how AI can benefit their research goals. It could be as simple as streamlining operational processes that have nothing to do with humans, such as rewriting all of your SOPs to be in a different tone. It doesn’t have to be this big scary complicated thing.