Mike Wenger, chief innovation officer at CRIO, explains how AI can responsibly support data quality and monitoring with proper oversight, and why advancing eSource and EHR systems remains critical to strengthening data integrity and remote trial operations.
In a recent video interview with Applied Clinical Trials at the 2026 SCOPE Summit, Mike Wenger, chief innovation officer at CRIO, explained how eSource adoption requires a nuanced understanding of site workflows, with different approaches needed for EHR-based health systems versus protocol-driven research sites. He highlighted where AI can responsibly add value once data pipelines are established, particularly in supporting quality review and monitoring, while stressing the importance of risk mitigation and human oversight. Wenger also underscored how continued advancement of eSource and EHR integration strengthens data integrity, enables structured capture, and expands remote monitoring capabilities across clinical trial operations.
Editor's note: This transcript is a lightly edited rendering of the original audio/video content. It may contain errors, informal language, or omissions as spoken in the original recording.
ACT: Where does the use of AI fit into data capture?
Wenger: Yeah, AI is a powerful tool. It also has risks associated with it. So whatever you’re trying to do when it comes to AI, the question is: what is the risk if something goes wrong, and how do you de-risk that with things like human-in-the-loop processes or some sort of review or QA process to make sure that things are working as expected?
I think what’s nice is that with all the solutions that were talked about during that panel, AI was not a requirement. We were mostly talking about data pipelines from one system—which is a site system—to the sponsor system, and that’s a very straightforward API-type technology integration. So there’s not a lot of risk when it comes to some of the solutions we were discussing.
But once those pipes are set up, there’s a lot more opportunity for things like AI to come in and start to automate some of the quality checks that a CRA might be manually doing when they go and look at a site—for example, identifying things that maybe the site did by mistake. There’s a huge opportunity for AI to come in and provide intelligence and insights that can have an impact there.
ACT: Looking forward, can you explain the importance of industry continuing to advance the uses of EHRs and eSource?
Wenger: So before these technologies existed, we had paper. And what’s wrong with paper? There are a lot of things wrong. For anyone who’s familiar with ALCOA, it’s really hard to adhere to ALCOA when anyone could write something down and timestamp it at a different time. So knowing that something is attributable to this user, in this system, at this time—that’s a huge win right there.
Another benefit is that if you just have paper, you could write anything down. But if you have a structured data element—like a multi-select or dropdown—you’re getting much better structured data going into the system. You can also have tools, such as what we have in CRIO, where you can add rules around that. So if a value is out of range, how do we either create an alert or require a comment to make sure there’s annotation around why something is the way it is when it shouldn’t be?
And then lastly, once you’re digital and you’re in eSource, you have the ability for remote access. This allows site customers, for instance, to do much better remote monitoring of the quality of their own source. It also unlocks the ability to give CRAs access for remote monitoring—things that historically, if it was on paper, required someone to physically be in the office to look at what was written down.
