How AI Can Ease Recruitment Burdens in Resource-Strained Trials

In a recent video interview with Applied Clinical Trials, Kyle McAllister, co-founder, CEO, Trially, discussed the impact of funding cuts on underrepresented populations in medical research, emphasizing the importance of diversity-focused research. He highlighted the challenges created by budget cuts, particularly in patient recruitment and retention, due to reduced support staff. McAllister noted that artificial intelligence (AI) and new technologies like telemedicine are crucial for addressing these challenges by automating tasks and improving efficiency. He provided examples of AI's success in reducing chart review time and increasing patient enrollment in studies.

ACT: How can AI help address some of the recruitment challenges caused by reduced funding?

McAllister: It's a difficult time in the industry for a lot of folks, but in a lot of ways, it's kind of AI's time to shine. We're seeing decreased human resources, but that's where AI is really supposed to pick up the lift, so I think AI can really augment a lot of basic tasks. It can leave, and ultimately, hopefully leave research teams to practice more at the top of their license. “Let's take the let's take the dirty work off your plate and let you focus on the things you came here to do, which is see patients and care for patients.” This is going to be very slanted based on a lot of the things we work on at Trially, but some of the big, big, exciting things that I see folks getting excited about for AI are, one, AI-powered patient identification. AI can sift through large amounts of data from EHRs, from CTMS systems, even from PDF documents, taking what is, 30 plus minutes, hours of time, reading through patients’ charts down to moments. That’s huge. We've had sites where we have teams that do 90% less chart review than they did before the system gets implemented. Quick example, I spent my entire career before this, before basically OpenAI and Sam Altman happened, telling physicians that if you write it in a note, it's gone forever. In the time I spent at Epic and Cerner, that was the message, it's got to be documented discretely. That's just not true anymore. We can sift through a lot of that information and do it really quickly, which is powerful, that's industry shifting capability, so that's one.

The other would be, if we're already in there sifting through all that data, you are resource constrained as a site. Picking the right study, picking the studies that you're going to crush it on, is really important, so the ability to actually do real-time feasibility on that data set, and understand, do I have a population that supports our research? Us taking on this study is really big, so real-time feasibility, real-time business development and site select, or trial selection without guesswork, are really big things.

Then the last I'd call out is, you hear a lot about automated prescreening and engagement of patients with AI voice in call screen, and actually engage these patients directly with very human like capability, and that's a huge lift for sites. In my time actually working at a site network, it was a massive pain for all of our research coordinators to have to call, say we're doing a vaccine study and we have 10,000 potential candidates. How do you call 10,000 of those people to try to screen them into a trial. AI can really take the front-end lift of that, take the initial outreach, and then hand it off to a human to really close the deal and bring them in for the study. There's probably a thousand more use cases that I didn't hit, but those are some of those that are closest to our hearts.