How an AI-Human Partnership Drove 50% More Qualified Clinical Trial Participants

For every medical breakthrough that hits the market, several more go into clinical trials to test similar or adjacent treatments.

Take the three FDA-approved Alzheimer’s Disease drugs that were released over the last two years. After more than a couple of decades of Alzheimer’s studies suffering setbacks—including investigational drugs being put on hold—there is now a development pipeline of nearly 200 Alzheimer’s trials. Considering that each of these trials requires north of 100 participants and so many studies are taking place simultaneously, there’s suddenly a lot of demand for similar patient populations. And that’s just Alzheimer’s studies.

Across health conditions and treatments, there are currently more than 450,000 clinical trials in progress around the world. Around eighty percent of these are at risk of delays because trial sponsors simply can’t find the volunteers they need to proceed. Once sponsors do get participants in the door, getting them to stay for the duration of the trial is another feat entirely.

Participant recruitment and retention are the biggest sources of manual labor, delays and costs for clinical trials. This is what makes them perfect candidates for AI’s immense potential to facilitate—and even reimagine—otherwise cumbersome processes.

As clinical trial sponsors, recruiters and managers determine AI tools’ part in improving these workflows to boost participation and retention, it’s important to not completely shift away from necessary, human-centered interactions. Doing so can have regrettable consequences on the quality of the patient experience.

Here’s a look at the hybrid approach we developed to achieve up to a 50% increase in patient randomizations into clinical studies.

Also Read: Humanoid Robots And Their Potential Impact On the Future of Work

AI identifies and targets candidates to find new potential trial participants that sites don’t already know about.

Given that multiple studies are often tapping into the same patient populations during similar timeframes, many trial sites are often limited in their ability to effectively enroll all trials simultaneously.

This is due in part to the fact that while sites’ databases might be comprehensive, they don’t account for every existing potential participant. In many cases, patients simply haven’t been identified yet.

For example, in the U.S. alone, there are over 6.5 million people living with Alzheimer’s. Many of these people may qualify for clinical trials that could lead to life-changing treatments for themselves and others, but either don’t know what trials are being conducted or how to get involved.  And sites likely don’t know about them either.

In response to these limitations, sites have begun increasingly relying on outside vendors to assist with recruitment through direct patient advertising and advocacy methods. As a result, social media and digital advertising have become essential tools for accessing these as-of-yet unaware patient populations by reaching them where they’re already looking.

Now, with the addition of AI and machine learning, campaigns can factor in thousands of historical data points from recruitment firms or pharmaceutical companies to identify potential matches for trials–all based on highly specific trial criteria. And AI can do this at an exponentially higher scale and speed.

With AI working hand in hand with this historical data–and learning even more as it goes–trial sponsors can go beyond simply targeting ads at, say, populations within 10 miles of the trial site who meet the trial age criteria. While this is no doubt a good and logical place to start, the potential created by using advanced technology is infinitely greater.

AI assists in pre-screenings to expedite the enrollment process.

Once potential participants have been identified pre-screening is necessary to determine whether or not they meet trial requirements. Screening participants before they go to the trial site ensures sites can focus their time and resources on only the most qualified patients.

Not only can we leverage AI to pre-screen exponentially more candidates faster than any method before it, it also creates more accurate patient recommendations so the right participants are getting to the right trials. This increases the chances that every trial gets what it needs and makes more efficient use of existing and new patient pools.

Through this approach, we have been able to increase the number of highly qualified participants we match to trials every month by 15%. This is a huge advantage for sponsors that need more patients than ever before.

Humans then step in to transform candidates into participants.

Once candidates are pre-screened, the technology passes them on to skilled nurses, who personally walk candidates through the rest of the enrollment process. These nurses conduct thorough health assessments, making sure that potential clinical trial participants match protocol criteria. They are also responsible for understanding participants’ individual lifestyle considerations and how those intersect with trial participation requirements.

By pairing the efficiency of technology with human-centric care, we have seen an increase in patient randomizations into trials by up to 50%. This is even higher than when pre-screening is conducted solely by online technology.

Also Read: AI Strategies for Accelerating Clinical Trial Timelines and ROI

Human care plays a critical role in the patient experience, as patients want to work with providers they trust.

Healthcare is deeply personal and intimate, which is one of AI’s largest limitations.

For those struggling with diseases that they may not fully understand, or who are confused and anxious about the full implications of their diagnoses and treatment plans, receiving guidance from well-trained human nurses and clinicians can make all the difference. That human touch fills in gaps that AI may never be able to identify or understand.  Picking up on emotional nuances and meeting them with human compassion often falls well outside of AI’s domain.

While technology is constantly evolving and some predict it will adopt more human-like attributes and autonomy, keeping clinical trial participants engaged and comfortable for the duration of trials will continue to be the domain of trained humans—with a gentle touch, a calm voice, and a steady demeanor–for the foreseeable future.

We’ve doubled down on our convictions here by hiring 50 on-staff nurses who work with patients throughout trials. By clearly explaining trial requirements, communicating expectations, and staying with them every step of the way, we’re able to increase retention beyond anything we could achieve by simply automating the process with technology.

AI is not ready to be ‘solely’ relied on for core clinical trial functions and may never be. The future holds many uncertainties for a technology that has changed the landscape of every industry within a few years.

The healthcare industry as a whole is today challenged to perform a tightrope act, toeing the line between using AI to optimize processes and reserving those areas that require the dignity that the human touch can provide.

Striking that balance is a win-win for companies, patients, and our path toward progress for life-changing conditions.

[To share your insights with us as part of editorial or sponsored content, please write to psen@itechseries.com]