Integrating AI is a game-changer in the ever-evolving landscape of clinical trials, and it is now being applied to clinical trial recruitment processes to help study sites process referrals more efficiently. I had the opportunity to interview three industry experts: Kurt Mussina, Chief Executive Officer at Paradigm Clinical Research; Paul Neyman, Co-Founder at Areti Health; and Ilya Gluhovsky, Ph.D., Co-Founder and CEO at Areti Health. Our discussion focused on the implementation and impact of novel AI-assisted technologies in clinical trials, particularly in patient engagement and pre-screening processes.
Moe: What are the limitations of traditional patient recruitment methods in clinical trials?
Paul Neyman: One significant limitation of traditional patient recruitment methods is their inability to scale efficiently. When a patient expresses interest in participating in a trial, the process usually begins with them filling out a form on a website. However, it can take hours or even days for a human coordinating team to reach out and make contact. This delay is problematic because their initial interest may have waned when the coordinators connect with the patient, leading to a drop in engagement. The manual nature of this process—often involving multiple phone calls and follow-ups—makes it time-consuming and labor-intensive, limiting the number of patients that can be processed simultaneously.
Additionally, the average conversion rate from initial interest to actual trial participation in the industry is notably low, typically around 2% to 3%. This low conversion rate is partly due to the inefficiencies and delays inherent in traditional recruitment methods. Not promptly engaged patients may lose interest or seek alternative options, resulting in a high dropout rate. Furthermore, the manual processes can lead to missed opportunities as coordinators may struggle to keep up with the volume of inquiries. This inefficiency slows the recruitment process, increases costs, and extends the timelines for clinical trials. Traditional methods are often not equipped to handle modern clinical research’s dynamic and fast-paced needs, highlighting the need for more efficient, scalable solutions like AI-assisted recruitment.
Moe: How does AI improve the patient recruitment process?
Paul Neyman: AI enhances patient recruitment by engaging potential participants right when they express interest. This immediate engagement is crucial as it captures the patient’s attention when their interest peaks. The AI system allows patients to ask questions and gather information in a manner that suits them, providing detailed and personalized responses that help them make informed decisions. This customized interaction builds trust and ensures patients feel valued and understood, which is essential for maintaining their interest and motivation.
Furthermore, by automating the initial stages of interaction, AI significantly streamlines the recruitment process. It efficiently prequalifies patients based on their responses to critical questions, instantly identifying those who meet the eligibility criteria. Once prequalified, the AI can seamlessly schedule on-site screening visits, ensuring patients are promptly moved to the next stage while their interest is still high. This rapid and efficient process reduces the time and effort required by human coordinators, increases patient engagement, and improves conversion rates. In addition, the AI automates rejections, further freeing up coordinator staff to focus on promising leads.
Moe: Can you describe the workflow of this AI technology from a patient’s perspective?
Paul Neyman: The AI-assisted workflow is meticulously designed to integrate seamlessly with existing site operations, strengthening their processes while maximizing efficiency. It interfaces smoothly with the current acquisition channels used by sites and clinical research organizations (CROs), such as social media campaigns, online ads, principal investigator (PI) EHR, and lead databases. When a patient engages with the AI system, it answers any questions and addresses their concerns—from compensation details to getting a single room for an overnight stay. Answers are sourced from informed consent forms (ICF), phone scripts, site information, etc. While providing these answers, the AI pitches take the next step, such as scheduling the on-site screening visit and pulling real-time availability from the site’s clinical trial management system (CTMS). Once the time is chosen, the prequalification process begins, gathering essential information and assessing eligibility. The process takes minutes, resulting in a pre-qualified lead with their scheduled appointment and their data logged in the CTMS or an automated rejection.
Furthermore, this integration is not just limited to patient acquisition and scheduling. The AI system continuously updates the CTMS with real-time data, providing coordinators with up-to-date patient interactions and status information. This ensures that when human coordinators step in, they comprehensively understand each patient’s journey, preferences, and queries already addressed by the AI. This detailed handover reduces redundancy, enhances communication, and allows coordinators to focus on higher-value tasks such as personalized patient care and complex decision-making. By plugging into existing processes and enhancing them with AI capabilities, the technology supports a more streamlined, patient-centric approach to clinical trial management, ultimately improving efficiency and outcomes.
Moe: What has been the feedback from coordinators regarding the efficiency of AI-assisted pre-screens?
Kurt Mussina: Our coordinators have found the AI-assisted pre-screening approach to be a significant step in efficiency and effectiveness. One of the primary benefits is that by the time coordinators interact with a patient; the individual is already prequalified and well-informed about the study. This pre-qualification
process means that coordinators can focus on high-quality leads, significantly reducing the time spent on unqualified candidates. This efficiency accelerates the recruitment process and ensures that patients who reach the coordinators are genuinely interested and better prepared for the next steps, enhancing the overall workflow.
Additionally, the initial feedback from our coordinators has been overwhelmingly positive. They have appreciated the streamlined process and the high quality of patient leads generated by the AI system. Coordinators have reported that the AI-assisted pre-screens provide a more targeted approach, allowing them to allocate their time and resources more effectively. While we are still collecting comprehensive data on long-term outcomes, the early indications suggest that this technology is making a tangible difference in improving the efficiency and success rate of patient recruitment for clinical trials.
Moe: What are real-world examples of how patients interacted with the AI system?
Paul Neyman: Patients have consistently expressed surprise and satisfaction with their interactions with the AI system. One noteworthy example involved a patient who, after successfully scheduling an appointment, remarked that it was “the best AI conversation of my life.” The patient’s positive remark indicates that the AI could address their needs promptly and accurately, creating a highly satisfactory experience that surpassed their expectations.
Another illustrative example involved a patient who immediately requested to sign up their spouse after signing up for a clinical trial. The AI system effortlessly managed this request, demonstrating its capability to handle complex, multi-step processes smoothly. This instance highlights the AI’s flexibility and efficiency in managing real-time, personalized interactions. It shows that patients are not only using the AI for individual needs but are also extending its use to their family members, further emphasizing the trust and reliability they place in the system.
Moe: How does AI help schedule and manage patient appointments?
Ilya Gluhovsky: AI revolutionizes the scheduling and management of patient appointments by streamlining the entire process from initial engagement to confirmed scheduling. If the patient meets the criteria, the AI seamlessly schedules an on-site screening visit directly into the CTMS. This process, which typically takes human coordinators hours or even days, is completed by AI in mere minutes. This rapid turnaround significantly reduces patient drop-off rates, ensuring that more individuals follow through with their initial interest in participating in the trial.
Furthermore, AI’s efficiency in managing appointments extends beyond just scheduling. The system can send automated reminders and updates to patients, keeping them informed and engaged throughout the trial process. This consistent communication helps maintain patient motivation and adherence to study protocols. Additionally, AI can dynamically adjust schedules to accommodate patient availability and site capacity, optimizing the use of resources and reducing the likelihood of missed appointments.
Moe: How will you scale the AI-assisted technologies across more sites and studies?
Paul Neyman: Given the positive feedback and initial success, we plan to expand the use of AI-assisted technologies to more sites and studies. This expansion will involve a strategic integration of AI into our existing recruitment and operational processes, allowing us to handle the increasing complexity of clinical trial protocols more effectively. By leveraging AI’s capabilities, we aim to streamline patient recruitment, enhance data accuracy, and reduce the burden on human coordinators, making the entire process more efficient.
We will implement a phased approach to scale successfully, starting with pilot programs at select sites to fine-tune the technology and gather further insights. This will be followed by a broader rollout across additional sites and studies. We also plan to continuously update the AI algorithms to adapt to new challenges and use cases, ensuring the technology remains cutting-edge. By focusing on patient-centricity, we hope to improve patient engagement and retention, ultimately leading to more successful clinical trials and faster development of new treatments.
Moe Alsumidaie is Chief Editor of The Clinical Trial Vanguard. Moe holds decades of experience in the clinical trials industry. Moe also serves as Head of Research at CliniBiz and Chief Data Scientist at Annex Clinical Corporation.