Clinical trials have a median cost of $19 million, with the average cost exceeding $41 million per trial. Considering the majority of trials face delays—and these can cost anywhere from hundreds of thousands to millions of dollars per day—there’s no shortage of incentive to identify opportunities to appropriately accelerate drug development.
Artificial intelligence represents a transformative shift in the way clinical trials are managed and how quickly they can be started and completed. From patient recruitment to patient retention, AI has the power to help address bottlenecks at critical junctures in clinical research and help trials cross the finish line on time.
How AI Can Support Faster Patient Recruitment
Recruitment delays are one of the biggest hurdles in clinical trials, with nearly 80% of studies failing to meet their enrollment timelines. By analyzing vast datasets like electronic medical records (EMRs), AI can identify patients that meet clinical trial eligibility criteria in far less time.
Algorithms can scan EMRs for specific markers and conditions, ensuring that no patient goes overlooked. This can be especially helpful in rare disease clinical trials, where patients tend to be harder to identify with traditional recruitment methods.
AI can also assist in prescreening potential trial participants by deploying intelligent bots on research sites’ websites. These bots use decision trees and a series of automated questions to assess patient eligibility with a certain degree of probability. This provides more immediate answers for patients while saving administrative staff from having to spend time prescreening patients on the phone.
How AI Can Support Better Patient Retention
While enrolling patients in clinical trials is challenging, keeping patients engaged is another feat. Per the Clinical Leader resource above, the average dropout rate across all clinical trials is about 30%.
AI can analyze patient data to identify early warning signs of patient frustration or noncompliance. Being able to anticipate these risks and proactively intervene prevent the study participant attrition, and by extension, minimize the costs of recruiting new ones.
For instance, AI may detect patterns of late medication logging in an eDiary, which can suggest the participant is struggling to keep up with dosing schedules. The trial team can use these insights to send customized reminders or provide a simplified medication schedule if trial protocols allow for it. Meanwhile, if AI systems find that participants from a specific region are frequently rescheduling appointments, they can arrange for virtual visits to reduce that burden.
Additional AI Use Cases
Alongside recruitment and retention efforts, AI may help accelerate trials in other respects too:
- Streamlined Clinical Trial Design: AI can identify optimal protocols based on historical clinical trial data. This reduces the need for protocol amendments later.
- Enhanced Data Collection and Analysis: AI-driven systems can collect, clean, and analyze trial data in real-time, minimizing the risk of errors and improving data quality. This is particularly useful for hybrid clinical trials involving multiple data sources.
- Regulatory Compliance Optimization: AI can analyze regulatory requirements and ensure trial designs and documentation meet global compliance standards, expediting approval timelines.
- Quality Assurance: AI can review source document creation for missing data and assess visit data against protocols to identify incomplete, incorrect, or outlying data. AI can also generate informed consent forms that comply with regulations and present information in a clear, easy-to-read format for participants.
The Challenges of AI in Clinical Trials
While AI’s potential is immense, its adoption in medical research is not without challenges.
The FDA has yet to issue comprehensive guidelines on the use of AI in clinical trials. Sponsors and research sites are treading carefully to stay compliant with privacy laws and data protection standards, knowing that regulatory dynamics may shift.
At the same time, implementing AI solutions is a significant technology and talent investment. Skilled professionals are needed to train AI systems, maintain data integrity, and engage with vendors. Costs extend beyond implementation, requiring ongoing maintenance and support.
The other layer to all this: the clinical research industry is inherently cautious when adopting new technologies. Integrating AI into trials requires careful validation and alignment with existing workflows, a process that can take a considerable amount of time.
The Road Ahead for AI
AI holds promise for accelerating drug development, with its use showcasing positive outcomes.
In one case study, AI-assisted recruitment was shown to reduce the time for identifying participants for relevant clinical trials by up to 50%.
The road ahead of AI in clinical trials may require collaboration, investment, and adaptation, but faster access to therapies and improved patient outcomes make it a journey worth taking.
At Remington-Davis, we’re committed to staying at the forefront of clinical trial innovation. We’ve been a leader in the use of other key study technologies like eSource, and we look forward to being part of the AI conversation and finding ways it can support future clinical trial success.