Three Ways AI Can Accelerate Innovations in Clinical Research

Contributed Commentary by Jayme Strauss, RN, MSN, MBA, SCRN, Chief Clinical Officer, 

April 14, 2023 | Bringing a new therapy to market typically exceeds $1 billion dollars and takes 10 years. How can pharma companies safely and successfully accelerate the drug development timeline and reduce cost? Artificial intelligence (AI) enabled technologies are showing potential to increase efficiencies by improving both site and patient recruitment in clinical trials. 

AI Can Improve Diversity In Clinical Trials 

More than 56 million Americans lack access to basic medical care, according to a study by the National Association of Community Health Centers and the Robert Graham Center. In 2016, 33% of U.S. adults went without recommended care, did not see a doctor when sick, or failed to fill a prescription because of costs. 

It is concerning that a significant portion of the US population lacks access to basic medical care. That underserved patient population also faces significant challenges participating in clinical trials, due to distance, time and cost barriers to reaching trial sites. Furthermore, the underrepresentation of non-white patients in clinical trials can lead to biased or incomplete understanding of drug efficacy and safety in those populations, which may have serious consequences for public health. To address these issues, strenuous efforts must be made to increase access to medical care for underserved populations and to ensure that clinical trials are designed to be inclusive and representative of the diverse patient population. This can involve targeted outreach and recruitment strategies to reach underserved populations, as well as efforts to remove barriers to participation, such as reimbursing travel expenses and offering language translation services. 

Additionally, it is important for researchers and clinicians to consider the impact of genetic diversity on drug metabolism and treatment outcomes, as this may require tailored dosing and treatment regimens for different patient populations. By working to address these issues, we can help ensure that all patients have access to high-quality medical care and that clinical research accurately reflects the needs and experiences of diverse patient populations. 

During the COVID-19 pandemic, many clinical trial study starts were delayed and a greater reliance was placed on remote patient monitoring for ongoing studies, which helped to remove some of the barriers faced by underserved patient populations. After the pandemic, many industry stakeholders are advocating for continuing to leverage digital technologies to reach more diverse patients.  

Here are some examples of ways in which AI-enabled technologies are helping to diversify and decentralize clinical trials: 

  • AI-driven patient identifiers: Quickly identifies potential participants for clinical and observational studies based on clinical attributes 
  • AI-enabled remote patient monitoring: Collects objective, real-world health data and improves participant retention by reducing the frequency with which they have to travel to the trial site 

  • AI-enabled collaboration: Pools and analyzes large quantities of data across hospital hub and spoke networks, improving clinical data sharing and shortening time to diagnosis 

AI Can Accelerate And Increase The Accuracy Of Screening 

Research teams often use manual processes when screening potential participants for clinical trials, which takes time and increases the risk of human error.  

Compounding this issue, sponsors are reporting unprecedented difficulty getting their usual sites to sign on for clinical trials. This is due in part to increasing clinical trial demand coinciding with staff shortages after the pandemic. Therefore, many staff members lack familiarity with these processes. In fact, only 3% of physicians are investigators, and 45% of them participate in only one clinical trial.  

Additionally, screen failure rates are traditionally high, approximately 36% and the cost on average across the industry is about $1,200 per failure.  

Employing AI to conduct real-time automated assessments of potential participant eligibility can improve clinical trial enrollment efficiency. AI-enabled pre-screening can reduce the patient screening time by 34%, creating cost efficiencies, and reducing the burden on research staff.  

AI Can Bridge Gaps Between Trial Research And Clinical Care 

Most people rely on their doctors to inform them about suitable clinical trials and studies. Yet, U.S. physicians and nurses refer only small numbers of patients each year to clinical studies due in part to lack of time and access to information to confidently discuss these options with their patients. 

Tufts Center for the Study of Drug Development surveyed 589 U.S.-based physicians and 1,255 US-based nurses and found that:   

  • 89% of physicians and 71% of nurses felt comfortable discussing clinical trial opportunities with patients,  
  • 55% of physicians and 58% of nurses lacked access to clinical trial information,  

  • 49% of both physicians and nurses were unsure where to refer patients, and  

  • 35% of physicians and 25% of nurses did not have enough time to learn about the trial. 

An alternative strategy is clearly needed to bridge this knowledge gap and increase participant access to clinical trials. AI-powered tools are generating evidence and  gaining credibility and popularity as a new approach to accelerating clinical trial participant recruitment and enrolment.  

By automating analysis of hospital imaging at sites and referring institutions and comparing those results against trial inclusion and exclusion criteria in real time 24/7, AI is improving and expediting candidate pre-screening for clinical trials. Once a potential candidate is identified, AI software can employ a HIPAA-compliant messaging system to share their information with the sites and referring facilities. Then, the same platform can be used to coordinate next steps, which increases collaboration, while also simplifying the enrolment process.  

Employing AI software has been shown to accelerate enrolment rates by up to 3X, while also removing communication barriers between clinical and research teams. AI-powered capabilities can also help with clinical trial data integration and interpretation.  

The average projected return on investment (ROI) in drug research and development (R&D) for global pharmaceutical companies fell from 6.8 percent in 2021 to 1.2 percent in 2022. By harnessing digital technologies such as AI, pharma companies can improve clinical trial efficiency and diversity. As a result, AI adoption will help increase R&D productivity, bringing a better ROI and accelerate clinical research innovations - creating a faster path to market for life saving therapies. 

Jayme Strauss is the Chief Clinical Officer at Jayme has more than 10 years’ experience as an executive building clinical service lines in the areas of neuroscience and oncology. Jayme previously was the Executive Director of Neuroscience at Piedmont Healthcare in Georgia, and Assistant Vice President of Neuroscience and Oncology at Baptist Health South Florida. Jayme is passionate about evolving the way that healthcare is delivered globally, ensuring patients have access to life saving therapies, and moving science forward through research. She can be reached at