Article Impact Level: HIGH Data Quality: STRONG Summary of JAMA. https://doi.org/10.1001/jama.2024.28047 Dr. Ozan Unlu et al.
Points
- A randomized clinical trial at Mass General Brigham Health System evaluated the RECTIFIER AI tool for clinical trial patient screening, comparing AI-assisted and manual methods across 4,476 patients.
- AI-assisted screening significantly reduced the time to eligibility determination (subdistribution hazard ratio 1.78, 95% CI 1.54–2.06; P < 0.001) compared to manual screening.
- The AI method identified eligible patients at a higher rate (20.4% vs. 12.7%, P < 0.001).
- AI-assisted screening led to more patient enrollments (35 vs. 19) with a subdistribution hazard ratio 1.79 (95% CI, 1.02–3.15; P = 0.04).
- The RECTIFIER tool improves clinical trial efficiency, reducing labor, accelerating recruitment, and enhancing patient access to novel therapies.
Summary
In a recent randomized clinical trial, researchers evaluated the effectiveness of the RECTIFIER tool, an AI-assisted method for screening patients for clinical trials. The study was conducted at the Mass General Brigham Health System from May 31, 2024, to September 28, 2024, and compared the time to eligibility determination and enrollment rates between manual and AI-assisted screening. A total of 4,476 patients were randomized, with the majority (71.0%) being male and a mean age of 74.9 years. The trial showed that the AI-assisted method was significantly more efficient, with patients screened using AI achieving eligibility determination more rapidly (subdistribution hazard ratio 1.78, 95% CI, 1.54–2.06; P < 0.001) and had a higher eligibility rate (20.4% vs. 12.7%, P < 0.001). The AI-assisted screening led to more patient enrollments (35 vs. 19), with a subdistribution hazard ratio of 1.79 (95% CI, 1.02–3.15; P = 0.04).
The primary outcome of this trial was the time to eligibility determination, assessed through survival analysis, which accounted for ineligibility determinations as competing risks. The study showed a clear advantage for the AI-assisted tool, significantly reducing the time for eligibility determination compared to manual screening. Furthermore, the trial demonstrated that the use of AI-assisted screening sped up the process and resulted in a higher number of eligible patients being enrolled in the clinical trial.
This study’s findings suggest that AI-assisted tools, like RECTIFIER, can significantly enhance the efficiency of patient screening in clinical trials. By streamlining the eligibility process, these tools reduce the labor intensity of manual screening, potentially accelerating trial timelines and reducing costs. While requiring further validation in multiple settings, this approach offers promise for improving recruitment and patient access to clinical trials, ultimately contributing to the faster availability of novel therapies.
Link to the article: https://jamanetwork.com/journals/jama/fullarticle/2830514
References Unlu, O., Varugheese, M., Shin, J., Subramaniam, S. M., Stein, D. W. J., St Laurent, J. J., Mailly, C. J., McPartlin, M. J., Wang, F., Oates, M. F., Cannon, C. P., Scirica, B. M., Wagholikar, K. B., Aronson, S. J., & Blood, A. J. (2025). Manual vs ai-assisted prescreening for trial eligibility using large language models—A randomized clinical trial. JAMA. https://doi.org/10.1001/jama.2024.28047
