SBIR Phase I: Detecting clinical trial communication behavior and preference patterns at a large scale to predict and improve clinical trial participant retention

Award ID: 2350202

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project may be to improve the success rates of clinical trials by possibly enhancing the engagement and retention of participants. Poor clinical trial communication causes participant disengagement and attrition, resulting in incomplete data, failed trials, and associated economic losses for the pharmaceutical industry The dynamic communication behavior prediction tools that will be developed by this research may improve participant engagement through tailored communication strategies. This technology combines unsupervised machine learning and operations research models to predict participant communication and optimize contact protocols to increase engagement and retention. This is a data-driven approach to improve clinical trial decision-making, schedule flexibility, and participant outcomes, and reduce no-shows and dropout rates. This Small Business Innovation Research (SBIR) Phase I project will develop a large language model that will improve the communication between clinical researchers and the participants in clinical trials with a focus on optimizing engagement and retention to prevent trial failures. The project will use cluster analysis of communications data from several clinical trials to understand and model group behavior for key variable detection. These data will be integrated to design customized communication strategies for identified behavioral clusters. The clustering and group assignment models will be tuned to develop a synergistic model for employing optimal communication with clinical trial participants. Increased research staff productivity, improved data collection efficiency, and advances in clinical trial research scientific and technological understanding are predicted. This new technology could solve a major problem in the industry, improve patient outcomes, decrease healthcare costs, and increase the success rate of clinical trials by achieving response rates close to 95% total participation. The ultimate goal is to improve treatment efficacy and healthcare delivery quality by incorporating a multi-objective machine learning methodology to increase patient engagement in their care. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Alastair Monk