Title: Patient Reported Outcomes in Diabetes and Obesity Trials using Natural Language Processing
Dates and times: 4 November 2024, 2:30 pm – 3:30 pm CST
Authors:
Abstract:
Background: Participant reported outcomes (PROs) are an important part of clinical trials to measure symptoms and experiences. Given the FDA’s recent emphasis on patient-focused drug development, this study evaluates trends in endpoints that target PROs using natural language processing (NLP) methodology.
Methods: ClinicalTrials.Gov was programmatically accessed through its API on March 25, 2024 for drug trials targeting obesity or diabetes. NLP was used to tokenize the titles of 325 validated PRO assessments into bigrams (two-word phrases) and trigrams (three-word phrases). In addition, 32 key tokens (single words) were chosen to capture non-validated PRO assessments.
Results: PROs were rarely targeted as primary outcomes but were increasingly included in secondary outcomes since 2000.
Conclusion: The experience of hunger and satiety is critical to differentiate among new drug classes coming to market, their side effects, and their impact on patients’ everyday lives.
Senior Scientific Advisor at Clario
Dr. Emerson is a scientist with over 15 years of experience in behavioral and social science research focusing on the role of technology in measuring and intervening on health behaviors. She has extensive experience in eCOA database design and management. She is currently a Sr Scientific Advisor at Clario, a global data and technology company that helps to minimize risk in clinical trials. In her current role she supports sponsors and CROs in eCOA design best practices and preliminary study set-up to optimize the reliability of eCOA data.