Search

Accrual success is one key determining factor for the success of clinical trials. Global data analyses of all terminated trials reported that 55% of trials were terminated due to low accrual rates. Failure to meet accrual goals have a significant impact on costs for sponsors, academic institutions, investigators, and society at large. The ability to predict trial accrual success with high precision before the trial starts would be highly valuable, preventing the allocation of critical resources for trials unlikely to meet accrual goals. In the present study, we constructed a dataset for predicting clinical trial failure based on poor accrual using clinicaltrial.gov data containing information on 57,846 trials. Features of the dataset were informed by prior literature and constructed using data-driven natural language processing methods. We built predictive models for accrual failure using state-of-the-art supervised machine learning protocols and methods. Models resulted in good predictive performance that was stable over a 10-year time period, with predictive performance of cross-validation AUC = 0.744 (+/-0.018) and prospective validation AUC = 0.737 (+/-0.038). We also improved model calibration and examined model performance with the reject option. These modifications enable model translation into decision support tools for various real-world settings. To the best of our knowledge, this is the first study to develop models for predicting clinical trial failure due to accrual based on a large dataset with a comprehensive set of trial features., Competing Interests: Declarations. Competing interests: The authors declare no competing interests., (© 2025. The Author(s).)

Health literacy is a prerequisite to informed health-related decision making. To facilitate understanding of information, text should be presented at an appropriate reading level for the reader. Cognitive studies suggest that the coherence of a text - the interconnectedness between the ideas it expresses - is especially important for low-knowledge readers, who lack the background knowledge to draw inferences from text that is implicitly connected only. Prior work in cognitive science has yielded automated methods to estimate coherence. These methods estimate the proximity between text representations in a semantic vector space, with the underlying idea that units of text that are poorly connected will be further apart in this space. In addition, recent work with large language models (LLMs) has produced probabilistic methodological analogues that have yet to be evaluated for this purpose. This work concerns the relationship between these automated measures and layperson comprehension of biomedical text. To characterize this relationship, we applied a range of automated measures of text coherence to a set of text snippets, some of which were deliberately modified to improve their accessibility in a series of reading comprehension experiments. Results indicate significant associations between reader comprehension - as estimated using multiple-choice questions - and LLM-derived coherence metrics. Interventions designed to improve the comprehensibility of passages also improved their coherence, as measured with the best-performing LLM-derived models and shown by improved reader understanding of the text. These findings support the utility of LLM-derived measures of text coherence as a means to identify gaps in connectedness that make biomedical text difficult for laypeople to understand, with the potential to inform both manual and automated methods to improve the accessibility of the biomedical literature., Competing Interests: Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Author TC serves on the editorial board for the Journal of Biomedical Informatics If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024. Published by Elsevier Inc.)