Sociotechnical feasibility of natural language processing-driven tools in clinical trial eligibility prescreening for Alzheimer’s disease and related dementias
Presenter
Betina Idnay, PhD, RN
Statement of Purpose
Current clinical trials continue to suffer from low recruitment rates. Improving the recruitment process is a crucial step towards eliminating barriers that inhibit participation and expediting potential treatments as they hold significant implications for enhancing better health outcomes. This challenge is potentially addressable via natural language processing (NLP) technologies for researchers to effectively identify eligible clinical trial participants. However, implementing informatics systems in clinical research, just like in any other setting, requires a deep understanding of sociotechnical systems. This present study evaluated the sociotechnical feasibility of electronic prescreening tools, both NLP and non-NLP-driven systems, for ADRD clinical research eligibility prescreening. We focused on how cognitive complexity influences system usability, identified sociotechnical gaps in system implementation, and recommended the necessary cognitive support for optimal use of NLP systems.
Learning Outcomes
Participants will be able to:
- discuss how NLP technology can optimize the prescreening process for clinical research by extracting data from both structured and unstructured EHR sources,
- interpret the feasibility of electronic prescreening tools within sociotechnical systems, and
- evaluate the impact of cognitive complexity on the usability of informatics systems, including the subjective experience of users interacting with NLP tools.