Grant: Master's Award: Canada Graduate Scholarships

This research project investigates the potential of Large Language Models (LLMs) in improving patient understanding of laboratory test results, building on studies that highlight the challenges patients face in interpreting these results. The digitization of healthcare has given patients access to their health data, yet many struggle to interpret lab results. Graphics have been shown to communicate the urgency of lab results more effectively than traditional tabular formats. LLMs, such as ChatGPT, show promise in simplifying complex health information and may help patients better understand their lab results.The objectives include identifying opportunities and perceptions around using AI for lab result interpretation, assessing whether LLM-generated summaries can help distinguish between urgent and non-urgent results, and evaluating understanding of urgency when results are presented graphically. An online quantitative survey will be conducted with a diverse sample of 1,000 Canadian healthcare consumers. Participants will evaluate four formats for presenting lab test results: Tabular, Graphical, Tabular + AI summary, and Graphical + AI summary. Outcome measures will focus on perceived urgency, behavioral intentions, and preferences for display formats.This research aims to inform the development of AI tools that enhance the interpretation of online lab test results, making them more accessible and understandable for patients with limited health literacy. The hypothesis is that visual displays will be preferred and that graphical formats will support a better understanding of urgency compared to tabular displays. By addressing patient needs and concerns, these tools could improve health outcomes and enhance the user experience with lab results.