Chatbots for cascade screening

Cascade genetic testing (CGT) is a valuable strategy for identifying individuals at risk for hereditary cancer syndromes (HCS) and implementing cancer prevention measures. Despite the life-saving potential of sharing such information and encouraging screening, less than 30% of eligible relatives ultimately undergo CGT. Current methods rely on patients to inform their relatives directly, but this approach is often ineffective. The low uptake of cascade screening in cancer is multifaceted and includes several barriers the burden on patients to inform and educate their relatives, potentially complex family dynamics, a national shortage of adequately trained genetics specialists, and concerns about genetic discrimination, future insurance access, alongside associated health insurance coverage and testing costs. Better strategies are needed to enable and ease communication within families and to support downstream decision-making about CGT. Notably, new generative artificial intelligence (AI) technologies may hold promise in facilitating such interactions. The use of large language models (LLMs) have been used in multiple domains to provide conservational interfaces for patient engagement. Our objective is to prototype and evaluate an AI-based chatbot to facilitate communication of genetic results with family members, answering questions and providing curated information regarding CGT, supporting inquiries about the rationale and potential value of CGT, and providing tailored information about CGT and local resources for genetic services and cancer screening.