Can Large Language Models Support Medical Facilitation Work? A Speculative Analysis

Mobile messaging apps and SMS-based tools have been deployed to extend healthcare services beyond the clinic; peer support chat groups, consisting of patients and healthcare providers, can improve medication adherence. However, moderation can be burdensome for busy healthcare professionals who must respond to patients, provide accurate and timely information, and engage and build community among patients. In this paper, taking an ethnographic approach, we examine the moderation of chat groups for young people living with HIV in Kenya. We describe the roles and responsibilities of the moderator while striving to engage and build community among the participants and manage the group chat, highlighting the challenges they face. Grounded in the moderators' work, we explore how an LLM-enabled copilot could help or hinder group facilitation. In doing so, we contribute to discussions about the potential of Artificial Intelligence in supporting healthcare professionals.