Applying Innovative Methods to Develop Health Education Text Messages in Cancer Survivorship

Comprehensive Summary

This study, conducted by Amanda Chriswell et al., explores the effectiveness of AI in creating SMS-based educational messages for cancer survivors and caregivers. The researchers used ChatGPT 3.5 to produce 180 SMS messages: 60 were diet related, 60 were physical activity related, and 60 were about sleep hygiene. Results showed that although ChatGPT could synthesize information at a much faster rate than humans, its output was often outdated and repetitive. Further, recommendations given by AI were vague, (e.g, telling participants to “eat more veggies” without specifying amounts), and mostly focused on weight loss instead of a holistic approach to health. Chriswell argues that AI should be used as a tool rather than a replacement for human judgment.

Outcomes and Implications

Technology is being used everywhere in the health care system, a prominent example being tele-health. However, this poses limitations for patients who do not have reliable internet or WiFi access. Amanda Chriswell et al. therefore focused on SMS-based interventions in order to make technology more accessible for all patients. However, for real clinical applications, AI and SMS must be supplemented with experts in the field in order to ensure culturally appropriate advice for patients.