Effect of Static vs. Conversational AI-Generated Messages on Colorectal Cancer Screening Intent: a Randomized Controlled Trial
🤖 AI Summary
This study investigates how the format of AI-generated health information—static versus conversational—impacts intention to undergo fecal-based colorectal cancer screening, relative to expert-authored materials. A randomized controlled trial compared three interventions: (1) brief, demographically tailored static messages generated by a large language model; (2) a motivational interviewing–inspired chatbot; and (3) conventional expert-written educational materials. Results showed that the static AI intervention increased screening intention by 12.9 points (on a 0–100 scale), significantly outperforming expert materials (+7.5 points) and performing non-inferiorly to the more resource-intensive conversational system. The key contribution is empirical validation that lightweight, non-interactive AI interventions can achieve high efficacy and clinical scalability in personalized health communication. This establishes a novel paradigm for precision health promotion in resource-constrained settings, balancing personalization, effectiveness, and implementation feasibility.
Application Category
📝 Abstract
Large language model (LLM) chatbots show increasing promise in persuasive communication. Yet their real-world utility remains uncertain, particularly in clinical settings where sustained conversations are difficult to scale. In a pre-registered randomized controlled trial, we enrolled 915 U.S. adults (ages 45-75) who had never completed colorectal cancer (CRC) screening. Participants were randomized to: (1) no message control, (2) expert-written patient materials, (3) single AI-generated message, or (4) a motivational interviewing chatbot. All participants were required to remain in their assigned condition for at least three minutes. Both AI arms tailored content using participant's self-reported demographics including age and gender. Both AI interventions significantly increased stool test intentions by over 12 points (12.9-13.8/100), compared to a 7.5 gain for expert materials (p<.001 for all comparisons). While the AI arms outperformed the no message control for colonoscopy intent, neither showed improvement xover expert materials. Notably, for both outcomes, the chatbot did not outperform the single AI message in boosting intent despite participants spending ~3.5 minutes more on average engaging with it. These findings suggest concise, demographically tailored AI messages may offer a more scalable and clinically viable path to health behavior change than more complex conversational agents and generic time intensive expert-written materials. Moreover, LLMs appear more persuasive for lesser-known and less-invasive screening approaches like stool testing, but may be less effective for entrenched preferences like colonoscopy. Future work should examine which facets of personalization drive behavior change, whether integrating structural supports can translate these modest intent gains into completed screenings, and which health behaviors are most responsive to AI-supported guidance.
Problem
Research questions and friction points this paper is trying to address.
Compare effectiveness of static vs. conversational AI messages for CRC screening intent
Assess scalability of AI-generated health messages in clinical settings
Evaluate AI persuasion for different screening methods (stool test vs. colonoscopy)
Innovation
Methods, ideas, or system contributions that make the work stand out.
Used AI-generated tailored messages for screening
Compared static vs. conversational AI interventions
Leveraged demographics for personalized content
🔎 Similar Papers
2024-05-09Citations: 11
