Information leaflets vs artificial intelligence: comparing perceptions of stroke survivors and professionals in a...

Area of Science:

• Neuroscience
• Digital Health
• Health Communication

Background:

• Stroke survivors report inadequate post-discharge information access.
• Existing stroke information resources may not fully meet survivor needs.

Purpose of the Study:

• Compare user perceptions of AI-generated stroke information versus third-sector websites.
• Identify preferred attributes of stroke information formats.

Main Methods:

• Searched UK third-sector stroke websites for relevant materials.
• Generated AI responses using ChatGPT-4o.
• Evaluated responses by stroke professionals, survivors, and caregivers on empathy, trustworthiness, reliability, comprehensibility, and usefulness.

Main Results:

• AI provided relevant responses for 87% of queried topics.
• Participants correctly identified AI responses 61% of the time.
• AI was preferred in 46% of cases; clinicians showed least preference (34%), while researchers (54%) and survivors (49%) showed higher preference. Stroke professionals found third-sector responses more empathetic.

Conclusions:

• AI-generated stroke information is positively received by survivors and researchers.
• Clinicians exhibit more skepticism towards AI-generated stroke information.
• Accessible AI tools can potentially enhance current stroke information resources.