Table 2.

Definitions of the scoring system.

	Score
	1	2	3
Accuracy	Fundamentally inaccurate or incorrect information, including critical errors, omissions and/or entirely incorrect treatment advice	Partially correct and accurate information, including non-critical errors and/or omitting relevant information or failing to provide specific guideline advice	Fully accurate and correct information, answering the specific question asked with no significant errors or omissions
Relevancea	Irrelevant and/or entirely tangential material, not addressing the specific question asked	Generally relevant material although including significant extraneous and/or tangential information	Relevant and focused information directly addressing the question asked, including an appropriate expansion on the relevant topic
Readability	Incoherent, unintelligible and/or garbled text, ± severely misformatted and/or oxymoronic material resulting in compromised legibility	Generally coherent and intelligible material with significant formatting and/or parsing errors	Fully coherent, well-parsed and constructed material, easily and clearly intelligible

^aInclusion of a disclaimer that the answer was provided by an AI/LLM and cannot be taken as medical advice and/or that any information or questions should also be addressed to a qualified medical practitioner was not scored negatively—as this represents a legitimate and appropriate legal disclaimer.