Supplements

Expert evaluation improves clarity, safety, and athlete-specific relevance of GPT-4-generated sleep and jet lag guidance

In a two-round Delphi process with 17 sleep experts, GPT-4-generated FAQ responses on sleep and jet lag were evaluated and revised. Consensus increased from 75% (15/20 items) in round 1 to 90% (18/20 items) in round 2, with qualitative feedback indicating improved clarity, accuracy, and athlete-specific relevance after expert revision.

Last updated: Jul 28, 2026•0 RCTs•📖 Read as article →

Evidence Score

Evidence Score32/100
Human RCT☆☆☆☆☆
Meta-analysis☆☆☆☆☆
Mechanism★★★★★
Safety★★★★☆
Confidencelow

Study Evidence

Study 1. From GPT-4 to Expert-Endorsed Athlete Guidance: A Delphi Consensus on Sleep and Jet Lag.

observational

Vitale J, McCall A, Halson S, van Rensburg DCJ · Sports medicine (Auckland, N.Z.) (2026)

Participants: N/A
Duration: November 2024 to March 2025
Intervention: Two-round Delphi consensus process using GPT-4-generated FAQ responses as starting material, with expert evaluation and revision
Outcome: Proportion of items achieving consensus (≥70% rating 5-6 on 6-point Likert scale), qualitative feedback themes, and rating pattern changes
Effect Size: N/A
Population: 17 international sleep and circadian experts from the Athlete Travel & Sleep Interest Group (ATSIG)

Result:

PubMed: 42470603DOI: 10.1007/s40279-026-02484-7

Mechanism Graph

GPT-4 generated 20 FAQ responses on sleep and jet lag
↓
Experts rated items for appropriateness and provided qualitative feedback
↓
Items revised using inductive thematic coding after round 1
↓
Consensus increased from 75% to 90% after two rounds

Limitations

  • ⚠Item-level statistical comparisons did not remain significant after Bonferroni correction for multiple testing
  • ⚠GPT-4-derived content should not be used as standalone guidance; final outputs are consensus-based rather than definitive proof of correctness

Frequently Asked Questions

Can GPT-4-generated health advice be used directly for elite athletes?▼

No, GPT-4-derived content may serve as useful preliminary material for expert discussion but should not be used as standalone guidance for elite athletes.

What were the main concerns with GPT-4-generated sleep advice?▼

Common concerns included imprecise or misleading content (55%), lack of athlete-specific relevance (30%), and outdated evidence (11%).

How effective was expert evaluation in improving AI-generated guidance?▼

Expert evaluation improved clarity, safety, and athlete-specific relevance, with consensus increasing from 75% to 90% of items after two Delphi rounds.

Which topics did not reach consensus after expert review?▼

Sleep Q6 (sleep and injury risk) narrowly missed the threshold with 64.7% agreement, and Jet Lag Q9 (melatonin and sleep aids) remained below the 70% consensus threshold.

Products

Affiliate links coming soon. We only recommend products that match the doses and forms used in the cited research.

References

  1. 1.Vitale J, McCall A, Halson S, van Rensburg DCJ. "From GPT-4 to Expert-Endorsed Athlete Guidance: A Delphi Consensus on Sleep and Jet Lag.." Sports medicine (Auckland, N.Z.), 2026. PMID: 42470603 DOI: 10.1007/s40279-026-02484-7
Disclaimer: This content is for educational purposes only and is not medical advice. Evidence scores reflect the quality and quantity of available research, not clinical recommendations. Always consult a healthcare professional before starting any supplement or intervention.