Tele-Ophthalmology-Enabled and Artificial Intelligence-Ready Referral Pathway for Community Optometry Referrals of Retinal Disease: HERMES Cluster Randomised Trial with a Diagnostic Accuracy Study

AI Generated Summary*

Twenty-six community optometry practices were randomised to standard referral or hospital specialist remote review (294 adults). Among referred patients, unnecessary urgent referrals were 59.5 percentage points lower with remote review. Overall unnecessary referrals were too few for a conclusion. Incorrect referral decisions occurred in 30.2% of standard-care patients versus 4.4%. Time to consultation averaged 89 days with standard care and 84 with remote review, a nonsignificant gap; it fell to 53 days only after excluding delays, in the remote arm alone, that the investigators judged external to the pathway. Modelling found remote review cheaper and more accurate. An AI model could assess only 204 of 396 participants, performed worse than hospital experts, and resembled optometrists on urgent decisions.

*This summary was generated by AI and is published unedited. Oku does not alter these summaries. It may contain errors or omissions and is provided for general informational purposes only. Oku does not guarantee its accuracy, completeness, or reliability. For authoritative information, please refer to the original, peer-reviewed article.