ARVO2026 Recap #5 – The big picture in myopia management

Authors: Mary K. Durbin¹, Corrina McElduff³, Ian Flitcroft², James Loughman²

1. Topcon Healthcare, Oakland, New Jersey, United States
2. Centre for Eye Research Ireland, Ireland
3. Ocumetra, Ireland

This study evaluated the clinical and operational impact of a myopia management decision-support software tool. In this pilot study, myopia management software (Ocumetra, Ireland) was integrated with standard operating procedures (SOPs) across 20 locations with 186 clinicians. Clinicians completed a survey after three months addressing confidence, chair time, treatment adherence, and overall user experience. 

Of 168 respondents (90%), 89% reported improved experience, 86% reported improved confidence, 81% agreed it improved the patient journey, and 62% agreed it facilitated treatment uptake, with average chair time savings of 5 minutes per consultation. The pilot demonstrates that digital integration of SOPs with myopia management software can drive meaningful improvements in clinical confidence, operational efficiency, and treatment uptake.

[Link to abstract]

Authors: Aragaw Assaye¹, Sheela Kumaran¹, Kate Gifford², Nagaraju Konda³, P. S. Roopa Reddy⁴, Konrad Pesudovs¹

1. Optometry and vision science, University of New South Wales, Sydney, New South Wales, Australia
2. Optometry and vision science, Queensland University of Technology Faculty of Health, Kelvin Grove, Queensland, Australia
3. Optometry & Vision Science, University of Hyderabad, Hyderabad, Telangana, India
4. Ayudha Eyecare, Hyderabad, Telangana, India

This study explored the lived quality-of-life (QoL) experiences of individuals undergoing different myopia control treatments to identify associated QoL themes. 61 participants aged 8–21 years across 8 treatment modalities underwent semi-structured online interviews. 

Thematic analysis identified that myopia control treatments influence multiple aspects of QoL, with the most prevalent concerns being treatment inconvenience, myopia progression despite treatment, activity limitations, economic burden, and symptoms including blurred vision, headaches, and dryness. These findings will inform development of a myopia control–specific QoL measurement instrument to support patient-centred clinical decision-making and outcomes research.

[Link to abstract]

Authors: Bingjie Wang¹, Kathleen Watt¹, Zhi Chen^2,3, Pauline Kang¹

1. University of New South Wales, Sydney, New South Wales, Australia
2. Department of Ophthalmology and Visual Science, Fudan University Eye Ear Nose and Throat Hospital, Shanghai, China
3. Key Laboratory of Myopia, Fudan University Eye Ear Nose and Throat Hospital, Shanghai, China

This study investigated parental willingness to participate in pre-myopia interventions among Australian parents and the factors associated with this willingness. An online cross-sectional survey was conducted among Australian parents/guardians (n=800) with at least one child aged 4–18 years, assessing myopia awareness, knowledge, and willingness to consider pre-myopia interventions. 

Parents with myopia or who have a myopic child had higher awareness and knowledge of myopia. A high proportion (76.7%) of parents expressed willingness to consider pre-myopia intervention, and higher myopia knowledge consistently predicted increased willingness (OR=1.33). On the other hand, parents with both myopic as well as non-myopic children showed reduced willingness for their non-myopic children (OR=0.40). Parents who declined pre-myopia intervention cited cost as the major barrier. As knowledge of myopia is the key driver of engagement with pre-myopia interventions, promoting myopia education may help improve uptake of prevention strategies.

[Link to abstract]

Authors: Annegret Dahlmann-Noor^1,2, Maria Sifaki^1,2, Peter Thomas¹, Robert Haynes³, Paul J. Foster^2,4, Anthony P. Khawaja^2,4

1. Paediatric Ophthalmology, NIHR Moorfields Biomedical Research Centre, London, England, United Kingdom
2. Institute of Ophthalmology, University College London, London, England, United Kingdom
3. NHS Business Services Authority, Newcastle upon Tyne, England, United Kingdom
4. Translational Data Science, NIHR Moorfields Biomedical Research Centre, London, England, United Kingdom

This study described the number of spectacle prescriptions for myopia in children aged 5–15 years in England, and explored associations with socioeconomic deprivation and access to green space. This cross-sectional analysis utilised a pseudonymised General Ophthalmic Services database of spectacle prescriptions issued in England during 2024-2025, supplemented with satellite-derived vegetation index (NDVI) data. 

Of 1,248,226 prescriptions issued, 52% were for myopia. The commonest age group with myopia were 10–14-year-olds (64%), and 72% of prescriptions were issued for mild myopia (-0.50 to -2.99D) as opposed to high myopia (5%). Myopia prevalence showed a bimodal distribution across Index of Multiple Deprivation (IMD) deciles, being highest in both the poorest and wealthiest areas. There was a weak inverse correlation between green space access and myopia prescriptions. Myopia in children is more prevalent in both low and high deprivation areas, suggesting differing lifestyle risk factors for myopia development, while access to green space may offer a modest protective effect.

[Link to abstract]

Authors: Yuening Shen¹, Yadan Quan¹, Wei Lu¹, Jianguo Sun^1, Xinghuai Sun¹

1. Eye & ENT Hospital, Fudan University, Shanghai, China

This study evaluated whether general-purpose, unmodified large language models (LLMs) can distinguish glaucomatous changes from physiologic cupping in highly myopic eyes. This retrospective study enrolled colour fundus photographs from 110 eyes (55 with normal-tension glaucoma, 55 with high myopia) to three LLMs (ChatGPT-5.1, DeepSeek-V3.2, and Gemini 3 Pro), alongside patient clinical data. For each image, each model generated a diagnosis, any additional differential diagnosis, and a confidence score (0-1). 

Overall diagnostic accuracy was poor across all three models (DeepSeek 49.5%, ChatGPT 49.1%, Gemini 42.7%), with sensitivity for glaucoma detection ranging from 32.7–36.4% and AUC values of approximately 0.50, consistent with random classification. General-purpose LLMs currently show no clinically meaningful ability to differentiate glaucomatous from physiologic cupping in high myopia and cannot substitute for specialised imaging or expert clinical evaluation.

[Link to abstract]