Factors Associated With Independent Completion Of An Ai-Enabled Cataract Assessment Pathway: A Real-World Analysis
Published 2026 - 44th Congress of the ESCRS
Reference: FP25.11 | Type: Free paper | DOI: 10.82333/ke80-7j54
Authors: Dr Yeshaswini Nagaraj* 1 , Ms Nicola Young 1 , Dr Ola Spencer 1 , Prof. Clare ODonnell 2
1Optegra Eye Sciences,Optegra Eye Health Care,Manchester,United Kingdom, 2Optegra Eye Sciences,Optegra Eye Health Care,Manchester,United Kingdom;Faculty of Biology, Medicine and Health,Manchester University,Manchester,United Kingdom;College of Health & Life Sciences ,Aston University,Birmingham,United Kingdom
Purpose
To evaluate whether language needs and cognitive vulnerability predict the adoption of an AI-enabled cataract pre-operative assessment pathway compared to age and gender.
Setting
Optegra Eye Health Care
Methods
A multi-site evaluation was conducted after implementing an AI-supported cataract pre-operative assessment. 14,000 consecutive patients progressing through the pathway between January 2025 & January 2026 were included. Mean patient age was 73.4 years (SD=9.6; range 42–96). 55.8% were female. Independent pathway completion was defined as successful booking after the AI assessment. Patients needing language support were identified if “Translator Required" was recorded as the final workflow step. Cognitive vulnerability was defined by noting “mental capacity assessment required”. Multivariate logistic regression was used to determine whether language needs, cognitive vulnerability, age, & gender independently predicted AI pathway completion.
Results
Of the 14,000 patients, 3,633 (25.95%) completed the assessment independently. Language support was needed by 1,182 patients (8.44%), and a cognitive review was triggered for 218 (1.56%). Language needs were linked to lower independent completion (OR 0.21, 95% CI 0.18–0.25, p<0.001). Cognitive vulnerability showed a stronger effect (OR 0.03, 95% CI 0.008–0.12, p<0.001). Increasing age was also associated with lower uptake (OR 0.98 per year, 95% CI 0.976–0.984, p<0.001). Gender was not independently related to completion (OR 1.06, 95% CI 0.98–1.14, p=0.146).
Conclusions
In real-world cataract pathway use, language and cognitive needs are stronger factors affecting AI uptake than age or gender. Digital pathway design should prioritise translation accessibility and supported assessment models to ensure fair implementation in elective eye care.