Association of Retinal Age Gap with Socioeconomic Deprivation: A Cross-Sectional Analysis of a Large Diverse Urban Cohort

PURPOSE: Socioeconomic deprivation is a profound determinant of health, driving systemic physiological dysregulation that may accelerate biological ageing. Retinal age gap (RAG) is a validated deep-learning-derived biomarker for accelerated ageing and systemic disease risk. This study investigates the relationship between socioeconomic deprivation and RAG in two large, distinct United Kingdom cohorts. METHODS: We performed cross-sectional analysis within the UK Biobank (UKB) (N = 38,412) and the London-based AlzEye (N = 49,681) cohorts. Socioeconomic deprivation was defined using Townsend Index for UKB and Index of Multiple Deprivation (IMD) for AlzEye. RAG was generated from colour fundus photography using a pre-trained deep learning model. We applied linear regression modelling to examine associations between deprivation and RAG, sequentially adjusting for chronological age, sex, ethnicity, hypertension, and diabetes. RESULTS: The AlzEye cohort was older, more diverse, and had more comorbidities compared to the UKB cohort. Modelled per decile of decreasing deprivation, there was significant association with decreased RAG in both UKB (-0.11 years per decile; 95% CI -0.13 to -0.10; p < 0.001) and AlzEye (-0.43 years per decile; 95% CI -0.46 to -0.39; p < 0.001). Following full adjustment, association remained significant in UKB (-0.04 years per decile; p < 0.001) but was attenuated and non-significant in AlzEye (-0.002 years per decile; p = 0.82). CONCLUSIONS: Socioeconomic deprivation is associated with accelerated retinal ageing, although influenced by cohort characteristics and underlying health status. Signal attenuation in AlzEye suggests comorbidities may mediate impact of deprivation on retinal ageing. Longitudinal research is required to disentangle complex interplay, ensuring equitable utility of oculomic biomarkers in precision medicine.