Abstract
BACKGROUND: Recreational screen time (RST) has been found to be associated with cognitive decline and neurodegenerative diseases. However, the association between RST and age-related macular degeneration (AMD), an ocular neurodegenerative disease, is still unclear. We aimed to elucidate the association between RST and AMD. METHODS: This study consisted of three parts: 1) a prospective cohort study with 482 939 UK Biobank participants and a 12.8-year median follow-up to explore the association between RST and incident AMD; 2) a two-sample Mendelian randomisation (MR) analysis using summary-level genome-wide association analysis data for RST based on 526 725 European individuals to assess causality between RST and AMD; and 3) a cross-sectional study involving 38 478 UK Biobank individuals with optical coherence tomography data to investigate the link between RST and retinal thickness. RESULTS: Multivariable Cox proportional-hazards models showed that an increase in total daily RST was associated with a greater risk of developing AMD (hazard ratio (HR) per standard deviation (SD) increase = 1.05; 95% confidence interval (CI) = 1.03, 1.07, P < 0.001). With further analysis, we revealed that daily high RST (>4 h/d) significantly increased the risk of AMD compared with low RST (0-3 h/d) (HR = 1.09; 95% CI = 1.04, 1.15, P = 0.001), while moderate RST (3.5-4 h/d) had no significant effect on AMD. Restricted cubic spline plots revealed a linear dose-response association between RST and AMD. With MR analysis we confirmed the potential causal association between RST and AMD (odds ratio per SD = 1.26; 95% CI = 1.01, 1.59, P = 0.042). Multivariable linear models suggested that an increase in RST was associated with thinning of the outer and inner retinal layers and total macular thickness. CONCLUSIONS: Longer RST may be a potential causal risk factor for AMD and may lead to pathological retinal thinning. Reducing RST could be beneficial for preventing AMD, and future research should identify the most effective time points for intervening on RST.