PURPOSE: To assess the association of potential risk factors with early and late age-related macular degeneration (AMD) in the Thessaloniki Eye Study (TES) population Design: Population-based, cross-sectional study of subjects over age of 60 living in Thessaloniki, Greece Methods: Subjects without any AMD features and subjects with early and late AMD (neovascular AMD or geographic atrophy) were identified in the TES cohort using standardized procedures and masked grading of stereo color fundus photos. Demographic, lifestyle, systemic and other ophthalmic covariates were also collected during a detailed examination process. Their association with AMD was investigated using univariate and multivariate adjusted logistic regression models.RESULTS: Among the 2108 participants with gradable photos, the grading process identified 1204 subjects with no AMD, 848 subjects with early AMD, and 56 subjects with late AMD (24 with geographic atrophy and 32 with neovascular AMD). In multivariate analysis, compared to no AMD, late AMD was positively associated with older age (OR:1.16; 95%CI:1.10-1.22 per year of age), current smoking (smoking vs. never smoking, OR:2.34; 95%CI:1.12-4.90), prior cataract surgery (cataract surgery vs. no cataract surgery OR:2.06; 95%CI:0.96-4.40), marital status (divorced/separated vs. married, OR:3.10; 95%CI:1.08-8.93) and with 60% lower odds when sleeping in the afternoon (yes vs. no, OR:0.40; 95%CI:0.22-0.72). Early AMD was positively associated with older age (OR: 1.03; 95%CI:1.01-1.05 per year of age) and negatively with higher pulse pressure (OR:0.99; 95%CI:0.98-0.99 per mmHg).CONCLUSIONS: In TES, apart for well-known risk factors for AMD like age, smoking, and cataract surgery, two novel behavioral risk factors for prevalent late AMD were suggested. Sleeping in the afternoon was associated with 60% decreased odds for late AMD and 67% decreased odds for neovascular AMD. Being divorced/separated compared to married was associated with 3-fold higher odds for late AMD. Large longitudinal population-based studies will be necessary to further establish the potential late AMD risk effects of these two novel factors, to demonstrate potential implications of underlying pathogenetic mechanisms, and to explore preventive measures and therapeutic targets.