Diabetes mellitus is both defined by and is a dreaded condition because of microvascular involvement of the eye and the kidneys. The level of glucose at which retinopathy occurs forms the basis for diagnosing diabetes. Despite non-invasive methods being available for its diagnosis, retinopathy is often underdiagnosed and untreated. Eventually it leads to visual loss often in working-age populations. There appears to be ethnic variation in susceptibility to diabetic retinopathy (DR), with prevalence of DR, sightthreatening DR and macular edema being higher in people from South Asia, Africa, Latin America and indigenous tribal populations [1]. The variations are ascribed to differential susceptibility to known risk factors, as well as differences in lifestyle, access to health care and genetic and epigenetic phenomenon. A number of prevalence studies were reported from different parts of India. In a hospital-based report from Kashmir (n, 500 with diabetes), DR was identified in 27 % (n, 135) [2]. Increasing age was a risk factor, and milder forms accounted for the most cases, suggesting that early screening and treatment is necessary [3]. Prevalence studies from rural India are fewer [4]. In a population-based Central India Eye and Medical Study involving more than 4500 subjects, DR was diagnosed by fundus photography utilizing criteria of Early Treatment of Diabetic Retinopathy Study. DR was identified in 15 subjects (0.33 %) of the entire cohort. In subjects with diabetes, DR was present in 9.6 % [5]. Risk factors were increasing age and higher glucose concentrations. All eyes showed only nonsevere forms of DR. There was no significant association with other conventional risk factors or with ocular parameters. In the SN-DREAMS III report no 2 (Sankara Nethralaya Diabetic Retinopathy Epidemiology and Molecular Genetic Study III), population-based cross-sectional screening was carried out in a rural population of south India (n, 13,079) [6]. DRwas identified using 45° four-field stereoscopic digital photography, along with 30° seven-field stereo digital pairs in those with DR. Among those with diabetes, DRwas identified in 10.3 %, with greater risk in men, insulin users, those with longer duration of diabetes, systolic hypertension and poor glycemic control [6]. In urban Chennai, the CURES Eye Study I, a populationbased screening for diabetes, was carried out in adults aged 20 years and above (n, 26,001). Among 1382 known subjects with diabetes who agreed to undergo four-field stereo colour photography, and 354 newly diagnosed subjects with diabetes, DR was present in 17.6 % [7]. The prevalence of DR was higher in men and those with proteinuria. Another population-based study in South Kerala assessed the prevalence of diabetes and DR in a community-based screening programme. One hundred and sixty camps were held in five southern districts of Kerala (Project Trinetra) [8]. In a pilot hospital-based screening for DR from western India (n, 168) using ETDRS classification, prevalence of DR was 33.9 %; non-proliferative DR was 25.5 % and proliferative DR 8.33 % [9]. The current issue of the Journal reports on the risk factors for DR in sub-Saharan Africa, prevalence in a rural south Indian population, and on biomarkers for DR and screening methods. * G. R. Sridhar sridharvizag@gmail.com