This retrospective, population-based, cross-sectional study evaluated the association between vitamin D deficiency and retinopathy severity in diabetic patients with poorly or well controlled glycaemia. Other potential risk factors for diabetic retinopathy severity were also assessed. The National Health and Nutrition Examination Survey (NHANES) 2005-2008 data were used for the study. Outcomes assessed included retinopathy severity, HbA1c levels, socioeconomic, behavioral, and biological factors. Univariate and multivariate regression analysis was used to evaluate association of different parameters with retinopathy severity. The interaction among HbA1c control, vitamin D deficiency, and retinopathy severity were also explored. The population included 842 adults (52.8% women) with mean age of 61.2 years. Retinopathy was detected in 301 subjects (35.7%). Mild non-proliferative retinopathy (NPR) was present in 195 subjects (23.2%), severe non-proliferative and proliferative retinopathy in 106 subjects (12.6%). Multivariate ordinal regression analysis found being male (odds ratio (OR): 1.602, P=0.001), increased duration of diabetes (OR: 1.072, P=3.77E-7) and poorly controlled HbA1c (OR: 3.522, P=2.00E-5) were associated with greater retinopathy severity. The association between vitamin D deficiency and retinopathy severity only found in diabetic patients with well controlled glycaemia. The findings of this study indicate that vitamin D deficiency associated with severe diabetic retinopathy in patients with well controlled diabetes. The findings provide possible relationship for the previous conflict results, and highlight the need for controlling modifiable risk factors to reduce the development of sever diabetic retinopathy.