Purpose The aim of this study was to audit the knowledge and practices regarding eye care and eye complications of diabetes mellitus (DM) among diabetic patients 18 years and above in Kampala, Uganda. Methods. A cross-sectional study was done to collect data on the demographics, level of awareness and practices of the 409 diabetic patients regarding eye care and eye complications of DM. Data collected was captured in EPIDATA version 3.1, exported to STATA version 15.0 for further management and analysis. Participants characteristics were summarized using summary statistics and graphs. Using a standard questionnaire, scores for knowledge and practice for diabetes; knowledge and practice on diabetic retinopathy were generated and in all the four scores aforesaid, participants were classified as having good or poor knowledge and practice1. Proportions of participants demonstrating good awareness and good practice were reported. Fishers and Pearson chi- square tests were used to test for associations between patient’s characteristics and knowledge and practice on DM. Bivariable and logistic regression analysis was performed and variables with a p-value of < 0.2 of the unadjusted odds ratio were further analyzed at multivariate logistic regression analysis to find out factors that significantly predict patient’s knowledge and practice on diabetes mellitus. Results. A total of 409 participants were interviewed in the study, majority were females 293 (71.6%) and mean age (SD) was 50 (12) years. A high proportion of participants 314 (76.9%) was aware that DM could affect the eyes but only 24 (5.9%) stated diabetic retinopathy as an eye complication in diabetic patients. Good knowledge about diabetes mellitus was demonstrated by 178 (43.5%) of the study participants. However, only 33.3% had good knowledge on eye care and diabetic retinopathy. It was determined that female diabetic patients and those who stayed with DM for 10 years and beyond were less likely to have good practice on DM compared to male patients and those who had been with DM for less than five years (OR, 95% CI: 0.58, 0.36-0.95, P=0.029: OR, 95% CI: 0.53, 0.32-0.87, P=0.011). It was also found that diabetic patients with good knowledge of DM were at least three times more likely to have good practice compared to those with the poor knowledge (OR, 95% CI: 3.2, 2.1 -4.8, P <0.001). Conclusion Lack of knowledge regarding the importance and need for periodic eye check-up for diabetic retinopathy was a significant finding in his study. Good knowledge on diabetes, gender and duration of DM had significant association with the patients practice patterns.