BACKGROUND: At present Diabetes Mellitus is a global phenomenon with the disease topping the list, comprising of about 32 million cases, India is in the forefront with 30% of the cases. The disease affects multiple organs and is a leading cause of much morbidity and mortality. Since it is a multi-factorial disease a major step would be to identify different associated factors, for an early diagnosis and prompt treatment. The ABO blood groups are often associated with several diseases, with one blood group more often seen with the patients of a particular disease. Our study will help to determine the frequency and distribution of blood groups in correlation with Diabetes Mellitus. MATERIAL & METHODS: This study was conducted in the Gandhi Medical College, Secunderabad, during a two year period. A random study involving every third diabetic patient was chosen and their blood group was determined. A total of 300 patients were selected with 150 male and 150 female patients. Another 300 volunteers who were not diabetics were chosen as controls and their blood groups were also determined. A pro-forma was given to both diabetics and controls which included the following variables: 1. Demographic data 2. Blood grouping 3. Fasting and post prandial blood sugar. Following this, blood groups of both cohorts and controls were determined by antigen antibody agglutination method. Data analysis was done after data was entered into excel sheet and double checked for errors using SPSS Software RESULTS: Our analysis showed that O group was significantly more among diabetic patients when all patients were compared to control.(2) there was a preponderance of blood group O among female diabetics and B among male diabetics. CONCLUSION: ABO blood groups have been determined in 300 diabetic patients and compared with the controls comprising of a series of 300 voluntary blood donors. When the results were analysed on the basis of sex, there was preponderance of group O in female diabetics. It can be observed that no uniform association has been found between blood groups and diabetes although the preponderance of one or the other blood group has been reported from time to time.. The results have been so variable and often inconclusive that the possible role of ethnic and social factors cannot be ignored.