Background: Dietary intake is a known determinant of body mass index (BMI) among different populations and is therefore a useful component for BMI control. To our knowledge, no study has investigated the usual dietary intake and its association with BMI in type 2 diabetes patients among the Ugandan population. This study aimed to analyse the usual dietary intake of newly diagnosed type 2 diabetes patients and determine the association between the different dietary nutrients and BMI. Methods: We conducted a cross sectional study among 200 newly diagnosed type 2 diabetes patients in two major diabetic clinics of Kampala district. Sociodemographic, lifestyle, clinical measurements and dietary intake data were collected using a pretested structured questionnaire and a 24-h dietary recall respectively. Patients were divided according to quintile of nutrient intake. The association between dietary intake and BMI was investigated using multiple linear regression. Results: The average energy intake was 1960.2 ± 594.6 kilocalories/day. Carbohydrate, protein and fat contributed 73, 12.6 and 14.4% of the daily energy consumption respectively. We observed an inverse association between protein intake and BMI. Slopes (95% C.I) of average BMI for patients in the respective quintiles were: 0.0, -2.1 (-4.2, -0.06), -4.4 (-6.9, -1.9), -5.6 (-8.2, -3.0), and -7.3 (-10.6, -4.0); p trend <0.001. In contrast, the findings showed a positive association between carbohydrate intake and BMI. Slopes (95% C.I) of average BMI for patients in the respective quintiles were: 0.0, 3.0 (0.6, 5.4), 3.5 (0.5, 6.4), 5.2 (1.9, 8.6) and 9.7 (5.3, 14.1); p trend <0.001 after adjusting for sociodemographic, clinical and dietary intake variables. We found no significant association between the dietary intake of fibre, fat, saturated fat, polyunsaturated fat and monounsaturated fat with BMI in the final adjusted model. Conclusion: Higher intake of carbohydrate was associated with higher BMI while higher intake of protein was associated with lower BMI.