The Diabetes Control and Complications Trial (DCCT) demonstrated that in insulin dependent diabetes mellitus, the incidence of retinopathy, nephropathy and neuropathy could be reduced by intensive treatment and reduction of blood glucose concentration and glycosylated haemoglobin values to normal. the present methods of assessing diabetic control ed. urineglucose measurements or random blood glucose estimations tend to be unreliable. They only relate to blood glucose control over the preceding few hours, or minutes respectively. The newly avail-able measurement of the degree of glycosylation of haemoglobin provides an index of integrated plasmaglucose levels over a longer period of time, and a completely new tool to aid clinicians. Ninety-six diabetic cases from Medical Unit II and diabetic clinic, Mandalay General Hospital were studied over a 10-month study period from October 1994. Fastingblood sugar and HbA1c levels of patients with and without diabetic complications and risk factors were compared. There is a positive correlation between fastingblood sugar and HbA1c levels in this study. However, there is no significant difference in fastingblood sugar and HbA1c levels of diabetic patients with and without complications as patients were taking regular treatment and follow-up at the diabetic clinic. The mean duration of diabetes in diabetics with and without complications was 6.58 years and 2.69 years respectively. Patients with complications had a significantly longer duration of diabetes although their fastingblood sugar and HbA1c levels are level with those of patients without diabetic complications. This supports the evidence that long-term complications of diabetes are not solely dependent on glycaemic control and duration of diabetes plays an important role. Statistical analysis was done by correlation coefficient and Student's 't' test.