The prevalence of QT interval prolongation is higher in people with diabetes and its complications. Sudden death has been reported as a common cause of death in insulin-dependent diabetic patients affected by autonomic neuropathy. It has been postulated that QT prolongation predisposes to cardiac arrhythmias and sudden death. In this analysis the prevalence of QT interval prolongation and its relation with diabetic complications were evaluated in the EURODIAB IDDM Complications Study (3250 insulin-dependent diabetic patients attending 31 centres in 16 European countries). Five consecutive RR and QT intervals were measured with a ruler on the V5 lead of the resting ECG tracing and the QT interval corrected for the previous cardiac cycle length was calculated according to the Bazett's formula. The prevalence of an abnormally prolonged corrected QT was 16 % in the whole population, 11 % in males and 21 % in females (p < 0.001). The mean corrected QT was 0.412 s in males and 0.422 s in females (p < 0.001). Corrected QT duration was independently associated with age, HbA_{1 c} and blood pressure. Corrected QT was also correlated with ischaemic heart disease and nephropathy but this relation appeared to be stronger in males than in females. Male patients with neuropathy or impaired heart rate variability or both showed a higher mean adjusted corrected QT compared with male patients without this complication. The relation between corrected QT prolongation and autonomic neuropathy was not observed among females. In conclusion we have shown that corrected QT in insulin-dependent diabetic female patients is longer than in male patients, even in the absence of diabetic complications known to increase the risk of corrected QT prolongation. [Diabetologia (1999) 42: 68–75]