The aim of this study was to examine the influence of pre-pregnancy care and its effect on early glycaemic control and also the effect of glycaemic control in later pregnancy on risk of pre-eclampsia in women with type I diabetes. A prospective cohort study of 290 consecutive nonselected pregnancies in women with type I diabetes was performed from 1991 to 2002. We examined the relationship of monthly glycosylated haemoglobin (HbA1c) level, pre-pregnancy care, parity, diabetes duration, microvascular complications, maternal age, weight and smoking with risk of pre-eclampsia. Pre-eclampsia developed in 31/243 singleton births (12.8%). HbA1c level at 24 weeks was significantly increased in women with pre-eclampsia compared with women without pre-eclampsia (6.0 versus 5.6%, P= 0.017) and was, after nulliparity, the strongest independent predictor of increased risk (OR 1.65 for each 1% increase in HbA1c; P= 0.01). In contrast, there was no relationship between pre-pregnancy care or HbA1c level at booking and risk of pre-eclampsia.