Gestational diabetes mellitus (GDM) has an incidence of 2±6 % and is associated with maternal morbidity, adverse perinatal outcomes and many other conditions [1±4]. Unmodifiable known risk factors associated with GDM are ethnicity, age, number of previous pregnancies, and a family history of diabetes. Recently, a short stature has been identified as an independent variable associated with gestational diabetes [5±7]. A modifiable known risk factor is obesity [4]. However, women without conventional risk factors for gestational diabetes also develop the disease [8]. Discovering additional risk factors would help identify those women who need to be tested for gestational diabetes.Additional modifiable risk factors could be a lack of exercise and dietary fat as well as life-style habits that adversely influence insulin-resistance, such as smoking and certain drugs [4]. Diabetologia (2001) Abstract Aims/hypothesis. The purpose of this study was to investigate the relation between life-style habits and glucose abnormalities in Caucasian women with and without conventional risk factors for gestational diabetes. Methods. A total of 126 pregnant women with gestational diabetes, 84 with impaired glucose tolerance and 294 with normal glucose tolerance, identified by sequential screening, were interviewed to determine their usual weekly food pattern, amount of exercise, smoking habits and alcohol intake. Results. Patients with glucose abnormalities were older and shorter in height and had significantly higher BMI before pregnancy, percentage of diabetic firstdegree relatives and higher intake of saturated fat. Patients without known risk factors for gestational diabetes (i. e. younger than 35 years of age, BMI < 25 kg/m 2 , no first-degree diabetic relatives) included 40 with impaired glucose tolerance or gestational diabetes. In a multiple logistic regression model age, short stature, familial diabetes, BMI and percentages of saturated fat were associated with impaired glucose tolerance or gestational diabetes in all patients, after adjustment for gestational age. In patients without conventional risk factors only percentages of saturated fat (OR = 2.0; 95 %-CI = 1.2±3.2) and polyunsaturated fat (OR = 0.85; 95 %-CI = 0.77±0.92) were associated with gestational hyperglycaemia, after adjustment for age, gestational age and BMI. Conclusion/interpretation. Saturated fat has an independent role in the development of gestational glucose abnormalities. This role is more important in the absence of conventional risk factors suggesting that glucose abnormalities could be prevented during pregnancy, at least in some groups of women. [Diabetologia (2001) 44: 972±978]