Aims Screening and prevention strategies for Type 2 diabetes require insight into the aetiological and potentially different risk factors leading to early impairments of fasting plasma glucose (FPG) and 2-h post-load plasma glucose (2hPG) levels. We studied whether risk factors predicting subtle elevations of FPG levels were different from those predicting elevations of 2hPG levels in men and women. Methods We used baseline and 5-year follow-up data from middle-aged men and women with normal glucose tolerance (NGT) at baseline in the Danish population-based Inter99 study (n = 3164). Anthropometric and non-anthropometric baseline predictors of the 5-year FPG and 2hPG levels were estimated in linear regression models stratified by gender. Results In men, but not in women, smoking and family history of diabetes predicted increased FPG levels, whereas high physical activity predicted a decline in 2hPG levels. Among the anthropometric variables, large waist circumference was the strongest predictor of increased FPG levels in men, whereas high body mass index (BMI) was the strongest predictor of increased FPG levels in women. In both men and women, BMI and waist circumference were equally strong in predicting 2hPG levels. Furthermore, short height predicted increased 2hPG levels in men, and short height and low hip circumference predicted increased 2hPG levels in women. Conclusions Risk factors that predict future FPG levels are different from those predicting future 2hPG levels. Furthermore, different risk factors predict glycaemic levels in men compared with women. These findings indicate that different aetiological pathways may lead to Type 2 diabetes in men and women.