The aim of the study was to analyse the data obtained from a 2-year follow-up study of middle-aged Finnish subjects (n = 183) with a previous history of impaired glucose tolerance (IGT) in order to elucidate the longitudinal relationships between hypertension, fasting hyperinsulinaemia and obesity. Hypertension was defined as either a systolic blood pressure (BP) of > or =160 mm Hg or a diastolic BP of > or =95 mm Hg or being on anti-hypertensive medication regardless of the BP value. Multiple logistic regression analysis adjusted for glucose tolerance status, serum lipids, exercise behaviour and alcohol consumption shows that the odds ratios of one unit (mU/l) increase in the baseline fasting insulin concentration were 1.13 (95% confidence interval 1.00-1.28) for the 2-year incidence of hypertension in subjects with IGT at baseline. Baseline body mass index (BMI) also predicted the 2-year incidence of hypertension, with an odds ratio of 1.20 (95% CI 1.02-1.42). BMI correlated positively with fasting insulin level (r = 0.54, P