Objective: To define the prevalence and describe the natural history of microalbuminuria (MA) in a population-based sample of children with type 1 diabetes mellitus (T1DM).Methods: Children with T1DM diagnosed <= 16 yrs and screened for MA were identified through the Western Australia diabetes register. Three-monthly hemoglobin (HbA1c) was performed from diagnosis. MA screening (albumin excretion rate in three timed overnight samples) was performed yearly after 10 yrs of age or after 5 yrs from diagnosis. Cox proportional model assessed the risk of different variables on the occurrence of MA. Kaplan-Meier survival analyses (log-rank test) estimated the probability of developing MA.Results: Nine hundred and fifty-five T1DM children (462 male), mean diabetes duration of 7.6 yrs and mean age at onset of diabetes of 8.5 yrs were selected for the study. MA, mean albumin excretion rate >= 20 and <200 mu g/min, developed in 128 subjects (13.4%) at mean diabetes duration of 7.6 yrs. Cumulative probability for MA was 16% after 10 yrs. Determinants for MA were HbA1c [hazard ratio (HR) 1.21; 95% confidence interval (CI) 1.05-1.38; p = 0.007], onset of puberty (HR 8.01; 95% CI 3.18-20.16; p < 0.001) and age at diagnosis (HR 1.25; 95% CI 1.18-1.33; p < 0.001). Females had a higher probability for MA during puberty than males (p = 0.03). The total incidence of MA (subjects with MA/100 person-years) was 1.26, 1.85 and 2.44 for those who developed diabetes at ages < 5 yrs, 5-11 yrs and >11 yrs, respectively.Conclusions: Onset of puberty, diabetes duration and metabolic control are major factors predisposing the development of MA. Children diagnosed with T1DM at younger ages have a prolonged time for developing MA.