Increased urinary albumin-excretion (UAE) predicts cardiovascular events and clusters with the metabolic syndrome. The aim of this population-based, prospective study was to assess the relationship between baseline and longitudinal changes in cardiovascular risk-factors and 7 years' increase in UAE. Three thousand and four hundred non-diabetic participants (1838 men, 1562 women) of the Tromsø studies in 1994/1995 and 2001/2002 were included. In each survey, first-void spot-urine-samples were collected, and albumin-creatinine ratio (ACR) was calculated. Change in ACR (DeltaACR) was dichotomized into upper vs. the three lower quartiles. Median UAE in the population did not increase during follow-up. Baseline predictors for DeltaACR in the upper quartile were: age (OR 1.32 per 5 years, 95% CI 1.22-1.43), HbA1c (OR 1.43 per %, 95% CI 1.08-1.91) and waist circumference (OR 1.11 per 5 cm, 95% CI 1.04-1.19) in men, and age (OR 1.14 per 5 years, 95% CI 1.04-1.25) and current smoking (OR 1.71, 95% CI 1.27-2.30) in women. Systolic blood pressure and estimated glomerular filtration rate were predictors without gender-specificity. Clustering of three or more metabolic traits did not predict ACR increase independently. Protective factors against ACR increase were initiation of antihypertensive treatment in women (OR 0.59, 95% CI 0.39-0.87) and hard physical activity in men (OR 0.70, 95% CI 0.51-0.96). In summary, cardiovascular risk-factors at baseline predicted ACR increase, but initiation of antihypertensive therapy (women) and physical activity (men) seemed to protect from ACR increase during follow-up. Endpoint-data are needed to explore the clinical significance of low-grade UAE increase.