The relationship between cardiovascular risk factors and urine albumin excretion were studied in 474 healthy office workers. Albumin concentration was measured fresh in first morning midstream urines. Lifestyle details, oscillometric BP and lipids were assessed. Subjects with urine albumin concentration above the median (5.30 mg/l) were compared with those with albumin concentration below the median. Subjects with above median urinary albumin concentration had higher systolic blood pressure (mean 115.2 vs. 113.1 mm Hg for above median, respectively, P = 0.06), were more likely to be male (56.8 vs. 45.0%, respectively, P = 0.01) and to have lower levels of high density lipoprotein (HDL)-cholesterol (mean 1.34 vs. 1.41 mmol/l, P = 0.006). Multivariate analysis following adjustment for urine creatinine concentration to allow for urine volume confirmed the relationship with systolic blood pressure (P = 0.01) and sex (P = 0.02), and in addition revealed a relationship with alcohol intake approaching significance (mean intake 70.8 and 76.0 g/week, respectively, P = 0.06). The univariate finding of increased albuminuria with lower HDL-cholesterol appeared to be attributable to the associated relationships with male sex and lower alcohol intake. The relationships between albumin excretion and BP, male sex and alcohol intake may reflect the effects of asymptomatic developing arterial disease. The relationship with BP may also be a consequence of effects on glomerular hydrostatic or interstitial renal pressure on albumin filtration or resorption. Very low level urine albumin excretion in healthy subjects is associated with factors which predict arterial disease. Urine albumin excretion may prove to be a useful early marker of cardiovascular disease in population studies.