OBJECTIVE: To assess direct medical costs in patients with varicose veins (VV) and the association between VV and costs in patients with venous disorders. METHODS: 416 patients included in the VEINES cohort study in Belgium were followed for a mean of 11.34 months, with health service utilisation measured three times from physicians and patients. Items measured included consultations to physicians and health professionals, diagnostic tests, drugs, compression material and interventions (sclerotherapy, surgery, wound care, physiotherapy). Total costs (T) and costs for patients (P) were valued. RESULTS: Age and gender-standardised one-year adjusted mean costs for all patients with VV were US$ 311.24 (T) and 107.36 (P). Interventions represented 46.2% of T costs and 14.5% of P costs, whereas drug costs represented 17.1% and 22.6% respectively. In patients without VV (only symptoms or telangiectasia), T and P costs were US$ 134.72 and 75.67. Costs in patients with VV alone were US$ 304.9 (T) and 81.8 (P). Patients with VV and ulcer had the lowest costs (T: 189.1, P: 111.2) due to the small number of surgical interventions. In semilog covariance model adjusted for other venous disorders and determinants of health service utilisation in all patients with venous disorders, a cost ratio (CR) of 2.32 (95% CI: 1.25 to 4.30) was observed for patients with VV vs. without VV. Other main cost determinant was baseline measure of disease-specific quality of life (VEINES-QOL) with inverse relationship (CR for highest vs. lowest QOL scores: 0.14, 95% CI: 0.07-0.27). CONCLUSIONS: Costly interventions (e.g. surgery) are predominantly used by patients with VV alone for cosmetic reason, and with VV and skin changes possibly for ulcer prevention. Disease-specific QOL (VEINES-QOL) could be an important outcome in baseline evaluation of clinical and pharmaco-economic studies on VV, as it was associated with level of health services utilisation over 12 months.