Background & Aims: Effectiveness of surveillance for hepatocellular carcinoma is controversial. We here explore its effects in "real life'' clinical practice. Methods: Patients with hepatocellular carcinoma diagnosed in the period 2005-2012 in five Dutch academic centers were evaluated. Surveillance was defined as >= 2 screening tests during three preceding years and at least one radiologic imaging test within 18 months before diagnosis. Results: 295 (27%) of 1074 cases underwent surveillance. Median time interval between last negative radiologic imaging and hepatocellular carcinoma diagnosis was 7.5 months. In the surveillance group, cirrhosis (97% vs. 60%, p < 0.001) and viral hepatitis were more frequent, and non-alcoholic fatty liver disease or absence of risk factors less frequent. In case of surveillance, tumor size was significantly smaller (2.7 vs. 6.0 cm), with lower alpha-fetoprotein levels (16 vs. 44 mu g/L), earlier tumor stage (BCLC 0 and A combined: 61% vs. 21%) and resection/transplantation (34% vs. 25%) or radiofrequency ablation (23% vs. 7%) more often applied, with significantly higher 1-, 3-, and 5-year survival rates. Survival benefit by surveillance remained significant after adjustment for lead-time bias based on assumed tumor doubling time of 90 days, but not with doubling time of >= 120 days. In multivariate analysis, surveillance was an independent predictor for mortality (for interval <= 9 respectively >9 months: adjusted HRs 0.51 and 0.50, 95% confidence intervals: 0.39-0.67 and 0.37-0.69). Conclusions: Surveillance for hepatocellular carcinoma was associated with smaller tumor size, earlier tumor stage, with an impact on therapeutic strategy and was an independent predictor of survival. (C) 2015 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.