BACKGROUND: Evidence is scarce on the efficacy of long-term human albumin (HA) administration in patients with decompensated cirrhosis. The human Albumin for the treatmeNt of aScites in patients With hEpatic ciRrhosis (ANSWER) study was designed to clarify this issue. METHODS: We did an investigator-initiated multicentre randomised, parallel, open-label, pragmatic trial in 33 academic and non-academic Italian hospitals. We randomly assigned patients with cirrhosis and uncomplicated ascites who were treated with anti-aldosteronic drugs (≥200 mg/day) and furosemide (≥25 mg/day) to receive either standard medical treatment (SMT) or SMT plus HA (40 g twice weekly for 2 weeks, and then 40 g weekly) for up to 18 months. The primary endpoint was 18-month mortality, evaluated as difference of events and analysis of survival time in patients included in the modified intention-to-treat and per-protocol populations. This study is registered with EudraCT, number 2008-000625-19, and ClinicalTrials.gov, number NCT01288794. FINDINGS: From April 2, 2011, to May 27, 2015, 440 patients were randomly assigned and 431 were included in the modified intention-to-treat analysis. 38 of 218 patients died in the SMT plus HA group and 46 of 213 in the SMT group. Overall 18-month survival was significantly higher in the SMT plus HA than in the SMT group (Kaplan-Meier estimates 77% vs 66%; p=0·028), resulting in a 38% reduction in the mortality hazard ratio (0·62 [95% CI 0·40-0·95]). 46 (22%) patients in the SMT group and 49 (22%) in the SMT plus HA group had grade 3-4 non-liver related adverse events. INTERPRETATION: In this trial, long-term HA administration prolongs overall survival and might act as a disease modifying treatment in patients with decompensated cirrhosis. FUNDING: Italian Medicine Agency.