ObjectivesThe emergence of SARS-CoV-2 variants raised questions over the extent to which vaccines designed in 2020 have remained effective. We aimed to assess whether vaccine status was associated with the severity of Omicron SARS-CoV-2 infection in hospitalised patients.MethodsWe conducted an international, multicentric, retrospective study in 14 centres (Bulgaria, Croatia, France, Turkey). We collected data on patients hospitalised ≥24 hours between 01/12/2021 and 03/03/2022, with PCR-confirmed infection at a time of exclusive Omicron circulation, with hospitalisation related or not to the infection. Patients who had received prophylaxis by monoclonal antibodies were excluded. Patients were considered fully vaccinated if they had received at least 2 injections of either mRNA and/or ChAdOx1-S, or 1 injection of Ad26.CoV2-S vaccines.ResultsAmong the 1215 patients (median [IQR] age 73.0 [57.0; 84.0]; 51.3% males), 746 (61.4%) were fully vaccinated. In multivariate analysis, being vaccinated was associated with lower 28-day mortality (RR=0.50 [0.32-0.77]), ICU admission (R=0.40 [0.26-0.62], and oxygen requirement (RR=0.34 [0.25-0.46]), independently of age and comorbidities. When co-analysing these Omicron patients with 948 Delta patients from a study we recently conducted, Omicron infection was associated with lower 28-day mortality (RR=0.53 [0.37-0.76]), ICU admission (R=0.19 [0.12-0.28], and oxygen requirements (RR=0.50 [0.38-0.67]), independently of age, comorbidities and vaccination status.ConclusionsmRNA- and adenovirus-based vaccines have remained effective on severity of Omicron SARS-CoV-2 infection. Omicron is associated with a lower risk of severe forms, independently of vaccination and patient characteristics.