Background: In late 2022, the SARS-CoV-2 omicron (B.1.1.529) BA.5 sublineage accounted for most of the sequenced viral genomes worldwide. Bivalent mRNA vaccines contain an ancestral SARS-CoV-2 strain component plus an updated component of the omicron BA.4 and BA.5 sublineages. Since September, 2022, a single bivalent mRNA vaccine booster dose has been recommended for adults who have completed a primary SARS-CoV-2 vaccination series and are at high risk of severe COVID-19. We aimed to evaluate the effectiveness of a bivalent mRNA vaccine booster dose to reduce hospitalisations and deaths due to COVID-19. Methods: We did a retrospective, population-based, cohort study in Israel, using data from electronic medical records in Clalit Health Services (CHS). We included all members of CHS who were aged 65 years or older and eligible for a bivalent mRNA COVID-19 booster vaccination. We used hospital records to identify COVID-19-related hospitalisations and deaths. The primary endpoint was hospitalisation due to COVID-19, which we compared between participants who received a bivalent mRNA booster vaccination and those who did not. A Cox proportional hazards regression model with time-dependent covariates was used to estimate the association between the bivalent vaccine and hospitalisation due to COVID-19 while adjusting for demographic factors and coexisting illnesses. Findings: Between Sept 27, 2022, and Jan 25, 2023, 569 519 eligible participants were identified. Of those, 134 215 (24%) participants received a bivalent mRNA booster vaccination during the study period. Hospitalisation due to COVID-19 occurred in 32 participants who received a bivalent mRNA booster vaccination and 541 who did not receive a bivalent booster vaccination (adjusted hazard ratio 0·28, 95% CI 0·19–0·40). The absolute risk reduction for hospitalisations due to COVID-19 in bivalent mRNA booster recipients versus non-recipients was 0·089% (95% CI 0·075–0·101), and the number needed to vaccinate to prevent one hospitalisation due to COVID-19 was 1118 people (95% CI 993–1341). Interpretation: Participants who received a bivalent mRNA booster vaccine dose had lower rates of hospitalisation due to COVID-19 than participants who did not receive a bivalent booster vaccination, for up to 120 days after vaccination. These findings highlight the importance of bivalent mRNA booster vaccination in populations at high risk of severe COVID-19. Further studies with longer observation times are warranted. Funding: None.