To examine the effectiveness of one dose of the COVID-19 vaccine on care-home residents. Natural experiment. We compared the effectiveness of single doses of Pfizer/BioNTech BNT162b2 (effective at 10 days) and AstraZeneca ChAdOx1 (effective at 14 days) vaccines in vaccinated and control (unvaccinated) homes. Using routine data, all care-homes reporting COVID-19 outbreaks between 11/12/2020 and 12/3/2021 in a sub-region of North West England were included. Of 126 care-homes (4042 residents), with outbreaks, 55 (44%, 1686 residents) reported onset dates before vaccination commenced; 38 (30%, 1304 residents) reported onset < 10 (BNT162b2) and < 14 days (ChAdOx1) after vaccine administration; and 33 (26%, 1052 residents) reported onset > 10 (BNT162b2) and > 14 (ChAdOx1) days after vaccination. Eighty-nine (71%) homes used ChAdOx1 vaccine. A single dose of vaccine before the outbreak onset significantly lowered the risk of symptoms (reduced by 48%), positivity (by 65%), hospitalisation (by 68%), and death (by 81%). Some vaccine effectiveness was also noted in care-homes that received one dose of vaccine within 10–14 days of outbreak onset. The number needed to vaccinate to prevent one resident from COVID-19-related hospitalisation was 34, and death was 17. This real-world, natural experiment adds to the evidence of COVID-19 vaccine effectiveness from different studies using varying designs. In the context of lockdown's impact on infection rates and on-going care-home outbreaks, a single dose of either ChAdOx1 or BNT162b2 vaccine had a significant impact on reducing COVID-19 related hospitalisation and death in care-home residents. Natural experiments should be used more in public health. Display omitted