Researchers behind the React trial from Imperial College London, which has been tracking the disease throughout the pandemic, said the percentage of those who tested positive for Covid-19 was 0.4% among those who were double-jabbed compared to 1.21% among those unvaccinated – a three-fold difference.
After adjusting for various factors they found that vaccine effectiveness was 49% among people who reported that they had been double-jabbed.
“In other words, people who are double-vaccinated are half as likely to be infected,” said Professor Paul Elliott, director of the React programme, and chair in epidemiology and public health medicine at Imperial College London.
The study also found that the Delta variant (formerly known as the Indian variant) had “completely taken over” from the Alpha variant, with 100% of swabs provided showing that people were infected with the newer variant.
The researchers said that even if double-jabbed people come into contact with someone who has Covid-19, only one in 25 (3.84%) will go on to catch it themselves.
And cases are generally milder among double-jabbed people who do get infected, they added.
The study draws information and swab samples from more than 98,000 people randomly selected across England.
The scientists said that it was uncertain whether or not there would be an increase in infections in September when schools return and there is more indoor socialising.
But they stressed that every additional person that gets vaccinated “is taking a decent chunk of potential transmission out of what may or may not happen in September”.
The React study draws on data for those with and without symptoms across the general population.
When the authors examined data on symptomatic individuals only, the efficacy rates were higher – 59%.
Among those studied, 40% of people were found to have Covid-19 but did not have any symptoms – a figure which has varied slightly throughout the course of the study.
Prof Elliott added: “Our estimates are a bit lower than Public Health England have shown – remember they were focusing on the routine testing of people, so nearly all of them would be symptomatic, and we’re looking at effectiveness in a random sample of the general population, which includes asymptomatic individuals”.
He said that vaccinated people were likely to have less severe infections, with double vaccinated people having a lower viral load compared to those who got infected after a single dose or the unvaccinated.
But he added: “There are some double-vaccinated people who will get infected because even with very high protection, it’s not 100%”.
Researchers found that around 44% of infections were among vaccinated people but cautioned: “As a higher proportion of the population are vaccinated, then the proportion of infections in that group can go up”.
Steven Riley, professor of infectious disease dynamics at Imperial, said that young people were driving infection rates prior to the recent dip in cases, adding: “So every additional person that gets vaccinated is taking a decent chunk of potential transmission out of what may or may not happen in September.
“There’s been a drop, a plateau, now, and I think it is challenging to make any kind of prediction over the summer months.
"There is uncertainty around what might happen in September when schools return and increased indoor mixing”.
Commenting on the study, Health Secretary Sajid Javid said: “Our vaccination rollout is building a wall of defence that means we can carefully ease restrictions and get back to the things we love, but we need to be cautious as we learn to live with this virus.
“I urge anyone who has yet to receive a vaccine to get jabbed and take up both doses – the vaccines are safe and they are working”.
Listen to our coronavirus podcast