For the past year, there has been great interest in measuring how much protection the covid-19 vaccines give against infections and severe illness – and now the focus has turned to how the vaccines will fare against the new omicron variant. But there is another question about immunity that is relevant: how much protection is provided by a previous covid-19 infection?
Two years after the pandemic began, there have been about 250 million confirmed coronavirus infections around the world, although the real number is probably much higher because so many cases go unrecorded. In the UK, a by the University of Cambridge suggests that 25 per cent of people over 65 have been infected, rising to as many as 81 per cent of those aged 15 to 24.
One source of data on how much immunity this gives is an analysis of figures from the UK’s Office for National Statistics (ONS) large, ongoing COVID-19 Infection Survey. By tracking people’s antibody levels over time, the authors estimated in October that people would be protected against infection for an .
Advertisement
People’s responses were variable, though, with nearly a quarter of participants in this study making no detectable antibodies against the virus. That doesn’t mean they would have no immunity, though, because this analysis didn’t measure their T-cells, the other main arm of the immune system.
We can also look at figures on actual reinfection rates, rather than estimates based on people’s immune system responses. show that a prior infection cut people’s chances of testing positive for covid-19 by 80 per cent – roughly the same as having had three vaccine doses.
A more sophisticated analysis from Israel compared people with similar characteristics – such as age and how much time had passed since their infection or vaccination. This study, , found that people with prior infections had a 13-fold lower chance of being infected than people who had had two Pfizer/BioNTech jabs, although this waned over time, in a similar way to immunity from vaccines.
And a  found that the risk of severe illness in someone’s second encounter with the virus was 1 per cent that of someone getting severely ill in their first infection.
Figures such as these don’t mean that getting infected is a better strategy than getting vaccinated, says at Imperial College London. “There’s an incredible cost to infection – are you going to be hospitalised, are you going to have long covid?” he says.
As with vaccines, a prior infection may not provide complete immunity against a second infection, but it is more likely to prevent severe illness and death, says at the Intensive Care National Audit & Research Centre in London.
A key question now is how much do the figures change if a person’s second encounter with covid-19 comes from a different coronavirus variant than the one that caused their first infection? The omicron variant is known to have many mutations in its spike protein, used to enter human cells, that mean it could overcome people’s immunity, and a wave of cases is now expected around the world.
The study from Qatar suggests that protection against severe illness is maintained against at least some new variants, because the analysis spanned a period when the country experienced three waves of the coronavirus, involving the original virus from Wuhan, followed by the alpha then beta variants.
A suggests that when considering purely the number of reinfections – and not their severity – prior infections give a reduced level of protection against the omicron variant. The study found that the risk of reinfection in November, when omicron was becoming the dominant variant in South Africa, was 2.4 times greater than when the beta and delta variants had been prevalent.
Yet looked at another way, these figures are cause for optimism, says Doidge. Protection from reinfection fell from 85-91 per cent during beta and delta waves to 75 per cent with omicron, says Doidge. “That’s still a significant level of immunity.”
However, we cannot assume that reinfection rates in South Africa will be the same as in other countries, such as the UK, says at Imperial College London. As well as societal differences between countries, people’s immunity to the coronavirus is somewhat affected by which variant they first encounter – and that varies from region to region.
Boyton’s team published showing that those who had been infected with the alpha variant and then been double jabbed against the original virus had less effective antibody responses to the original virus and beta than people who had first encountered the original virus and then been double jabbed. In contrast, the first group had better antibody responses to delta. “It’s very variant-specific and quite unpredictable,” says Boyton.
“If you have got people in different parts of the globe exposed to different waves of infection, you have got people with different immune repertoires,” says Boyton. “It’s dangerous to draw too many conclusions.”
Sign up to our free Health Check newsletter for a round-up of all the health and fitness news you need to know, every Saturday
Article amended on 9 December 2021
We clarified that some people with prior infections made no detectable antibodies against the virus
