Previous Covid-19 infection may not protect against Omicron – Expert Reaction

Early research from South Africa suggests the Omicron variant may carry a higher reinfection risk compared to the Beta and Delta variants.

Researchers looked at the data of almost three million South Africans who were previously infected with Covid-19 to see whether the risk of getting reinfected has changed with the arrival of the Beta, Delta, and Omicron variants. They found the Omicron variant is linked with a “substantial ability” to evade the immunity built up from a prior infection, while there was no such evidence of the Beta or Delta variants doing so.

This research is a preprint and has not yet been formally peer reviewed.

The SMC asked experts to comment on the research.

Associate Professor Siouxsie Wiles, School of Medical Sciences, Faculty of Medical and Health Sciences, University of Auckland, comments:

“Researchers in South Africa are doing an incredible job of trying to find out as much as they can about the new omicron variant as fast as possible and sharing it with the world. In this preprint, they’ve calculated the risk of reinfection of people who have previously had COVID-19 with this variant compared to the beta and delta variants that caused their second and third waves of disease.

“Their data suggests the risk of reinfection with omicron could be about three times higher than with beta and delta suggesting the constellation of mutations omicron possesses can help it evade our immune system.

“One confounding factor could be that the arrival of omicron may have coincided with people’s immunity waning after their previous infection. However, there is anecdotal evidence from other confirmed cases around the world that the omicron variant may more easily infect fully vaccinated people suggesting this increased risk of reinfection is real.

“What we don’t yet know is how well being vaccinated or having had COVID-19 is able to protect people from having severe disease after infection with the omicron variant. This is something it will take a few weeks to months to find out.”

No conflict of interest.

Professor Nigel French, Co-Director, One Health Aotearoa, comments:

“This is an important analysis of the available data from South Africa that provides evidence of immune escape associated with Omicron, which wasn’t evident for previous Beta and Delta variants. The authors use two approaches that come to the same conclusion and, if this were supported by ongoing experimental virus neutralisation studies, this would have major implications for global spread of the new variant. It would mean Omicron could reinfect previously infected individuals at a higher rate than previous variants of concern. There are limitations to the study acknowledged by the authors, mostly related to missing data and potential misclassification of individuals as at risk of primary or reinfection during certain periods, as well as the absence of data related to vaccination status.

“Concerning vaccination, the authors conclude that ‘increasing vaccination uptake may reduce the risks of both primary infection and reinfection’. This is important – it is still likely that two, and to a greater extent, three vaccinations (i.e., a booster) will reduce the risk associated with Omicron, compared to no vaccination. Further, we still don’t know whether this variant is associated with milder or more severe disease, so there is still much to learn.

“The study shows the importance of having high quality surveillance data that is made available for rapid epidemiological analyses.”

No conflict of interest.

Professor Michael Plank, Te Pūnaha Matatini and University of Canterbury, comments:

“This study shows that the risk of being re-infected with Covid has risen sharply in South Africa in recent weeks. At the moment, this is correlation rather than causation because the researchers didn’t have direct data on the variants involved. But nevertheless, it suggests that the omicron variant is better able to evade immunity and infect people who have had Covid previously. This is consistent with expectations based on the type of mutations omicron has.

“There is still a lot we don’t know yet about the omicron variant, including how well existing vaccines will work and whether it causes more severe disease or less. However, the early signs are concerning. New Zealand is in a good position to keep this variant out in the short-term because we still have our MIQ system. Vaccine manufacturers are confident they will be able to update vaccines for the omicron variant if needed, but this will take time. It would be prudent to maintain border settings designed to minimise the risk of importing omicron until more is known.”

Conflict of interest statement: Michael Plank is partly funded by the Department of Prime Minister and Cabinet for research on mathematical modelling of COVID-19.

Dr Matthew Hobbs, Co-Director, GeoHealth Laboratory, and Senior Lecturer in Public Health, School of Health Sciences, University of Canterbury, comments:

“Researchers in South Africa should be highly commended for the work they are doing. This study compares the risk of reinfection for people who have already had COVID-19 for the omicron variant. Risk of reinfection with omicron was estimated to be around three times higher. This is concerning as it suggests the various mutations around the spike protein that omicron has may help it get around our own immune system response.

“We still don’t know whether omicron is associated with milder or more severe disease. We also don’t know other things like how well being vaccinated is able to protect people from having severe disease after infection with omicron. However, this important evidence will emerge over the next few weeks. This study begins to piece together parts of the puzzle and perhaps confirms some of our suspicions around the heavily mutated omicron variant.”

No conflicts of interest declared.

Dr Dianne Sika-Paotonu, Immunologist, Associate Dean (Pacific), Head of University of Otago Wellington Pacific Office, and Senior Lecturer, Pathology & Molecular Medicine, University of Otago Wellington, comments:

“The recently emerged Omicron variant of the SARS CoV-2 virus classified by the WHO as a variant of concern, has already been identified in a number of different countries including Australia.

“The emergence of new variants sadly highlights some of the major inequities affecting countries associated with COVID-19 vaccine accessibility and the vaccination roll-out globally.

“Data released on 2nd December 2021 from a pre-print study involving 2,796,982 individuals from South Africa who’d previously tested positive for COVID-19, showed that of this number 35,670 persons had become reinfected with the SARS-CoV-2 virus.

“When compared with the original strain affecting South Africa, the authors of this work found no increased risk of reinfection was associated with either the Beta or Delta variants of the SARS-CoV-2 virus, however the risk of reinfection with Omicron was more than two times higher. This strongly suggests the Omicron variant is able to evade immunity, generated by previous infection.

“What is yet to yet determined however, is whether Omicron can overcome vaccine induced immunity, and whether or not Omicron leads to more severe illness, hospitalisation and death when compared to other variants. It’s important to note also that information is still needed to understand how protective the current COVID-19 vaccines are against the Omicron variant.

“Although Omicron hasn’t reached Aotearoa New Zealand yet, a maintained focus on vaccination, border controls and public health settings will be needed moving forward.

“In Aotearoa New Zealand, we’re continuing to experience the impact of the Delta variant on day 1 after movement into the new COVID-19 Protection Framework, or Traffic Light System. Lower COVID-19 case numbers are being reported in the Auckland Region, with growing case numbers outside the Auckland Region and across the country – and including in the South Island. There are currently 79 people in hospital, with nine people in ICU or HDU, with wastewater detection results indicating the SARS-CoV-2 virus is being picked up in Matamata and Gisborne.

“COVID-19 vaccination levels within Aotearoa New Zealand continue to increase with 93 per cent for first doses and 87 per cent full doses being reported. It was also encouraging to note vaccination rates in Northland have increased to 86 per cent for first doses, a testament to all the hard work being undertaken with targeted and tailored equity approaches especially with Māori-led efforts to address COVID-19 vaccine accessibility issues that have persisted.

“We all still have work to do to keep others safe from COVID-19. Please get vaccinated and tested, and reach out to help others do the same.”

No conflict of interest.

Professor Peter McIntyre, Medical Advisor, Immunisation Advisory Centre, comments:

“This is an enormously impressive study both for the size and quality of the SARS-CoV-2 infection databases maintained in South Africa and the speed with which it was produced. It confirms what was suspected – that Omicron is better able to reinfect previously infected people than was the case with Delta, consistent with it having three-fold more mutation sites, many of which were associated with presumptive immune evasion.

“However, interpretation of what this means at the population level is much more uncertain, especially with respect to the key question – to what extent is protection against severe disease (hospitalisation or death) preserved despite increased susceptibility to infection? We also do not know how many of the cases of identified re-infection occurred in people who had also received vaccine versus those who only had previous infection. This will be strongly linked to age in South Africa as – although overall vaccine coverage might only be around 25 per cent – in persons over 60 years it is close to 70 per cent. It would be important to know if re-infection risk was different in people with hybrid immunity (vaccine post infection) vs infection only or vaccine only.

“This is, however, difficult to know as in South Africa as opposed to New Zealand, where the intensity of testing is extreme, lots of people will have had infection that was not identified through testing. Serological studies have shown that it is likely that at least 70 per cent and probably a much higher percentage of the population in South Africa have been previously infected but have not been tested. It is also important to point out that despite a higher relative number of re-infections, primary infections (those with no record of previous infection) are much more numerous in their data. Although the authors do not highlight this, it is notable to me that the relative proportion of re-infections also looked high in Jan 2021 when Beta – which also had immune escape characteristics – was dominant in South Africa.

“The bottom line is that irrespective of the variant, people who are neither previously infected nor previously vaccinated are at by far the highest risk of severe disease. It may not be a problem at all to be re-infected by a new variant if you are previously infected (booster by infection rather than vaccine) as long getting a re-infection or an infection post two vaccine doses does not make you more than ‘influenza-like illness’ sick – i.e., not needing oxygen, to go to hospital, etc.

“There is now a lot of evidence that the people who have the best protection against infection or severe illness when a new variant comes along are those who have hybrid immunity due to infection then vaccination or – now becoming more common overseas – vaccine and then later infection. Either way this is a lot more protective than infection-derived immunity on its own. Unless Omicron causes a situation where previously vaccinated people get very sick, it will not be a significant issue for highly vaccinated countries except for the old, the frail, the immune compromised, for whom we need boosters and then more boosters and also to have effective antivirals available.

“Do not panic – this is a coronavirus doing what it needs to do from its point of view, which is to be able to infect new people and replicate. The virus isn’t interested in making people sick and especially not in killing them as that reduces likelihood of being passed on, so I will be very surprised if Omicron leads to a problem with severe disease, unless you are unvaccinated, as we saw with Delta. And it’s still important to get those previously infected people in South Africa vaccinated because that gives much better protection. The good news is they probably only need one dose to have better protection than we in New Zealand have even with boosters, if not old and/or frail.”

Conflict of interest statement: Professor McIntyre is on the COVID-19 Vaccine Technical Advisory Group (CV TAG) and WHO Strategic Advisory Group of Experts on Immunization (SAGE). NOTE: The Immunisation Advisory Centre (IMAC) has a contract with the Ministry of Health to delivery education and training to the healthcare sector for COVID-19 vaccines.


The UK Science Media Centre has also gathered comments in response to this research.


Professor Paul Hunter, Professor in Medicine, The Norwich School of Medicine, University of East Anglia, comments:

“The preprint posted by Pulliam and colleagues is the first such report on the transmission from the omicron variant. The authors were able to get this report in the public domain as they had only recently (11th November) published a report of reinfections in South Africa. What they have done here is updated this previous paper to include new data from recent weeks up to the 27th November to give the first indication of the epidemiology of omicron. That they have been able to get this report together so quick is a credit to the South African authors as well as a big helping of serendipity from having a paper already completed.

“Essentially the authors looked at the rate of reinfection in people (a reinfection with a positive sample occurring 90 days after the last positive specimen). The authors used two different modelling approaches to analyse national routinely collected data both at national and regional levels. The authors were not able to determine the severity of infection in the dataset they were using. The analysed data during the waves associated with the beta variant, and the delta variant and in this resubmitted pre-print the omicron variant.

“Essentially, they found that the second (beta) and third (delta) waves were not associated with increased risk of reinfection, though they point out that this observation is at odds with the results of laboratory based viral neutralisation studies which did suggest some immune escape. However, the risk of re-infection in the current wave or at least 1st to 27th November is more than twice as great as previously greater (Relative hazard Ratio 2.39 (95% confidence intervals 1.88 to 3.11).

“This finding is consistent with the hypothesis that unlike previous waves that were because the variants were intrinsically more infectious, omicron appears to be have substantial immune escape at least from immunity caused by a natural infection. Whether omicron is also more infectious is possible but cannot be answered by this analysis. This analysis is also not able to determine how much omicron can escape from immune control following vaccination, though if it can escape natural immunity it is also likely to have substantial escape potential for vaccine induced immunity.

“The implications of this paper are that omicron will be able to overcome natural and probably vaccine induced immunity to a significant degree. But, the degree is still unclear though it is doubtful that this will represent complete escape. The other big uncertainty is whether this increases the risk of severe disease, hospital admissions and deaths. With previous variants epidemiological studies showed that protection against severe disease from other variants was better maintained than protection against infection. It remains to be seen how much protection against severe disease is maintained for the omicron variant.

“But even if protection against severe disease is maintained, if case numbers increase dramatically then the pressure on hospitals will also probably rise.

“It remains the case that the extra value of the booster vaccination dose remains the most important step that we can take to reduce the probability of severe disease. I suspect new targeted vaccinations will be developed against omicron but it the infection spreads globally as rapidly as it seems to be taking off in South Africa then most of us may already have had the infection by the time a new vaccine is available.”

No conflicts of interest received.

Dr Michael Head, Senior Research Fellow in Global Health, University of Southampton, comments:

“The South Africa team deserve great credit for putting together a high-quality analyses in a short period of time, all the more so with the world watching and waiting. This analysis does look very concerning, with immunity from previous infections being relatively easily bypassed. Might this all still be a ‘false alarm’? That is looking less and less likely.

“We do not yet have up to date information on vaccine effectiveness. However, the situation should, to some extent, be different with vaccine-generated immunity. The immune response from vaccination is much stronger when compared with infection-acquired immunity. Whilst there is likely to be some impact, it is likely vaccines will still provide some level of protection. The booster dose may be key here in maintaining a high level of protection. Whilst we await more data to emerge over the coming days and weeks, the message to the general public has to be – go and get all the doses you are eligible for. Keep that protection as high as possible.”

No conflicts of interest received.

Professor Francois Balloux, Professor of Computational Systems Biology and Director, UCL Genetics Institute, UCL, comments:

“The study is based on nearly three million people with laboratory-confirmed SARS-CoV-2 in South Africa. It reports 35,670 suspected reinfections throughout the pandemic. Risk of reinfection by the omicron variant was estimated to be around three times higher than by the alpha and delta variant.

“The higher estimated reinfection ability of the omicron variant to cause reinfection is not overly surprising and could be largely anticipated based on the large number of mutations in the spike protein caried by the omicron variant, which increase the omicron variant’s ability to bypass host immunity.

“The study is competently performed and highly timely as it provides the first direct evidence for the increased ability of the omicron variant to partially bypass prior host immunity conferred by prior infection.

“Though, since the study is strictly correlative, some confounders such as waning levels of immunity could not be taken into account. It remains that it provides a plausible estimate for the increased rate of reinfection by the omicron variant.

“The study does not provide any insight on the robustness of vaccine induced immunisation against the omicron variant. It also does not report any data on infection severity. South Africa has a low vaccination rate but a large proportion of the population has been infected during previous Covid-19 waves. The population of South Africa also tend to be fairly young with a median age of 27.6 years.

“As such, the results from this study are not directly portable to other settings such as Europe or North America and more data will be needed before we can make more any robust prediction about the potential threat posed by a global spread of the omicron variant in different parts of the world.”

No conflicts of interest received.