select search filters
roundups & rapid reactions
factsheets & briefing notes
before the headlines
Fiona fox's blog

expert comments about herd immunity

There have been questions from journalists regarding herd immunity.


Prof Willem van Schaik, Professor of Microbiology and Infection, University of Birmingham, said:

“Herd immunity describes the phenomenon that at-risk individuals are protected from infection because they are surrounded by immune individuals.  The spread of the virus is thus minimised.  Currently, we talk mostly about herd immunity in the context of vaccines.  If a sufficiently high number of individuals in a population are vaccinated, they will provide herd immunity to the small number of people that are not vaccinated (e.g. for medical or religious reasons).  We have recently seen cases of measles outbreaks where herd immunity was not sufficiently high because children were not vaccinated out of completely unfounded fears against vaccination.

“Herd immunity exists for flu.  If large proportions of the population would get a flu vaccine that could protect non-immunised individuals.  The problem with flu is that it is difficult to know which strains of flu (think of strains as variations on a theme: it is flu but just a little bit different) will be causing infections at any given point in time and so that is why the flu vaccine is not always 100% effective.  The major problem with coronavirus is that this is a novel virus that has never spread before, which means that everyone is at risk for infection.  Herd immunity can only be reached by widespread vaccination (but there is currently no vaccine, and it may take a long time before an effective vaccine becomes available) or by individuals falling ill and recovering thereby developing natural immunity against the virus.

“Unfortunately, a very rough estimate suggests that we will only reach herd immunity to Covid-19 when approximately 60% of the population is immune (and remember that immunity is currently only reached by getting the infection as we have no vaccine!).  The major downside is that this will mean that in the UK alone at least 36 million people will need to be infected and recover.  It is almost impossible to predict what that will mean in terms of human costs but we are conservatively looking at 10,000s deaths, and possibly at 100,000s of death.  The only way to make this work would be to spread out these millions of cases over a relatively long period of time so that the NHS does not get overwhelmed.  Social distancing might contribute to this.  Clearly the government believes that this process is manageable and building up herd immunity is the most effective way to stem Covid-19.  I note that the UK is the only country in Europe that is following this strategy.  Other countries also use scientific advice to guide their research and it is unclear to me why the UK is alone in their laissez-faire attitude to the virus.  Perhaps the government has access to modelling data that suggests that the numbers I quoted above in numbers of cases and deaths are unavoidable in any scenario (e.g. even with prolonged social distancing strategies) but unfortunately these data, if they exist, have not been made available to the wider academic community so it is difficult to comment.  However, last night’s U-turn banning large events suggests that the government’s policy is still very much subject to change.  This change was perhaps influenced by the unprecedented outcry of scientists on the lax containment policy of the government.

“Finally, it is true that ‘Herd immunity makes it harder for a disease to spread due to people becoming immune after already catching it or getting vaccinated’, and this is why vaccinations have been the cornerstone of the prevention of infectious diseases.  Letting a potentially lethal infectious disease burn through a population is another way to reach herd immunity, but it comes with considerable risks and downsides (see above).  This is why we still need to all work together to try to slow down and minimise the spread of Covid-19 by practicing hand hygiene and social distancing.  This will hopefully reduce the pressure on the NHS, allowing for more people to be treated and thus reducing the number of people dying of this infection.

“While Covid-19 is not as bad as the Spanish flu, which had a considerably higher case fatality rate, it is probably going to be comparable to that outbreak in terms of the societal impact and how it will affect our daily lives for the foreseeable future.  Communities will need to come together to provide ‘social herd immunity’ by helping those out that are in quarantine (for example, by checking in on them via social media or direct messaging apps and doing some shopping for them).  We are currently living through an unprecedented health crisis but we can handle this if we act collectively.”


Prof Paul Hunter, Professor in Medicine, UEA, said:

“Immunity is when an individual has acquired resistance to infection with a particular pathogen (a virus, bacterium or parasite that causes disease) because they have already had an infection with that pathogen and recovered or have been immunised.  Herd immunity occurs when a large enough proportion of a population are immune that an infection does not spread so easily or it can actually die out.

“In determining what level of herd immunity is necessary to stop the spread of infection we need to know the R0 (or reproductive ratio) this is the number of people that are likely to be infected by a single case when a new pathogen appears in a community with no prior immunity.

“So assume a pathogen with an R0=2, this means that after the first case there will be 2, then 4, then 8, etc.  But by the time half the population is immune, on average half the people exposed from a single case will be immune and therefore only one person per infected person gets the infection.

“So the sequence 1, 2, 4, 8, 16, 32, 64, 128, 256 ….. becomes 1, 1, 1, 1, 1, 1, 1, 1, 1 and the disease will actually die out fairly quickly.  Estimates of the R0 for COVID-19 vary somewhat but are in the order 2.0 to 3.0.  Assuming that R0=3 then after about 66% of the population becomes infected then the virus will die out in the population.

“But it is not as simple as this as individuals are always moving into and out of a population, largely through birth and death.  But in this case future years will see a reduction in the average age of infection and so risk of more serious disease will be lower.

“This does not mean that achieving herd immunity through natural infection should be an objective as this does imply that a lot of people will suffer the infection and risk severe illness and even death.  But if our most vulnerable people are protected during this stage then it may well be possible to ultimately reduce the death rates as infection rates in subsequent years will be substantially reduced and a vaccine may become available that would protect our most vulnerable citizens even more.”


Dr Bharat Pankhania, Senior Clinical Lecturer with the University of Exeter Medical School, said:

“The concept of herd immunity is to try to create an extremely large group of people who have immunity against an infectious agent.  This means either vaccinating, or allowing people to get infected and recover, so they have developed memory against the infecting agent and for example can produce antibodies when encountering the infectious agent again.  This constitutes the concept of immunity.

“To create herd immunity, via an infection route, people get infected and recover, and hope that once they recover, they are hopefully also immune.

“With herd immunity, when a virus circulates, there’s the hope that it will invariably encounter people who are already immune.

“In general for a good, successful herd immunity, we need more than 90 per cent of the population to be immune in order to afford protection to the remaining 10 per cent who are not immune and are therefore vulnerable from a potential infection.

“Where the concept of heard immunity works, it can be very successful – this is how we’ve eliminated small pox virus and almost wiped out polio too.

“The concept of creating herd immunity by infection is similar to creating it by vaccination. The difference is that when you vaccinate, you are using tried, tested and extremely safe vaccinations.

“Trying to create herd immunity through Covid-19 brings in questions of safety.  You can’t control infection spread to “high risk” people.  Therefore, some people who become infected will develop very severe illnesses, and some of those would die.

“There is some herd immunity against flu, which is mediated by vaccination – but it’s not excellent as the majority of the population are not vaccinated, nor is the Influenza vaccine highly immunogenic.”


Prof Martin Hibberd, Professor of Emerging Infectious Disease, London School of Hygiene & Tropical Medicine, said:

“The current aim of the social distancing put in place by the UK government seems to me to be to delay and ‘flatten’ the peak so that those people who suffer the more severe forms of the disease are able to be cared for properly; whereas if they all occurred at the same time, in a very tall and sudden peak, the healthcare facilities would be overwhelmed and not able to provide proper care.  This might mean that by winter, when the disease might be expected to get worse, those who had not had the disease would be in the minority and would be protected by herd immunity.

“I think that the delay techniques promoted by the government are likely to achieve this aim.  We are not sure how the UK policies will work compared to other European policies but I suspect they will be similar.  I feel that no European country is putting in place the full measures that China was able to use to curtail the outbreak and in any case conditions are more difficult now as the disease has become a pandemic.

“The alternative, ‘contain’ approach, that has been implemented successfully in countries such as Singapore (and recently dropped in the UK), may already be too late for many countries.

“The evidence is increasingly convincing that infection with SARS-CoV-2 leads to an antibody response that is protective.  Most likely this protection is for life, although we need more evidence to be sure of this, people who have recovered are unlikely to be infected with SARS-CoV-2 again.

“As more and more people become infected, there will be more people recovered who are then immune to future infection.  As these numbers build up, it will be harder for an infected person to transmit to other people, because some of the people that meet the infected person will be resistant (because they had the infection previously and are now immune).  When about 70% of the population have been infected and recovered, the chances of outbreaks of the disease become much less because most people are immune (resistant to infection), this is called herd immunity.

“In a good scenario, the 70% infected, recovered and immune would be people who were expected to have mild disease and the 30% who were vulnerable to severe disease would be protected by this herd immunity.

“The Government plan assumes that herd immunity will eventually happen, and from my reading hopes that this occurs before the winter season when the disease might be expected to become more prevalent.

“However, I do worry that making plans that assume such a large proportion of the population will become infected (and hopefully recovered and immune) may not be the very best that we can do.  Another strategy might be to try to contain longer and perhaps long enough for a therapy to emerge that might allow some kind of treatment.  This seems to be the strategy of countries such as Singapore.  While this containment approach is clearly difficult (and may be impossible for many countries), it does seem a worthy goal; and those countries that can should aim to do.”


Prof Matthew Baylis, Institute of Infection, Veterinary and Ecological Sciences, University of Liverpool, said:

“What is herd immunity?  In a nutshell, everyone in a population is protected from infection before all of them are immune.  The reason is that at a certain level of immunity (i.e., a proportion of people are immune, from having had the disease or having been vaccinated), the point comes when – on average – one infected person does not manage to contact and pass the infection on to one other person.  Most of their contacts are already immune.  The occasional contact is still susceptible, and the odd transmission event happens, but not often enough to sustain the disease.  Transmission grinds to a halt, even though some or even many people have still not had the disease.  This is herd immunity.  It is one of the reasons boys are vaccinated against rubella: by vaccinating boys, boys are less likely to transmit to girls (an effect of immunity), and by vaccinating boys, girls are less likely to transmit to girls (an effect of herd immunity).  For herd immunity, it does not matter whether the immunity comes from vaccination, or people having had the disease; people just need to be immune.

“A key question is how much immunity is needed before we get herd immunity?  It varies per disease, depending on how transmissible it is.  For a highly transmissible disease, like measles, on average one person might infect up 20 others, and herd immunity kicks in at 95% immune – and so, the target coverage for MMR vaccine is 95%.  For flu, on average one person infects just 1.3 others; in this case herd immunity kicks in at about 25% immune or less; and so the target coverage for flu vaccine is much less than it is for measles (three quarters of over 65s).

“So what about COVID-19?  Estimates are that one person may infect as many as 2-3 others, on average, meaning herd immunity should kick in at 50 – 67% of the population immune.  And so in the absence of a vaccine, there would appear to be nothing to stop the spread of the virus until 50-67% of us have had it; and at that point herd immunity will kicks in and transmission will decline or stop.  This is where the 60% of the population statistic has come from.  And this is deeply concerning – taking the low fatality rate estimate of 1%, even 50% of the UK population infected by COVID-19 is an unthinkable level of mortality.

“But it doesn’t have to be – and it won’t be – this way.  By reducing the number of people that one person infects, on average, then we lower the point at which herd immunity kicks in.  If we reduce it to 1.3, COVID19 becomes more like flu, and herd immunity kicks in when about one quarter of the population has had the disease and is now immune.  So, from an epidemiological point of view, the trick is to reduce the number of people we are in contact with (by staying more at home), and reduce the chance of transmission to those we are in contact with (by frequent hand washing) so that we can drive down the number of contacts we infect, and herd immunity starts earlier.  The sweet spot comes at the point where one infected person infects one, or less than one, person on average.  But, importantly, we will need to sustain this until we have a vaccine: only at that point can we return to normal behaviour patterns, with herd immunity now achieved by vaccination, not disease.”


Professor Peter Openshaw, Past President of the British Society for Immunology and Professor of Experimental Medicine at Imperial College London, said:

“Herd immunity occurs when a large percentage of the population is protected against a particular disease, stopping the ability of that disease to spread within communities.  This protection can either be gained through methods such as vaccination (which induces the body to produce antibodies which protect you against catching the disease) or through enough people in the population having been infected and generating antibodies by their body fighting the pathogen directly.  Modelling studies show that, over time, we can expect 60-80% of the population to be infected with SARS-CoV-2.  Generating herd immunity in the population, and particularly in younger individuals who are less likely to experience serious disease, is one way to stop the disease spreading and provide indirect protection to older, more vulnerable groups.

“SARS-CoV-2 is a novel virus in humans and there is still much that we need to learn about how it affects the human immune system.  Because it is so new, we do not yet know how long any protection generated through infection will last.  Some other viruses in the Coronavirus family, such as those that cause common colds, tend to induce immunity that is relatively short lived, at around three months.  However, these viruses have co-evolved with the human immune system over thousands of years meaning they may well have developed methods to manipulate our immune responses.  With the novel SARS-CoV-2, the situation may be very different but we urgently need more research looking at the immune responses of people who have recovered from infection to be sure.”


Dr Erica Bickerton, The Pirbright Institute, said:

“When a large proportion of a population become immune to an infectious agent, such as a virus, there can be “herd immunity” or “community immunity” which provides some protection to others from catching the same virus.  It is more difficult for a virus to spread throughout a population if there are not many people who can be infected.  Herd immunity is usually achieved through vaccination rather than natural infection, although natural infection can contribute to it.  However, herd immunity does depend on how long individual people remain immune and there is no guarantee that herd immunity will protect people who are not immune from being infected.

“An example can be found in seasonal influenza.  A level of herd immunity against seasonal influenza virus is achieved by annual vaccination, which is required because the circulating strains of influenza virus vary each year.  Vaccination of a high proportion of the population against seasonal influenza virus helps to protect more vulnerable members of the population, who cannot receive the annual vaccine, from being infected.

“Immunity to SARS-CoV-2 is not yet well understood and we do not know how protective the antibody response to this new virus will be in the long-term.  This is a new coronavirus and there is a lot of work going on to understand immunity to this virus.  It is too early to say how long immunity lasts or how the virus will adapt to escape immunity.  There is still much to be learned.”


Dr Simon Gubbins, The Pirbright Institute, said:

“For a viral disease “herd immunity” refers to the indirect protection an uninfected individual receives if a proportion of the population is immune to infection.  This could be achieved due to previous infection with the virus or more likely due to vaccination.  The protection comes about because in a partially immune population infected individuals are less likely to encounter uninfected ones and so transmit the virus to them.  Consequently, infection chains are interrupted and spread is stopped or slowed.

“The proportion of the population that needs to be immune for the number of new cases to decline depends on the basic reproductive ratio of the virus, known as R0.  This is the average number of secondary cases that arise from each primary case when a virus is spreading in a wholly susceptible population.

“For SARS-CoV-2 estimates for R0 are around 2.5, so the proportion of the population that needs to be immune to achieve herd immunity is around 60%.

“Herd immunity acts as an evolutionary pressure for a virus to adapt so that it can escape immunity and can spread more easily.  Influenza viruses are very good at this and frequently mutate to produce new strains to which people are not immune.  This is the reason the seasonal flu vaccine needs to be updated annually.  There is no information to show whether something similar will happen with SARS-CoV-2.”


Dr Ed Wright, Senior Lecturer in Microbiology, University of Sussex, said:

“Herd immunity is the required proportion of a population that needs to be immune to a pathogen to stop it from spreading within that same community.  This immunity can be stimulated by vaccination or recovery following infection.  The level of herd immunity required depends on how transmissible the pathogen is.

“This can be gauged from its basic reproduction number (R0) – the average number of people a positive case will go onto infect.  For instance, to stop measles virus spreading within a population requires upwards of 90% of people to have immunity because the R0 for the measles virus is high (12-18) – it’s an airborne virus.  The latest R0 for SARS-CoV-2 puts it between 2 and 3 so estimates suggest around 60% of the population would need immunity to stop the virus from taking hold in a community.”


Prof Rowland Kao, Sir Timothy O’Shea Professor of Veterinary Epidemiology and Data Science, University of Edinburgh, said:

“Herd immunity is a potentially confusing term because it really has nothing directly to do with the immune system.  When everyone in a group (i.e. a ‘herd’) is susceptible to a disease, and able to transmit it once infected, this means that once anyone in the group becomes infected, then everyone else is at risk.  And so the disease has a good chance to propagate.  However, if some of the group are protected, for example by vaccination, then this means that at least some of the time, a contact that would have been infectious, isn’t infectious, because the contact was with someone who couldn’t get infected.  Because the number of contacts over the lifetime of an infection is limited, this therefore means that the disease’s ability to reproduce is impaired.  If there are enough individuals protected so that, on average, the disease when introduced can infect less than one other, this means that the disease will infect maybe a few, but won’t spread broadly through the population.  The important point is that not everyone needs to be protected in order for the group as a whole to have little chance of getting infected.  The concept is therefore called ‘herd immunity’ because it means that, at the group or herd level, there will be relatively few infections in the group, even if there is at least one infection introduced.”


Prof Christl Donnelly, Professor of Applied Statistics at University of Oxford, and Professor of Statistical Epidemiology and Deputy Director of the WHO Collaborating Centre for Infectious Disease Modelling at Imperial College London, said:

“The basic reproduction number (denoted R-nought, R0) is the average number of new infections a single infected person causes in a fully susceptible population.  If the proportion (1 – 1/R0) of people are immune to infection (whether through vaccination or through natural infection), then the average number of new infections per infectious person is 1, which would cause the daily incidence of new infections to plateau.  Once the proportion is greater than (1 – 1/R0), then the average number of new infections per infectious person will be less than 1, which will cause the daily incidence of new infections to decline.  This is how the immunity of populations (herd immunity) reduces transmission and can contribute to control.”


Dr Thomas House, Reader in Mathematical Statistics, University of Manchester, said:

“The R0 of a disease is equal to the number of infections that a typical case will cause before they recover early in the epidemic.  If it is bigger than 1, the epidemic grows, and if less than 1, it shrinks.  Suppose R0 is 3, so the epidemic initially grows, but that at a later time over two thirds of people have become immune.  Then the average case will make three infectious contacts, but we expect that two of these will not lead to new cases due to immunity and so the epidemic will no longer grow.  This is called herd immunity and can be achieved two ways.  Ideally it happens through vaccination, which does not involve illness.  It can also happen due to infection leading to disease and later recovery.  Social distancing measures do not lead to herd immunity, so when they are lifted the epidemic may grow again.  Whether we aim for it or not, herd immunity will happen at some point in the future since neither a growing epidemic nor social distancing measures can continue forever, and the aim of policy should be for this to happen with the minimum human cost possible.”


All our previous output on this subject can be seen at this weblink:

The SMC also produced a Factsheet on COVID-19 which is available here:


Declared interests

None received.

in this section

filter RoundUps by year

search by tag