During the pandemic, many scientific debates have become highly politicised. These include debates around fatality rates, the mortality impact of the pandemic, herd immunity, vaccination, and various drugs and treatments. It has been common for unlikely or absurd claims to be couched in the language of science and data.

So, it is not surprising that discussions of how the pandemic will end have political undertones too. In particular there has been much talk of the transition “from epidemic to endemic”. What exactly does this mean? Is this transition inevitable? Is it good news? And will it change how we look back at the Covid-19 epidemic?

Here’s my take on the key issues and debates.

What does ‘endemicity’ mean and is it good news?

Roughly speaking, a disease is endemic in some region if it occurs at a fairly steady level without dying away and without major flare-ups. There may be some fluctuations and seasonal variations but not large surges, namely epidemics. For example, there are four endemic human coronaviruses which tend to circulate more widely in winter, and cause symptoms of the common cold. Could SARS-CoV-2, the virus which causes Covid-19, eventually join them?

There are several common misconceptions which become apparent in discussions of Covid-19 and endemicity.

  • It is sometimes assumed, wrongly, that endemic must imply low-level. This need not be the case, although alternative terms such as “hyperendemic” are sometimes used to describe diseases which persist at high levels.
  • Low and/or steady cases for a few months do not automatically imply endemicity. If cases are steady for some time, we need to consider why and whether the situation is stable in a long term sense.
  • Endemicity need not imply that there are no efforts to control the disease. For example, malaria is endemic in many parts of the world, despite efforts to control it in these areas.
  • It may be the most likely possibility, but it is not yet certain that endemicity is the future of Covid-19 everywhere.

Since we still do not have a good idea of what burden of disease to expect if SARS-CoV-2 becomes endemic, it doesn’t really make sense to celebrate this possibility. Of course if there comes a point when the pandemic can be declared over, that will be good news.

Is there disagreement about the future of Covid-19?

As early as May 2020, Michael Ryan, executive director of the World Health Organisation’s Health Emergencies Programme suggested that SARS-CoV-2 might eventually become endemic. In a survey by the journal Nature, reported in early 2021, the scientists polled nearly all agreed that the virus is likely or very likely to become endemic in at least some regions of the world. The majority felt that elimination from even some regions is unlikely.

Certainly, there is broad consensus that eradication of the virus is looking unlikely. There are also other possibilities different from either eradication or endemicity everywhere: for example, some regions might see low levels of disease most of the time with occasional major epidemics, a situation which is harder to predict and control.

Why does eradication seem unlikely at the moment?

The answer lies in the ease with which the disease spreads, and the nature of the immunity we acquire after infection or vaccination.

Crucial to the eradication of smallpox was the fact that vaccines provide effective long-lasting protection against infection. Covid-19 vaccines, on the other hand, provide highly effective and probably long lasting protection against severe disease; but protection against infection is imperfect and wanes more quickly. The same is likely true for the immunity we get after natural infection – it provides some protection against reinfection, but probably stronger protection against severe disease.

Based on current knowledge, it seems unlikely that many places could drive transmission down to zero and keep it there. Even if some region does achieve elimination for a while, the question is whether it will be able to keep the virus out forever in an interconnected world.

Steady states of disease

Hidden behind the idea of endemicity is the mathematical notion of a stable “steady state” or “equilibrium”. Very roughly, a steady state is a state of a system which does not change too much over time. And it is stable if small changes in inputs do not cause major changes in the state. It is worth noting that different mathematical and scientific communities use all of these words in subtly different ways.

What would a steady state look like in practice? We would expect Covid-19 cases to remain fairly constant. Hospitalisations and deaths would probably remain steady too, unless new treatments become widespread. “R”, the effective reproduction number of the disease, should hover around 1: on average, each infected person would infect one more. Any more would lead to exponential growth in infections, and any less would cause infections to fade away.

Stability is important too, implying predictability and robustness. Intuitively, if the state is stable, then a few infected people coming into a population won’t provoke a large new outbreak. By contrast, if a region is Covid-free but people have little immunity to the disease, this is not a stable state: a few infected people could set off a major outbreak.

Do steady cases imply endemicity?

No. Endemicity should imply a steady state, but the converse need not be true: there are reasons why Covid-19 cases, hospitalisations and deaths could remain fairly flat for some time, perhaps even months, without endemicity.

For example, when there are (imperfect) measures to control the spread of infection, it may leak from locality to locality in a fairly steady way. There could be local flare-ups, but when you look at the whole picture, infection might appear approximately constant. We have probably seen something like this in parts of India at various times – one example was Kerala’s long plateau in daily cases during late 2020 and early 2021.

Additionally, governments may tighten measures as cases start to rise, and relax them as they fall. This is an example of so-called feedback control, and can lead to a situation where infection levels stay roughly constant. But we should be cautious about referring to such a state as endemic, unless we are sure that relaxing the control measures would not lead to a major surge.

Why is immunity important for endemicity?

When thinking about endemicity we need to think about immunity and vulnerability of the population to new outbreaks.

In order to be confident that we are looking at an endemic situation, we would like to know that levels of immunity in the population are roughly constant. Broadly, people acquire immunity after infection or vaccination, but can become susceptible to infection again as their immunity wanes. In an endemic state, these two processes should balance out, with the net result being no change in population immunity.

But things are complex in practice. Six months after infection or vaccination an individual exposed to the virus has a certain chance of getting infected, a certain chance of passing the infection on, and a certain chance of getting very ill. These probabilities change from person to person.

If we are concerned about hospitals being overwhelmed, we care about immunity against severe disease: the likelihood that a person exposed to the virus gets seriously ill. If we are worried about outbreaks, we care about the likelihood that a typical person exposed to the virus gets infected and infects others.

If Covid-19 becomes endemic in a region we would expect a high and steady level of population immunity – in all of its different senses. Crucially, it is immunity which should be the main factor stabilising transmission and preventing major surges. But confirming that this is indeed the case will not be easy.

Does endemicity mean there won’t be future epidemics?

In theory, in an endemic situation there will be some ups and downs in transmission, but no major outbreaks. In practice, with Covid-19, it’s going to be hard to be sure there won’t be more epidemics.

Consider a city like Mumbai where seroprevalence surveys, vaccination data, and modelling tell us there should be a high level of population immunity. At the time of writing, the city is seeing fairly constant cases, fatalities and hospitalisations. We can speculate that the city is close to an endemic steady state.

But we need to be cautious. We don’t know about slower trends in immunity – regular sampling of the population for antibodies to SARS-CoV-2 could provide more useful information than occasional large-scale serosurveys. Also, there is some ongoing mitigation reducing contacts between people. Further reopening could:

  • cause daily cases to rise a little but soon settle at a new steady state; or
  • cause a major new outbreak of disease.

In the first case, it would be reasonable to describe the current situation as an endemic steady state. In the second case, we shouldn’t really talk about endemicity.

Will measures to control transmission still be needed if Covid-19 becomes endemic?

Almost certainly. We don’t really know what the steady state level of infection would be in most places. How often would people get infected if life returned to pre-pandemic “normality”, and what fraction of these infections would be severe? What would the burden be on health systems and hospitals?

There are likely to be benefits to keeping disease at relatively low levels for the foreseeable future. Mitigation can continue to reduce transmission in an endemic situation. It is probable that the most effective measures in the long run would be structural changes such as improving ventilation in indoor spaces.

At the same time, ongoing vaccination may be needed to keep population immunity high. This too could become part of long-term control measures to keep transmission low.

Could new variants of the virus change the picture?

The evolution of the virus has already changed the global picture dramatically, and could do so again. A variant which spreads easily in vaccinated or previously infected populations could, in the worst case, provoke further epidemics. Even being optimistic, it would raise the level of disease at an endemic steady state. On the other hand, a new variant which causes more severe disease could push hospitalisations above an “acceptable” level, and force tighter measures to control the virus even if total transmission does not go up.

Of course, there is always the hope that vaccines will keep up with, or even outpace, variants. Just as variants could push up endemic levels of disease, or increase levels of illness, more effective vaccines could push these down. How this will play out is not yet clear.

What are the disagreements about?

Let’s turn to current discourse around the end of the pandemic. The title of a recent piece in the Wall Street Journal, “Covid Will Soon Be Endemic, Thank Goodness”, illustrates some of the issues.

First, there is the question of whether Covid-19 really will “soon” be endemic. Right now, the pandemic is still raging in many parts of the world. The United Kingdom, for example, is seeing a high and rising number of daily cases despite good vaccination coverage. In fact, the pandemic’s short-term future is very uncertain, not least because:

  • Vaccination coverage – and probably levels of population immunity – are highly uneven across the world;
  • There is still mitigation occurring in many places and we don’t know what surges could occur if control measures are lifted;
  • High and steady population immunity is crucial to endemicity, but data on long-term immune responses after vaccination or infection is still coming through.

The second questionable part of the above headline is the phrase “thank goodness”. In reality we don’t yet know what an endemic steady state would look like anywhere. How many deaths will occur every year? What will the burden on health systems be?

Any serious assessment of future endemicity will need to factor in inequality. Endemicity requires continual “topping up” of population immunity. To what extent will this be through vaccination, and to what extent through continual circulation of the virus? Endemicity could well mean low levels of disease in richer countries thanks to lower population density and periodic booster shots, and higher levels in poorer ones.

There is just too much that we don’t know about a possible endemic future to celebrate it at the moment.

Will endemicity vindicate a ‘let-the-virus-spread’ approach?

The Great Barrington Declaration, dated October 4, 2020, called for allowing infection to spread while somehow protecting the “vulnerable”. The goal was to rapidly reach herd immunity, thus allowing life to return to normal without the risk of further outbreaks of disease.

The scientific and ethical weaknesses of the declaration were clear from the start, and yet its core ideas became widespread and undoubtedly influenced policy, for example in the UK. I have previously discussed how some of its ideas likely also fed into flawed science and bad policy which exacerbated India’s second wave devastation.

If the future of Covid-19 is endemic, should this change our view of the Greater Barrington Declaration? Some of its proponents appear now to be claiming that an endemic future vindicates their call for herd immunity through natural infection. After all, if the virus will continue to circulate for the forseeable future, why not let it spread in the first place?

Such arguments are in fact absurd. If the future is of endemic disease, the GBD appears as a call to rush towards an end-point – herd immunity – is a mirage. A country following the Greater Barrington Declaration advice at the time would accelerate towards endemicity at huge human cost, only to find that, without vaccines, protecting the vulnerable from endemic SARS-CoV-2 is anyway impossible. Standing in contrast would be countries which slowed transmission and were eventually able to reach endemicity through vaccination, and without high levels of disease and death.

What will COVID-19 look like in India in the future?

Very substantial prior infection, and increasing vaccination, give some hope that India will not face another extreme and deadly surge of the kind seen during April-June 2021. But India’s population immunity comes at the cost of millions of pandemic excess deaths. We do not know how many lives could have been saved if government intervention had slowed the second wave as vaccination was being rolled out. We also do not know the long-term burden of illness as a result of this mass infection event.

It is certainly possible that endemicity nationwide is what the future holds. Indeed, elimination of the virus in any state or territory looks very unlikely in the short term. But we should be open to other possibilities too. For example, if circulation of the virus and vaccination are both limited in rural areas, the disease could become endemic in cities, with periodic rural outbreaks.

There is no strong evidence that the country as a whole is close to endemicity. The data is simply too inadequate. Weak and variable surveillance mean that we don’t have a clear idea of transmission in different parts of the country, let alone slower trends in population immunity. The picture will, no doubt, become clearer over time.


There is fairly broad agreement that the future of SARS-CoV-2 will be as an endemic virus in many, if not all, parts of the world. This would imply roughly constant levels of infection, without major outbreaks. A high and stable level of population immunity would be key to preventing surges.

Widespread endemicity reflects one possible future, and perhaps the most likely one. But the pandemic is not over, and there are many unknowns. We don’t know at what pace the transition to endemicity might occur, or whether it is already occurring in some parts of the world. Ultimately, disease could become endemic in some, but not all, areas. Or new variants could, once again, change the picture.

In an endemic future, we will need to learn what levels of disease and death to expect. What resources will be needed to control spread and top up population immunity? How will inequality, for example in access to vaccines, affect the picture? Will there be some communities forever at risk of severe outbreaks?

Ultimately endemicity is neither good nor bad news. What is bad news is that tens of millions of people have likely died in the pandemic, and that many of these deaths have occurred after vaccines were available to save lives. What will be good news is when Covid-19 is at low and predictable levels worldwide.

Murad Banaji is a mathematician at Middlesex University, London.