Matt Hancock is wrong about herd immunity Confusion about the Covid-19 science is hampering debate — and costing lives

Let’s have a look at the diseases he mentions. Measles, if it arrives on ‘virgin soil’, can devastate a population. In Tahiti and Moorea and the South-east and North-west Marquesas, between 20% and 70% of the population was lost to the first epidemic. Natural infection with measles provides lifelong immunity, and we now have a vaccine which provides similar solid, durable protection. We have not been able to eliminate the disease, but those who rather selfishly choose not to vaccinate their children are only able make that choice because the risks of infection are kept low by those who are immune — currently, a combination of those, like me, who caught it and recovered and many others for whom it is vaccine induced. The vaccine does not work in babies, which is why you have to wait till they are a year old before they get it. We can do this because herd immunity keeps the risk of infection down, so they are are unlikely to be infected in their first year of life. Without this herd protection, many under ones would die (as they regularly do in sub-Saharan Africa) despite a vaccine being available.

Malaria (a primary focus of my research) is caused by a parasite which carries at least 60 different outfits with which to disguise itself from immune attack. We, nonetheless, typically acquire sufficient immunity upon first infection so as not to suffer severe disease and death upon further attacks. This is a feature it probably has in common with Covid-19. But that is where the similarity ends. It has been difficult to make a malaria vaccine that covers all of its diversity, but fortunately that should not be a problem for Covid-19.

AIDS is caused by the Human Immunodeficiency Virus which also has an extraordinary capacity to change outfits during the course of infection. This, and other mechanisms of subverting the immune response, allow it to persist indefinitely in an infected person. Even so, rates of infection will slow down as the virus runs out of people to infect.

Fortunately, Covid-19 does not possess such an array of immune evasion mechanisms that could seriously compromise any possibility of vaccine development in the next few years. Like HIV, influenza also has the ability to change its outfit, and periodically a new strain emerges requiring a new vaccine. It is because sufficient immunity accumulates in the population that a radical change of outfit becomes worthwhile. What Mr Hancock means when he says we do not reach herd immunity to flu is that it finds ways around it; it is unlikely, given the nature of coronaviruses that the SARS Cov-2 virus would be capable of doing the same thing.

In a nutshell, the development of immunity through natural infection is a common feature of many pathogens, and it is reasonable to assume that Covid-19 does not have any tricks up its sleeve to prevent this from happening — it would pose a very serious problem for the development of a vaccine if it did. Having said this, the Covid-19 virus belongs to a family of viruses which do not typically give you lifelong immunity against infection. Most of us will never have heard of these other four ‘seasonal’ coronaviruses that are currently circulating in our communities. And yet surveys indicate that at least 3% of the population is infected by any single one of these corona cousins during the winter months. These viruses can cause deaths in high risk groups or require them to receive ICU care or ventilator support, so it is not necessarily true that they are intrinsically milder than the novel Covid-19 virus. And like the Covid-19 virus, they are much less virulent in the healthy elderly and younger people than influenza.

One important reason why these corona cousins do not kill large numbers of people is because even though we lose immunity and can be reinfected, there is still always a decent enough proportion of immune people in the population to keep the risk of infection low to those who might die upon contracting it. Also, all the coronaviruses in circulation — including the Covid-19 virus— have some features in common which means that getting one coronavirus will probably offer some protection against other coronaviruses. This is becoming increasingly clear from work in many labs, including my lab in Oxford. It is against this background of immunity from itself and its close relations that Covid-19 virus has to operate.

So what does Mr Hancock mean by “reaching” herd immunity? Herd immunity is a continuous variable which increases as people become immune and decreases as they lose immunity or die. He is perhaps referring to the threshold of herd immunity at which the rate of new infections starts to decrease. We do not yet have a very clear idea of what this threshold is for Covid-19 as the landscape in which it spreads includes people who are susceptible to it, people who have built up immunity to it, and people who have immunity to other coronaviruses.

Unfortunately, we do not have a good way of telling how many people have actually been exposed to the new virus, or how many people were resistant to start with. We are able to test for antibodies – and my lab in Oxford has been doing so since early April – but, as with other coronaviruses, Covid-19 antibody levels decline after recovery, and some people do not make them at all, and so antibody levels will not give us the answer. More and more evidence is accumulating that other arms of immunity, like T cells, play an important role.

Indications of the herd immunity threshold having been reached are available from the time signatures of epidemics in various parts of the world where death and infection curves tend to “bend” in the absence of intervention or to stay down when interventions were relaxed (in comparison with other locations where the opposite happened). But we do not know how far we are from it in most parts of the UK. It is important to bear in mind that the attainment of the herd immunity threshold does not lead to disease eradication. Instead it corresponds to an equilibrium state in which the infections lingers at low levels in the community. This is the situation we tolerate for most infectious diseases (like flu which kills 650K people every year globally). The situation can be vastly improved through vaccination, but it is very difficult to eliminate the disease even with a good vaccine.

We are of course also able to test for presence of the virus, and there is much attention on this with ‘test and trace’ strategies. However this test, known as the PCR test, is of limited value as it cannot tell us whether someone is infectious and can pass on the disease, whether they have the virus but cannot pass it on, or indeed whether the virus has been destroyed by the immune system and only fragments remain. This means that we need to make public health decisions based on only partial information, and in a changing environment, and is why assumptions of how many people have been infected and are immune are so important.

The Great Barrington Declaration proposes a solution for how we may proceed in the face of such uncertainty. It suggests that we exploit the feature of this virus that it does not cause much harm to the large majority of the population to allow them to resume their normal lives, while shielding those who are vulnerable to severe disease and death.

Under these circumstances, immunity will build up in the general population to a level that poses a low enough risk of infection to the vulnerable population that they may resume their normal lives. All of this can happen over a period of six months, and so this Focused Protection plan does not involve the permanent segregation of the vulnerable from the rest of the population.

It is important that any new proposal should receive close scrutiny and constructive criticism. I’ve addressed the short duration of immunity; but another worry is the ‘occult damage‘, or secondary effects sustained by those who are not obviously vulnerable. It is not at all unexpected that some people would suffer post-viral symptoms for extended periods of time (I believe I did!) and that it may be quite debilitating for some. Among the lessons we could learn from this crisis could be a wider recognition of the frequency and intensity of post-viral syndromes and an investment in support (leave of absence from jobs, help with daily activities) of those unfortunate enough to suffer in this way. But it is not a new phenomenon and cannot be a good enough reason to stop the world and potentially let tens of millions of people starve to death.

How such a plan may be put into place is obviously the next step to detail, and it is natural that some people might view with scepticism that it can be achieved other than in theory. Many components of protecting the vulnerable have already been enacted in the process of locking down so we should be discussing how these can be improved rather than dismissing them. Directing efforts at hospitals and care homes is one obvious priority. Other parts of the problem – such as the protection of vulnerable people within family settings – require careful discussion and thought, but it must always be borne in mind that these are temporary measures and in the long run could save more lives than cycling in and out of destructive lockdowns.

One colleague likened the GBD plan to “putting all your antiques in a room while your house is burning and fanning the flames”. A better analogy for the build up of herd immunity would rather be to douse the rest of the house with water (which could damage the Chippendale commode) — but it is hard to see how these could be the terms of a serious discussion.

Since the declaration last week, the Great Barrington Declaration has come under attack across the media, online (including Wikipedia and Google) from fellow academics as being part of a Libertarian conspiracy (my politics are not remotely libertarian) or being based in “pseudoscience”; others attempting to be less defamatory say that our views are “fringe”. The large number of serious scientists from top institutions taking part suggest otherwise. There are genuine good faith disagreements that must be aired and discussed — the impact on the world is too significant for us to fail to have this discussion in a serious way.