Hugh Pennington interview: Covid-19 is about superspreading, and there is no second wave

I was joined by Professor Hugh Pennington, an emeritus professor of bacteriology at the University of Aberdeen. He chaired inquiries into E coli outbreaks in Scotland and South Wales.

He had fascinating insights into the specific dynamics of the novel Coronavirus – its transmission is “overdispersed”, he argues, rather than homogeneous. It also depends on superspreaders.

The government’s thinking on “second waves” is based on flu modelling and is a poor way to understand the virus, he told me.

Alastair Benn: What are the special characteristics of SARS-CoV-2?

Hugh Pennington: It’s one of the characteristics of this virus that it causes big outbreaks in certain areas. Meatpacking plants, for example, in Europe and the US, or student halls, where people are in close quarters in large numbers. It’s a slightly curious phenomenon with this virus. We don’t see that to anywhere near the same degree with Influenza. We do have institutional outbreaks but it’s much more schools than universities. Children are very good amplifiers of Influenza but they’re not, fortunately, good amplifiers of this virus.

They don’t spread it easily to each other. They can catch it and transmit but they’re much poorer at transmitting the virus. They are also much less likely to have serious illness than with flu. A very small number have serious complications but this isn’t likely to be a direct effect of the virus – this is a delayed complication. It is very rare.

The one outstanding property of this virus which makes it unique among viruses and bacteria is the very strong relationship between age and having a hard time. The graph is a straight line – the older you are, the more likely you are to die of this. In most viruses and many bacteria, elderly people do have a harder time. That’s been the case for flu and we’ve known that for many years. Every so often flu gets into a care home and there is very high mortality. But flu is different because it has quite serious effects for young people. It affects children too.

Other respiratory viruses behave like that too. Pneumococcal pneumonia, for example. There is a vaccine against it. It protects but it doesn’t get rid of the bug. It is a tricky vaccine to make. That is a killer of old people. The 19^th century doctor William Ogle called it “the old man’s friend” because it carries you off quite quickly when you’re elderly. But most of the nasty infections are found in infants. That pattern is much more characteristic of these respiratory viruses and bacteria.

It’s easy to explain why the elderly are more likely to have a severe illness. They have other things wrong with them. Their immune systems are fragile. However, there is no simple explanation to why children are seemingly resistant to having any particular serious illness. It may be something to do with the distribution of the receptors to which the virus binds.

AB: Is it comparable to other coronaviruses that circulate in the human population?

HP: We know quite a lot about its basic virology because it is a coronavirus and we know quite a lot about coronaviruses in general. It has certain similarities with SARS which also appeared in China.

Like SARS, Covid-19 shows what it is called overdispersion. If you look at the transmission of the virus, most cases do not transmit to other people. A minority of cases transmit. We reckon that 10% of people infected account for 80% of secondary cases. It was the same in SARS. There were some people who were superspreaders and infected an enormous number of people. Most people who had SARS didn’t infect anybody at all. In a sense, this is good news – if you can control that aspect of the virus’s behaviour you can deal with it much more effectively than if it spread in an even sort of way, if every case was equally infectious.

This phenomenon of overdispersion was also true of Smallpox. How did we eradicate Smallpox? It wasn’t just a vaccine which didn’t always work. In the UK, we managed to eradicate Smallpox when we didn’t have the required level of herd immunity. It was contact tracing that played an equally important role in getting rid of smallpox. Finding cases and quarantining them and their contacts led to outbreaks fizzling out.

AB: What hopes do you have for a Coronavirus vaccine?

There will be a vaccine and probably more than one. There are so many in development. Some of them do seem to be stimulating an immune response that is satisfactory. Having said that, the likeliest sort of outcome will be a vaccine like we have for flu. It is far better than nothing but it only protects some of the people for some of the time. For many people who get the flu vaccine, it won’t stop them getting it but it will reduce the severity of the illness and perhaps save their lives.

Although we have an active flu vaccination programme every year we still have people dying of flu in care homes. It protects some of the people some of the time but it doesn’t get rid of the virus. It will continue to circulate.

A very optimistic outcome would be a vaccine that offers protection like the MMR vaccine. This gives good strong immunity for many years. Even then, remember that there was a virus circulating in universities just before we went into lockdown. It was being driven by students. This was mumps which is covered by the MMR vaccine. Their immunity had waned. Be careful when you say “the vaccine will solve all our problems” – this is very unlikely.

AB: Covid-related deaths went up very fast in March and April. Are we likely to see a peak like that again?

HP: It will be different. What happened in late February, early March, we had a lot of importations of the virus into the UK. We made the mistake of thinking that the virus would come from Wuhan – but hardly any of it came from Wuhan. It came from Italy, Spain and France. We know from doing RNA sequencing of the virus that we had at least 1,300 separate importations into the UK at that time. Some of these died out because of the overdispersion phenomenon I talked about. But a lot of it was seeded throughout the country.

Workers often without symptoms took it into care homes where there was very high mortality. That pattern was replicated across Europe, including in Italy, Spain, France and Sweden. In other countries the mortality wasn’t nearly as bad. Perhaps, they have better protections in care homes or there are fewer elderly folk. It is unclear and international comparisons are difficult to make.

I don’t think that will be replicated. We are much better in controlling importations of the virus so we aren’t setting up local chains of transmission all the time. We know where the problems are at the moment – in the north of England, in the Central Belt in Scotland and in Belfast. The virus is now far more localised than it was in March and April.

AB: Are national measures too blunt an instrument?

HP: The way the policy is generally presented is through national figures. The national figures don’t take account of local variation. And there is an issue here about presentation to the public. The role of the press conferences has been to generate fear. It is always the bad figures that are presented. The virus doesn’t behave in the way presented in Vallance’s exponential graph. It doesn’t grow exponentially like that. It might do that in a lab if you put it into a bottle of tissue culture.

In bacteriology, you plant the growth and it grows exponentially until it runs out of food. Viruses don’t even behave like that. It comes back to that dispersion point I made – not every case is going to transmit to another case. You might well get local situations where you get exponential growth. You are not going to have it affecting the population as a whole.

In a care home, the R number is probably 10. For the whole population, the R number may be below one. It is such a crude average. It is pretty useless as a measure of what is happening except in local circumstances. It is used as a stick to hit us round the head with. It has the ring of science about it so people say “it must be true”. It’s useful in an epidemiology textbook. It is not useful in public health measures.

AB: “Wave” is a poor metaphor for Coronavirus transmission then.

HP: The mathematical modellers, Neil Ferguson for example, are experts in modelling flu. One thing they haven’t been able to explain in their modelling is that flu appears in waves. It happened in the Spanish Flu in 1918. It arrived in the summer, went away and returned in the winter and returned the following year. In the second wave, it killed young people. In the third wave, it killed old people.

Nobody has been able to explain why that was the case so it’s a very poor comparison. It’s also unclear what people were actually dying of. It may have been other bugs coming in once the lungs had been damaged by the flu. This is where the idea of a second wave being nastier than the first comes from. This was much talked about earlier this year. This is not flu. This virus behaves quite differently.

A second wave in flu would be helped by schools. In the summer, it would circulate quite quietly. It would be harder to find serious cases. But then schools open and away it goes. That’s what happened with the Asian flu in the seventies. But in this case children are not amplifiers of the virus. Using flu as a model is a very poor model. The virus, in most of its properties, is very different. It affects different people. It affects different age groups.

AB: So the science was going in the wrong direction early on.

HP: I think many of the scientists who advised the government early on were very much wedded to the flu model because that’s the one we have in the ring binders as it were.

I’m a second wave sceptic because we are actually still in the first wave. We suppressed it through lockdown measures. We have released them and it is spreading more freely now. As far as we can tell it hasn’t mutated significantly. Again, unlike flu which has a much higher mutation rate.  Where are the virologists? There are hardly any of us left. It’s rather sad. Who is on SAGE? There were four virologists on SAGE out of more than fifty members. Three of them were flu experts.

We need to talk about superspreaders and the overdispersion factor, for example, which are characteristic properties of this virus. If you look at interviews of experts on Covid-19, hardly any are virologists. They are epidemiologists or public health experts. Virology has lost out. Perhaps, we were too successful for our own good. As a field, we had many triumphs in the 20^th century including the polio vaccine and treatment for HIV.

Pandemics are a virus speciality. There will be more pandemics. I don’t know what will cause it or when it will happen. All you can see is that there will be one. We have got to be ready for it.