Second wave scepticism

A “second wave” of COVID-19 cases has been repeatedly promulgated as a pessimistic but very real possibility at Government press conferences.

A very recent letter in the British Medical Journal signed by medical Royal College Presidents and journal editors says that it is a “real risk” It has even been said that it could have more catastrophic consequences than the current epidemic and would stress the NHS to breaking point. The 13 March minutes of SAGE, the main scientific committee advising UK governments, said: “SAGE advises that it is a near certainty that countries such as China, where heavy suppression is underway, will experience a second peak once measures are relaxed”.

But I am a “second wave” sceptic. I said so in evidence to the Scottish Parliament Health and Sport Committee on 28 April and was criticised by the First Minister for saying so.

Respiratory pathogen pandemic “waves” are an influenza phenomenon and have only ever been seen with that virus. Statistics from the past have to be treated with caution, but those based on the clinical diagnosis of diphtheria and smallpox are better than most.

Past pandemics

The great diphtheria pandemic of 1855 -63 had no second wave, neither did the big epidemic of smallpox in England and Wales caused by variola minor. The virus was imported in 1919, probably from the United States. Case numbers rose steadily, with a peak of 14,753 in 1927. They then fell as steadily as they had risen, reaching zero in 1935. There was only one subsequent importation, in 1952, which led to a localised outbreak of 135 cases in Rochdale. Its origin was never established. When working on a possible antiviral drug years ago I used a strain isolated in Rochdale to test its activity on smallpox.

Pandemic planners have focused their attention on influenza. That is right, because the majority of pandemics (by definition the global spread of an infection) in recent times have been caused by that virus. “Second wave” fears all go back to 1918, when the second wave killed far more than the first.

Mortality in Scotland was typical. In July 1918 768 deaths in which influenza was a contributing cause were recorded by the Registrar-General. There were 232 in August, 240 in September, 3,714 in October, 4,272 in November, 1,569 in December, then 781 in January. 50% of deaths were in people aged between 15 and 45 (quite different from COVID-19). None of these cases had a virological diagnosis, because influenza virus was not discovered until 1933, and the possibility cannot be excluded that the first wave of cases in July was caused by a different virus. 

Mathematical modellers have not offered any explanation as to why influenza cases in pandemics occur in waves. Their textbooks are silent about it. The timing of waves in 1918 cannot be simply explained as a seasonal effect, and it has nothing to do with the development of immunity. For all practical purposes there was no social distancing, so that cannot be given any credit for declines in virus activity.

Subsequent flu pandemics have killed far fewer than 1918. The Asian flu pandemic (1957) second wave was less lethal than the first. Hong Kong flu (1968-9) was even less lethal and its second wave was less lethal in the US. The swine flu pandemic in 2009 in the UK killed ten in its first wave and 137 in its second. So even in its own right, influenza is not a good model to follow regarding the likelihood that a dangerous COVID-19 “second wave” is a realistic prospect.

Forget ‘flu

Apart from infecting a prime minister (Lloyd George got the 1918 flu), and being spread by the respiratory route, COVID-19 and influenza have little in common. The illnesses they cause are different, exemplified by the relationship between age, the incidence of complications, and mortality. Unlike influenza, children nearly always escape scot-free when infected with COVID-19 and schools don’t seem to be amplifiers of infection. The linear increase in mortality with age is COVID-19 specific.

Although care homes for the elderly have long been known as bad places to be when E.coli O157 and influenza get inside, the scale of the impact of COVID-19 on their residents has been unprecedented, as has been its impact on meat and poultry processing plants; by the end of April 115 such facilities in the US in 19 states had had cases, with 20 deaths, and plants in Wales, France, Australia, Spain, Brazil and Germany have been affected.

The current giant outbreak at the very large Tonnies plant in Rheda-Wiedenbruck is to a degree reminiscent of the situation in Singapore; both administrations were seen as coping well with COVID-19, with functioning apps for contact tracing, but both have had a setback involving migrant workers living in crowded conditions. It is also relevant that current advice to meat plant workers in the US has been issued in 22 languages including Amharic, Burmese, Dari, Farsi, Haitian Creole, Kinyarwanda, Karen, Pashto, Tigrinya, and Vietnamese. All these things demonstrate that COVID-19 and influenza epidemiologies are very different and confirm that influenza, with its “waves”, is an unsuitable model to follow.

The current COVID-19 events are not a “second wave”, or a “second peak”, or “second spikes”. They are continuations of the ongoing epidemic.  There was no second wave with SARS, a closely related coronavirus, and we are still waiting for one in Wuhan.

Scotland imports

The Scottish COVID-19 epidemic was set off by imported virus. Retrospective fingerprinting of virus genomes detected by testing between 1 March and 1 April showed that the virus entered the Scottish population through at least 113 separate travel-related introductions, mostly at the end of February and early March, and mostly from Italy, Spain and Austria. These introductions set off trains of community transmission; the first case contracted in the community occurred on 2 March and community transmission was well established by 11 March. The rest is history.

For the virus to take off again and get out of control in the way and scale that it used to be, we would have to repeat these events and have more than a hundred uncontrolled and undetected virus hot spots across Scotland, to start with. What an indictment of the failure of Test and Protect that would be! 

Further reading: Mutating coronavirus, The Conversation, June 23 2020