Monday, 6 April 2020

Coronavirus: were science and statistics overruled by fear?

Niall McCrae
King's College London

Roger Watson
University of Hull


A crucial lesson for students in science and statistics is that correlation does not necessarily equate to causation (Thapa, Visentin, Hunt, Watson & Cleary, 2020). To consider a relationship between two variables as causative could lead to absurd conclusions. For example, the extent of damage caused by a fire correlates with the number of attending fire engines and firefighters. Obviously, it would be ridiculous to interpret this as evidence that less intervention by the fire brigade would lead to less destruction, and to act on such flawed logic accordingly. Clearly a large building in flames needs more hands to the pump.
The coronavirus crisis has exposed basic errors of numeracy and reasoning not only in the scientifically and statistically naive general public but also among experts. The impact of drastic decisions, made on dubious rationale, may endure for years if not decades. Global and national economies and health services may be harmed not so much by the virus itself, but from reaction to the virus. Here, we present some widely perpetuated and persuasive myths about the coronavirus outbreak.
1. Covid-19 is an unprecedented pandemic
Covid-19 has frequently been compared to the Spanish influenza outbreak of 1918-1919. But the death toll from Covid-19, just over 252 thousand at the time of writing, is 198 times smaller than the 50 million toll of Spanish flu. Mortality appears to be similar to that of seasonal flu epidemics, at around 0.1 to 0.3% (unlike the 3.4% rate suggested by the WHO in March). Although the daily and cumulative death totals are alarming, in 2014 almost 29000 died in England from a winter flu outbreak (Office for National Statistics, 2015), like the number dying from Covid-19 but with minimal public or media attention.
Misleadingly high death rates were reported in the media, based on a dubious denominator of cases, which depended on the amount of testing. The numerator of deaths may also be inconsistently measured; dying with the coronavirus is not necessarily the same as dying of the coronavirus. While in some countries cause of death was recorded as Covid-19 only if the person had been tested and died primarily from this disease, in other countries deaths were attributed to the virus without firm evidence. In the USA, hospitals receive higher payments from the government for Covid-19 deaths (Creitz, 2020).
2. The whole population needs protection
The Spanish flu targeted fit and healthy younger people; American troop ships arriving in France in 1918 were like floating coffins (Hasley, 2018). Covid-19, by contrast, mostly spares children, who are hardly affected by the disease and do not appear to transmit it. Most fit adults do not get seriously ill from Covid-19. Data from Iceland showed that 50% of cases were asymptomatic (Tara, 2020).
The profile of fatalities from Covid-19 is consistent: most are frail, old or have a major chronic condition such as diabetes mellitus (type I) or cardiovascular disease. Such people are more likely to contract an infectious disease that has become endemic in society, because their immune system is compromised. The death rate of those aged over 80 and 70 are estimated to be ten times higher and five times higher than the mean, respectively (Cuffe, 2020).  To put the numbers into perspective, consider a town with a population of 10000. On a national rate of deaths per million of 500, this town would have had five deaths (perhaps three in hospital, two in care homes). Most inhabitants would not know anyone personally who had perished. 
Clearly it was sensible to protect vulnerable people from the virus, by minimising contact with others while the virus was spreading. However, most European governments went far beyond protecting the weakest members of society.
3. Lockdown saves lives
The evidence for the lockdowns imposed in European countries and states of the USA is contentious. The rationale was to reduce the rate of transmission per person (R). But, as well as unavoidable transmission within households, lockdown probably increased multiple occupancy, as young people returned to the family home after being unable to study and work. In Italy, hundreds of thousands of people fled from northern cities to other parts of the country prior to the declared lockdown, unwittingly taking the disease with them. The only way to operate a lockdown effectively would be to isolate each person, which would be practically impossible.
In the UK, mortality reached a peak five weeks after lockdown was imposed. According to pathologist John Lee (2 May 2020), mass immobilisation of the people reverses the natural progress of a virus towards milder symptoms. The evolutionary purpose of a virus is to survive. A virus spreads wider if it infects active people with mild symptoms.
Lockdown has disrupted the normal operation of healthcare systems, with planned surgery postponed and screening for serious diseases such as cancer reduced or stopped. This will cause major collateral damage. In the UK, the slogan ‘Protect the NHS’ (National Health Service) deterred people from going to hospital when sick, and this combined with fear of infection resulted in emptying of wards. A rise in suicide has been predicted, particularly in socially isolated people with severe mental health problems (Hammond, 2020). Lockdown has increased domestic violence (Home Office, 2020). Furthermore, shutting down most businesses will have devastating effects on the wealth-generating parts of the economy, on which health service funding depends. 
Was this a necessary evil? The country receiving most attention for keeping its schools and businesses open is Sweden (Moore 2020), but Japan also maintained life as normal. To justify taking this risk, these countries would need to show mortality rates not substantially higher than in lockdown countries. In fact, they have had lower mortality, as shown in Table 1.

Table 1: Covid-19 mortality in selected countries (Worldometer, accessed 3 May 2020)

Deaths per million
Lockdown (most states)
No lockdown
No lockdown

Sophisticated analysis by Thomas Meunier (2020) indicates that the draconian European lockdowns may not have saved any lives. Comparing the UK, Spain, Italy and France with other countries with less restrictive interventions, there was no indication of any difference in the temporal pattern of the epidemic.  Some smaller countries have been able to contain the outbreak, particularly islands such as Iceland and New Zealand. But countries with high population density and cross-border transit have fared badly, as the virus quickly became endemic. Lockdown was like shutting the stable door after the horse had bolted. 
4. Testing is the priority
The World Health Organisation urged as much testing as possible. However, a major obstacle was the global shortage of testing kits (Smith, Wright, & Moody, 2020). The high frequency of tests in Germany was useful in showing the spread of Covid-19, but not necessarily in slowing it. In the UK, special testing centres were created for National Health Service (NHS) staff, following political and public pressure. However, there was confusion between diagnostic testing and antibody testing (ITV News, 2020). The former simply shows whether the person has the disease at the time of testing. An antibody test can show whether someone has had the disease, but antibodies take approximately four weeks to be detectable. Sending busy NHS practitioners to a testing centre would be almost futile. A negative test result would be no guarantee that the doctor or nurse does not get infected soon after. An antibody test would be more valuable to the health services and to society as this would show that people have had the virus and are consequently immune and not contagious and can work safely. The same principle applies to hepatitis B testing, which has been administered to all clinicians since the 1980s.
5. The more advanced the healthcare system, the more lives will be saved
This is not supported by evidence. Western European countries, with well-equipped intensive care facilities, have the highest mortality. As a ‘free at the point of delivery’ service, the British NHS may have been expected to save more lives than in countries with mixed public/private systems. However, despite its nationally cherished status, evidence indicates that the NHS was unable to prevent coronavirus deaths any more effectively than in countries with predominantly private insurance schemes. Ventilators and other modern intensive care apparatus are vital, but there is no cure for Covid-19.
While the NHS has been applauded for their efforts with a weekly public tribute, it may have contributed to the death toll. Most care in Britain is provided outside the NHS, particularly in residential homes for older people, where as many as half of Covid-19 deaths have occurred.
Discharge guideline was issued by the Department of Health on 17 March 2020 'to free up NHS capacity via rapid discharge into the community and reducing planned care'. An anonymous whistleblower involved in the care home sector claimed that many older patients were discharged from hospital before receiving vital treatment, and without being tested for Covid-19 (Richardson, 2020). Another whistleblower reported that in a care home in Edinburgh residents known to be infected with Covid-19 were allowed to wander freely (Richardson, 2020). The care home had nine deaths in two weeks and a further ten residents tested positively for Covid-19.
Myths are misleading and they can be dangerous. Lockdown does not appear to have rescued people from the virus, and its economic impact will be devastating to public finances and individual livelihood. The clamour for something to be done, and the use of information which is not necessarily evidence-based, may have misguided governments. We appreciate that measures taken to ‘flatten the curve’ of the pandemic were taken to avoid overwhelming hospitals with finite resources (although the hurriedly built Nightingale Hospitals around the country have mostly remained empty). By whatever means this is done we know that we cannot live safely with this virus in our midst until approximately 80% of the population have been exposed and developed the necessary antibodies (Sadarangani, 2016) and all we can do is control the rate at which this is achieved. Sceptics of government policies have been demonised in the mainstream media, but clearly there are major flaws in the rationale and practice of disease containment. Truth and accuracy should not be victims of the Covid-19 pandemic; we will need these precious commodities more than ever as we strive to return to normal life. A key question is: if we have Covid-21, would we do the same again?
BBC News (2020). Coronavirus: 'We do not recommend face masks for general wearing'. BBC News. ( (Published 3 April 2020; accessed 4 April 2020).
Cuffe, R. (2020). Coronavirus death rate: What are the chances of dying? BBC News. (; Published 24 March 2020; accessed 4 April 2020).
Devlin, H. (2020). Guardian (3 April 2020): Can a face mask protect me from coronavirus? Covid-19 myths busted. (; Published 3 April 2020; accessed 4 April 2020).
Creitz, C. (2020). Minnesota doctor blasts ‘ridiculous’ CDC coronavirus death count guidelines.; Fox News. Published 9 April 2020; accessed 4 May 2020.
Hammond, C. (2020). Coronavirus mental health warning: expert reveals key demographic most at risk. Daily Express. (; Published 26 March 2020; accessed 4 April 2020).
Hasley, A. (2018). A killer flu was raging. But in 1918, U.S. officials ignored the crisis to fight a war.; 3 February2018; accessed 4 May 2020.
Home Office. (2020). Coronavirus (COVID-19) and domestic abuse. GOV.UK (; Published 29 March 2020; accessed 4 April 2020).
ITV News. (2020). 'What's important is social distancing' coronavirus testing 'is a side issue', says Deputy Chief Medical Officer ITV.; Published 1 April 2020; accessed 4 April 2020.
Kington, T., Wilkinson, I., Bremner, C. & Sage, A. (2020) Italians singing from different hymn sheets as coronavirus fatigue sets in. The Times. (; Published 4 April 2020; accessed 4 April 2020).
Lee, J. (2 May 2020) Covid’s metamorphoses: is the lockdown causing more harm than good? Spectator.
Meunier, T.A.J. (2020): Full lockdown policies in Western Europe countries have no evident impacts on the COVID-129 epidemic. MedRxiv
Moore, T. (2020) Coronavirus: Sweden determined it has smarter strategy for dealing with COVID-19. Sky News (; Published 4 April 2020; accessed 4 April 2020).
Office for National Statistics. (2015). Highest number of excess winter deaths since 199/2000. Office for National Statistics. (; Published 25 November 2015; accessed 4 April 2020).
Richardson, H. (2020): Whistleblower claims hospitals are still deliberately sending elderly patients to care homes 'to die' by discharging them before Covid-19 test results come through - as part of 'soul destroying' culture of freeing up beds. Daily Mail. Published 1 May 2020; Accessed 4 May 2020.
Sadarangani, M. (2016). Herd immunity: how does it work? Oxford Vaccine Group. (; Published 26 April 2016; accessed 4 April 2020).
Smith, C., Wright, O. & Moody. O. (2020) Global shortage of coronavirus testing kits threatens expansion plan. The Times.; Published 31 March 2020; accessed 4 April 2020).
Spiegelhalter, D. (30 April 2020): Coronavirus deaths: how does Britain compare with other countries? Guardian.
Tara, J. (2020). Iceland lab's testing suggests 50% of coronavirus cases have no symptoms. CNN. (; Published 3 April 2020; accessed 4 April 2020).
Thapa, D.K., Visentin, D.C., Hunt, G.E., Watson, R. and Cleary, M. (2020). Being honest with causal language in writing for publication. Journal of Advanced Nursing doi:10.1111/jan.14311.
Hasley, A. (2018): The flu can kill millions: in 1918, a pandemic was fueled by World War I.; 3 February2018; accessed 4 May 2020
Worldometer (accessed 3 May 2020): Coronavirus.; accessed 3 May 2020.
Disclaimer: Editorials are opinion pieces. This piece has not been subject to peer review and the opinions expressed are those of the authors. RW is Editor-in-Chief of JAN. Neither NM or RW have relevant political or other affiliations to declare.

No comments:

Post a comment