The coronavirus pandemic: can we handle such epidemics better?

Heeding previous knowledge and doing research

What I missed the most in the early months of the pandemic were that the authorities did not heed sufficiently the knowledge we already had, and that researchers did not embark on experiments that could tell us what works, what doesn't and what is harmful.

Ground-breaking studies by Danish professor Peter Aaby have taught us that the infectious dose of a virus is very important for whether we will survive or die. ² If it is too high, the immune system will not have enough time to mount a defence. This is an important reason why so many people died in overcrowded hospitals in northern Italy and in China, where secret recordings showed dead people lying on the floor in a hospital where the staff was so busy that they didn't have time to remove the corpses while the only protection they had was face masks.

In South Korea, only severely ill patients were admitted to hospital, and they used aggressive early contact tracing and extensive laboratory testing. ³ Only four people per million inhabitants have died. ⁴ Countries that ignored the danger for far too long have had many more deaths, e.g. 358 and 193 per million in Italy and the UK, respectively (it is too early in the epidemic to say what will happen in USA that also ignored the danger, but deaths are rapidly increasing; currently, there are 87 deaths per million). ⁴ Only recently did the UK and USA introduce a lockdown, but with insufficient capability to test people, to trace contacts and to isolate them. I have not seen any country formulating a clear exit strategy from the lockdown.

Aaby's studies have taught us another important lesson. Virtually all countries have advised infected people to stay in their homes so that they do not infect people in the community. However, when the index person – the one who gets infected in the community – is ordered to stay at home, secondarily infected people in the household will have a considerably higher risk of dying. This is because the infectious dose is much higher when people live closely together. Aaby has shown that the mortality for secondary cases of measles in the home in Guinea-Bissau was 3–4 times higher than for the index case.

If the authorities had heeded this knowledge, they should not have sent infected people home but isolated them in quarantine centres, e.g. sports arenas and conference halls, till they were no longer infectious. This appears to have been the strategy only in China.

I miss randomised studies of school closure. If children are sent home to be looked after by their grandparents because their parents are at work, it could bode disaster for the grandparents because the median age of those who die is around 80. ¹

In some countries, people are only allowed to leave their house for shopping, walking their dog or going to a pharmacy. We don't know if these measures make more people survive during an epidemic, but we do know that they cause a lot of harm for whole populations.

We don't know either if face masks help when used outside hospitals. The panic has caused the public to purchase so many face masks that it caused a shortage at hospitals, which is where they are needed. Commercial vendors took advantage of the humanitarian crisis: prices of face masks increased six-fold, prices of respirators more than tripled and prices of gowns doubled. ⁵ Supplies could take months to deliver, market manipulation was widespread and stocks were often sold to the highest bidder.

Four months into the pandemic, it was finally announced that a randomised trial of face masks would be carried out in Denmark. A trial has started in Holland where health personnel are vaccinated against tuberculosis because this vaccine has positive effects also against other infections. ² The polio vaccine is another interesting candidate. ² Furthermore, the effect of various drugs on the virus are being tested.

Despite the wide range of social distancing policies adopted in various countries, the decline in the number of deaths has been remarkably similar. ⁶ It seems that, as long as social distancing is implemented, what remains is an inherent element of the epidemic's dynamic that is the same everywhere.

In contrast to Denmark, Sweden has remained a pretty open society that did not close down society, and it took till mid-March before they cancelled events with 500 or more participants, while Denmark, also in mid-March, made all gatherings of more than 10 people illegal, warning that law offenders would be fined by the police.

In mid-April, deaths associated with coronavirus had mounted to 119 per million inhabitants in Sweden and 53 per million in Denmark. ⁴ However, this cannot be taken as evidence that the draconian measures in Denmark work. There could be many other explanations, and differences of this size in an ecological comparison of two countries do not impress researchers. Furthermore, Sweden must have a much better herd immunity than Denmark because the virus was allowed to spread, which means that Swedes are better protected than Danes when the next wave of coronavirus arrives. We therefore need to wait a couple of years before we can compare Sweden and Denmark.

Is corona more contagious and deadly than influenza?

A key argument for the draconian measures has been that corona is much more contagious than influenza, but the transmission rates seem to be similar. The infection rate was only 19% on the Diamond Princess cruise ship where an entire, closed population of quarantined passengers and crew was tested for the virus, ⁷ even though people had crowded in bars, at buffets and at the dance floor. WHO's Director-General, Tedros Adhanom Ghebreyesus, has even announced, on 3 March, that COVID-19 was less contagious than influenza, ⁵ but five weeks later, he said that COVID-19 spreads fast and is 10 times as deadly as the swine flu.

It is puzzling why so many people have died because the best evidence we have indicates that the death risk is similar to that for influenza. But we know from previous virus epidemics that they can be particularly deadly when they hit a nonimmune population, e.g. in the measles epidemic in the Faroe Islands in 1846, 78% were attacked and the case fatality rate was 2.8%.⁸ As the risk of dying increases in settings with overcrowding, we can only estimate death rates approximately. In outbreaks of measles, a commonly used estimate is 0.2%, but it can be many times higher than this. ² In Denmark, we recently estimated the level of infection by testing blood donors for coronavirus antibodies and found a death rate of only 0.16%, ⁹ the same as for measles.

The case fatality rate on the Diamond Princess cruise ship was 1.0%,^3,7 in a largely elderly population in a crowded environment. Projecting the mortality rate onto the age structure of the US population, the death rate would be 0.125%. ³ This estimate was preliminary and based on only seven deaths but is similar to the Danish estimate.

On 3 March, the WHO Director-General stated that about 3.4% of reported COVID-19 cases have died and that seasonal flu kills far fewer than 1%. ⁵ The WHO mortality estimate for COVID-19 was likely exaggerated by a factor of more than 10 because the number of infected people was vastly underestimated. The result of this highly misleading announcement was to cause horror and even more panic than there already was. ³

For influenza, a systematic review reported huge variations in reported case fatality rates. ¹⁰ Based on a figure in the paper, I estimated a median rate of about 1% for laboratory confirmed influenza during the mild influenza pandemic in 2009 and the following years, but this is likely also exaggerated.

In Italy, 99% of those who died had at least one co-morbidity, half of them had three or more diseases, and the median age was 80 years. ¹ These are also the type of people who die during influenza epidemics. Italy has been hit very hard, and many factors have contributed. In Lombardy, the hospitals were overcrowded; they admitted even mild cases; the staff did not have much protection initially; there was a lack of personal hygiene; the population is older, on average, than in most other places and smoke more; and Italians have a tradition of generations being close together, and for hugging and kissing.

When comparing deaths in various countries, it is a further difficulty that there are several genetic variants. We would also expect additional mutations during the pandemic.

Should we do everything possible?

An often-heard argument in favour of draconian measures is that we should do our utmost to limit the number of deaths, as we should not set a price on a human life. But there is an economic and a societal limit as to what we can do. If we have no limits, we could use the entire gross national product on helping people survive from all sorts of diseases. We could avoid almost all traffic deaths if we lowered the speed limit for all vehicles to walking speed.

We could avoid millions of deaths from malaria, tuberculosis and other infections; other millions if we had effective drug regulation, as most of those who are killed by prescription drugs, didn't need them ¹¹ ; and yet other millions if we made tobacco illegal. Yet, we don't do this.

As we shall all die, we cannot save lives, only extend lives. If an 80-year-old person does not die from coronavirus, we might have extended life by about eight years, but with a quality of life below 100% because of the co-morbidity. ¹²

The price we pay to extend lives during the coronavirus pandemic exceeds by far anything we will usually accept. The UK National Institute for Health and Care Excellence will recommend funding medical interventions if they cost less than £30,000 per quality-adjusted life year. If 20,000 deaths are prevented in the UK, the cost per quality-adjusted life year is around £7 million. ¹² The UK government has used an estimated 250,000 for the upper limit of deaths if nothing is done to prevent spread, and if all these deaths are prevented (both assumptions are totally unrealistic in my view), the cost per quality-adjusted life year is around £400,000. ¹²

There seems to be something wrong with our priorities. There is also an opportunity cost. Many people with other diseases will not get the attention and treatment they need, which will increase suffering and deaths. Some elderly people with lack of support may die of dehydration and starvation, and after India introduced a lockdown, migrant labourers feared that hunger would kill them before the coronavirus did.

In USA, a huge number of patients with heart attacks and ischaemic strokes have not turned up at hospitals, likely because they are afraid of getting infected with coronavirus. ¹³ Since the chance of survival for both conditions is closely related to how fast you get treated with thrombolytics, the death toll could be greater than that directly caused by the virus. ¹³

On a much smaller scale but still relevant, when businesses go bankrupt and unemployment levels increase, suicides increase too. The panic has caused suicides among the staff and in an isolated patient, and a person in London died from malaria because he couldn't get through on the phone. During periods of crisis, infant injuries and deaths from abusive head trauma increase. ¹⁴

Is the mass panic justified?

It moves us all deeply when many people die from an epidemic in a short time. Even so, it is relevant to ask: Since there seems to be no major differences between COVID-19 and influenza in terms of infectivity and case fatality rates, why were draconian measures applied now and not during the 2009 influenza pandemic? It is always winter somewhere, and we cannot lock down the whole world more or less permanently. And even so-called mild or common cold type coronaviruses can have case fatality rates as high as 8% when they infect elderly people in nursing homes. ¹⁵

Part of the explanation is that no one will get in trouble for measures that are too draconian, only if it can be argued they did too little. Concern and anxiety may drive events more than the disease itself. ¹⁶ In 2005, the WHO stated that 2–7 million people would die from the swine flu, which they called a conservative estimate. ¹⁶ The UK government wasn't modest either; it predicted that as many as 65,000 citizens might die during the 2009 swine flu pandemic but fewer than 500 died. ¹⁷

International politics and Chinese censorship

Chinese doctor Li Wenliang warned about what looked like a new SARS-like illness in Wuhan in early December 2019 in a group chat with other health professionals. ¹⁸ A few days later, he was detained by the police for ‘spreading false rumours’ and they forced him to sign a document admitting that he had ‘seriously disrupted social order’ and breached the law. At least seven other people were similarly disciplined.

Li acquired the disease himself and died in early February. His treatment by Chinese authorities sparked outrage across China, and in March, an official report declared that Li had not disrupted public order, and that he was a professional who fought bravely and made sacrifices. ¹⁸ However, the Chinese rulers put the blame on the local police, which added fuel to the anger; on China's Twitter-like Weibo, news of the report had over 160 million views.

At the end of December, Taiwan alerted the WHO to the risk of human-to-human transmission of the new virus, but the WHO did not pass on the concern to other countries. ¹⁹ China had ensured that Taiwan is not a member of the WHO, and WHO's cosy relationship with China has been criticised, particularly when the WHO overly praised China's handling of the coronavirus outbreak despite the fact that China initially covered it up. ¹⁸

China's health ministry only confirmed human-to-human transmission on 20 January, after the WHO said in mid-January that there might be ‘limited’ human-to-human transmission, but it stepped back from this view on the same day.

International politics and censorship have likely increased the death toll substantially. It took almost two months before people reacted after they could have reacted based on the warnings from the Chinese whistleblower in early December, and a few weeks later from Taiwan.

Future directives

Since there is not much herd immunity, the pandemic will likely come back in another wave. It should be considered to close the open animal markets in Southeast Asia to reduce the risk of future virus epidemics.

Scientists will need to be involved right from the start so that we may gather important evidence, e.g. by repeated and widespread testing of random samples of the population and by performing randomised trials. Not only have few of the current measures been evidence-based, we have also not had enlightened public discussions. The politicians have ensured that it will be very difficult to analyse afterwards if the measures did more good than harm. This is not how healthcare should be.

Many companies and public institutions are working on developing a vaccine, which will undoubtedly be fast-tracked through drug agencies and perhaps approved based on surrogate markers, e.g. an increase in antibodies against the virus. When a vaccine comes into usage, we might therefore not know if it has any important effects. We should not be too optimistic about what a corona vaccine might accomplish if we compare with the influenza vaccines, which do not have documented effects on transmission, hospital admission, serious disease or death. ² We will know very little about the harms of the vaccine, and nothing about the rare but serious harms. An additional problem is that, if we let the drug industry develop a vaccine, it will almost certainly be sold at a price that many countries cannot afford.

Knowledge accumulates all the time, and the Centre for Evidence-Based Medicine in Oxford has a very helpful website with lots of analyses and reviews that are regularly updated. ²⁰

Some people have suggested that, rather than, or in addition to, isolating those known or suspected to be infected, we could focus in particular on social distancing and in some cases even isolation of the most vulnerable, and to be particularly careful when visiting and helping old people, while the virus is allowed to spread more freely and create better herd immunity, without locking down whole nations. It has been suggested that this might avoid more deaths than the lockdowns,^12,13 which we, furthermore, cannot afford to apply every time there is an epidemic.

However, the surprisingly low number of deaths per million in South Korea, which used aggressive early contact tracing, extensive laboratory testing, isolation of cases and avoided overcrowding of hospitals by allowing admission of only the most serious cases, suggests that this approach is the best one.