Understanding and planning for the pandemic

We are being bombarded with snippets of information about COVID-19, and short term government restrictions, but little information about what to expect through the year 2020. As a CEO, it's my job to plan ahead. Indeed, this week we made the sad and painful decision to part from about 20% or our colleagues and friends as the only way to make our cash last through the upcoming recession.

Our families feel the same need to understand and prepare for what's coming.

At first I thought that the UK, where I was living part-time before the pandemic, had found a better approach, but after studying it, the data doesn’t support that. Part of the reason for Freightos preparing for the worst this week is that it seems that no matter what strategy our countries adopt (and we are in several countries), we all face a few months of home isolation, and even longer restrictions on international travel, while a recession deepens around us. This is going to be a painful year, but if we know what is coming and prepare for it, we can make it through as businesses and as families. And as I say in the conclusion there is actually some very good news this morning. Here is my survey.

___________________

The first potential COVID-19 vaccine has started tests but experts agree it will take 12-18 months to complete tests, production and distribution of a vaccine. There are no signs of a wonder treatment. That places the end of the pandemic some time in 2021.

What strategy should a country pursue in the meantime? 

Do nothing

If a country does nothing the virus will spread exponentially. Each person will infect between 1.4 and 3.8 people depending on population density, weather and social habits. This number is known as R0. 

The number of infected people will double every 6.4 days on average, although again that might vary. In a small country with ten million people this rate implies going from 1 patient to the whole population in 22 weeks, for a billion people, 28 weeks. Unchecked, exponential growth can take the virus from patient zero (probably in November 2019) to the whole world population in about 30 weeks.

Will everyone get infected? There is no evidence that anyone has natural immunity. But as more people become infected and then immune, the spread will slow and eventually fizzle out. Depending on the actual R0, the population will achieve a herd immunity threshold (HIT) once 29–74% of the population are immune. An average estimate is therefore that around 50% of people will get sick before the epidemic stops spreading. No one knows for certain how long immunity lasts, but based on other viruses, if COVID-19 doesn’t evolve too rapidly, there is a good chance natural immunity will last at least for a year or two till vaccination is available.

The World Health Organization reports that 3.4% of diagnosed patients die. Fortunately, that is likely to be an overestimate of the mortality rate, because not all infections are tested or reported, while almost all fatalities are reported. In Korea, which has done the most aggressive testing of 4,800 per million, there is a mortality of just 0.6% which seems to match the best available estimate, and even that may be an overestimate since even South Korea cannot identify every infection, some of which are asymptomatic. Mind you there is also a risk that high incidence in a young mostly female non-smoking Korean church or cult skewed the Korean mortality rate downwards, so there is uncertainty in both directions.

However 0.6%-3.4% is the death rate only so long as hospitals are coping. In the do-nothing scenario, hospitals will very quickly be overwhelmed. In China, about 15% of those infected required hospitalization and 5% ended up in critical care. Again these are likely to be overestimates since many infections are not detected. The best estimate seems to be that 5% will require a hospital. In Korea 1% of cases are reported as serious or critical. There is no data for the mortality rate when no hospital care is available, it is probably in the 1-5% range.

In summary, in the do-nothing scenario we have an epidemic lasting 20+ weeks until 29-74% of the population are infected at which point it fizzles out, leaving anything from 1-10% of the population dead, the vast majority of whom never had a hospital bed. Therefore no country is pursuing this strategy.

Mitigate

Mitigation is a strategy which accepts that the virus will spread through the population, but attempts to reduce the impact by dampening the exponential growth, spreading the infection over time, and flattening any peaks. Models show that measures of social distancing can effectively slow the exponential spread of the virus. Is it possible to slow the spread of the virus to the point where hospitals can cope? 

This has apparently been the UK’s strategy at least until this week. Is it feasible?

Developed countries have about 2,000-7,000 hospital beds for acute care per million people, but mostly countries have 2,000-3,000. They have only 60-340 critical care beds per million. If we take an average estimate that the herd immunity threshold is 500,000 per million people, and based on the data that 5% require an ICU, 25,000 people per million will need a hospital over time. The average hospitalization time in China was 12.8 days. So a bed can treat just over 2 people per month, and in aggregate ICUs in most developed countries can treat 120-680 people per million population per month.

Treating 25,000 people per million in ICUs even in the US, which has the most critical care beds per capita, would take 3 years. In the UK it would take 17 years. That’s if all the beds were empty. However those beds are not empty. In the US, 68.2% of ICU beds are full on average. In Italy which has fewer beds 85% to 90% are occupied in the winter. 

Now the 5% number may be high. Comparing mortality rates in China (~3%) to Korea (5X lower at ~0.6%) suggests that perhaps 80% of cases are undetected in China (although other explanations for the discrepancy are possible such as the age group in Korea) so it could be that as few as 1% of people need an ICU. Perhaps treatments will improve. But even in the most optimistic scenario, and even with perfect throttling of the disease’s spread, it does not seem feasible that any country could allow herd immunity to emerge in the next year while providing an ICU to everyone who needs it.

Indeed recent modeling at Imperial College has shown that no amount of social distance mitigation will avoid the hospitals being overwhelmed:

But in fact there is no need to model the specifics of how the epidemic unfolds, it’s a matter of arithmetic that whatever the growth curve of the epidemic, there are not enough ICU beds to let it play out in the next year.

Mitigate and ventilate

A possible scenario is to mitigate the growth, while also increasing the number of critical care beds and especially ventilators. According to the Imperial College study, the total number of hospital beds equipped with ventilators would have to be increased at least 8X in most countries, within the first couple of months of the epidemic spread. Staff would have to be trained very rapidly to operate the equipment. This seems like an extreme but feasible scenario. China built new hospitals in 10 days. While this will be difficult, with a highly focused war-like effort, it does not seem impossible that ventilator capacity could be multiplied in weeks. Ventilators are not particularly expensive or hard to operate. They are usually found in ICUs, but it could be feasible to put one by every hospital bed and more still in makeshift hospitals or homes. Given that this is a global emergency, more doctors as well as nurses, medics and lay people could be trained to operate them.

Till this week it was surprising that no country had announced to my knowledge a dramatic effort to produce ventilators, the US yesterday paved the way for it by invoking the Defense Production Act of 1950 and the UK is finally acting too.

Age discrimination

Another variation of the mitigate strategy is strong age discrimination. The Imperial College model considered an option of quarantining everyone over 70, which proved insufficient. A more extreme scenario would be to quarantine everyone over 60 or even everyone over 50 and allow the younger people to develop herd immunity. A much smaller percentage of young people require an ICU, after adjusting for unreported cases probably under 1%, and it may be feasible for hospitals to cope. 

The under-50s represent over 50% of the population in every country besides Monaco. 

However, even once most of the under 50s have recovered from the disease and develop herd immunity, the older people would have to be released into society and the epidemic would be rekindled, especially assuming that older people often socialize with each other. As above, there is simply not sufficient hospital capacity to allow the older people to be treated even if they are infected over several months. The arithmetic suggests that this is not a feasible strategy.

Suppress to zero and isolate

So the data does not seem to support the UK’s mitigation strategy, unless it is accompanied with massive building of ICUs or at least makeshift ICU beds with ventilators. This week the UK seems to have realized that.

The alternative approach is to create such extreme social distancing that R0 goes below 1. At this point the disease rates reduce exponentially reaching zero within some weeks. 

Several countries are on their way to achieving this, particularly China, South Korea,Taiwan, Singapore. They are using various combinations of draconian social distancing, enforced curfews, and mobile phone tracking. They may have benefited from their experience with SARS.

Even if the suppression is working, if it is relaxed while a single person is contagious, or if a single contagious person is subsequently allowed to enter the country unchecked, a second wave of exponential growth can occur as occurred in the Spanish Flu of 1918.

So in this strategy the extreme social distancing would have to be maintained until there are literally zero unquarantined infected people. 99% of people will develop symptoms within 14 days, so once the entire population is symptom free for 14 days one might conclude that there are zero cases. However given that some people may be asymptomatic, and that 1% have a longer than 14 days incubation period, it might be necessary to wait 21 or 28 days. 

Once a country is confident that there are zero unquarantined infected people, it can relax social distancing, but it will have to keep its borders tightly controlled until the end of the pandemic. Any goods coming in from countries which still have an epidemic will have to be sanitized and any visitors from such countries will have to be quarantined for at least 14 days.

Today there was exciting news that China reported zero new domestic cases, about 5 weeks after their peak. In about 4 weeks China may start to relax social distancing while maintaining tight border restrictions. That would imply that with extreme measures a country can go from a major outbreak to zero cases in 2-3 months which is probably the best case we all have to prepare for. It remains to be seen if a democracy can achieve the same, although right now democracies are not shy to use state of emergency powers.

One might imagine a club of COVID-19-free countries reopening travel between them as we approach the summer, while they maintain strictly closed borders, or at least strict quarantines, with other countries. Other borders will have to remain tightly closed or controlled until there is a vaccine or new treatment, or until the virus is eradicated from all countries. 

Summary

Much is unknown. The virus may evolve. The summer weather may have some impact, although the virus did spread in the southern hemisphere’s summer. New treatments may become available. But based on the current data it seems that the only viable strategy for a country or group of countries is suppression through extreme social distancing till there are zero cases, and then isolation through border controls from any country which still has the epidemic, until the pandemic is obliterated by a vaccine or effective treatment is developed. It seems this state may be achieved in 2-3 months.

The alternative strategy of allowing the virus to spread while slowing it enough to flatten the peak, as the UK had been planning, could work only if countries prepare with a massive increase in ventilators and staff, and of course if countries must be willing to sacrifice an estimated 0.6% of the population. Then again months of extreme social isolation will take a significant human toll as well.

There will be an extreme impact on economies. Fortunately nowadays, many people can work from home, but other parts of the economy including most entertainment, leisure, and travel are rapidly shutting down.

Not everyone can work from home. Food and medicines require people to be physically present in factories, farms, raw material suppliers, transportation and warehouses, and utilities providing electricity, water, internet, cloud computing, and fuel, and the chain of support companies. Protocols are being introduced to allow these people to work without infecting each other. And those people at work in turn require transportation, while their kids must have some kind of care. 

All of this requires financial systems to keep operating.

So far China, South Korea,Taiwan and Singapore are providing a positive indication that countries can overcome the virus spread, by applying sufficiently draconian measures for some weeks, while still keeping a basic economy in operation.

We should all plan for extreme social distancing for at least 3 months, longer in countries with lighter controls, during this time and probably for some months after, there will be a severe global recession. After that there will be many more months of travel restrictions, probably lifting gradually over a year or more as more countries become virus free, until the pandemic is completely over.

Today there is room for significant optimism having seen China go from a roaring pandemic to zero new cases in 2-3 months. There is now a proven path to beat this virus by the time of the northern hemisphere summer, and then gradually restore normality over the rest of the year.