I assume my readers have a good idea of how Sweden has handled the COVID crisis compared to the rest of Europe. Briefly, they kept their schools, restaurants, and bars open with voluntary precautions in place. Critics have concluded this was a mistake, because there have been a lot more deaths there than neighboring countries. As of July 20, 2020, this is a true statement, but it is way too early to conclude that the Swedish approach was a mistake.
First, let’s review the facts up-to-date in total:
Country Population COVID cases COVID deaths
Sweden 10.2 million 78,000 5,639
Denmark 5.2 million 13,466 611
Norway 5.4 million 9,000 255
I tried to look up hospitalizations, but many countries don’t report this.
These are the reported national statistics, but there are many gaps. In the early days of the infection and there weren’t enough test kits, many people around the world with mild to no symptoms were never tested. Plus, I think there is still uncertainty about the accuracy of all COVID tests. There are many anecdotes of people testing positive with one method and negative with another, not explained by recovery time.
Clearly, a lot more people have been infected than the national numbers suggest. Antibody testing (still likely imperfect) recently in New York City and Stockholm found that 17%-18% of their populations have antibodies to the virus. These numbers are based on random sampling, not the positive test rate of people with symptoms or known exposures. Similarly, 19% of the people on the Diamond Princess cruise ship had evidence they contracted the disease (712/3711) and 13 people died (0.4%, and if we can make some stereotypical assumptions about the average age of a cruise passenger, then the death rate in younger people must be substantially less). Just for Sweden, this means the ratio of actual cases to known cases is likely at least 25 times higher. The daily death rate in Sweden for the past week has been in the 0-3 range. COVID has nearly burned itself out there. If only 17-18% of the population is protected, but 60% or more of a population must be immune for herd immunity to exist, how can this be?
The 2 explanations that make the most sense to me are 1) the tests still aren’t very accurate, and 2) lots of people have been infected by other corona viruses in the past that provide good cross-reactive protection. They don’t possess antibodies that are super specific to COVID-19, but they are still able to effectively bind to COVID-19 and call in the other immune system cells to destroy the virus. Corona viruses are not new. They are a common cause of common colds.
Another criticism of the Swedish approach was that in spite of its efforts to keep its economy going throughout the pandemic, its economy will still contract similar to surrounding countries. The latest articles I can find predict that all of the 3 Scandinavian countries’ economies will contract about 4% this year. The final numbers probably won’t be tallied for a few years. Several commentators say that Sweden should have been expected to take a harder hit, because more of its economy is based on manufacturing (Volvo, IKEA, e.g.). The Norwegian estimates apparently don’t include its ailing oil and gas industry.
What does all of this mean for Denmark and Norway? It means that they have a long way to go. If the ratios are similar, it means that they are still a long way off from 18% having detectable COVID-19 antibodies. The latest number I could find was that as of 3 weeks ago, 1% of Norwegians are antibody positive and 1% of Danes about 2 months ago.
What does this mean for the U.S.? It is clear that lots of people in the U.S. had mild to no cases of COVID infection already and were never tested, because the U.S. has had such a low supply of test kits. For the antibody prevalence approach, the CDC has launched a national study, but has released no results yet. An article from 3 weeks ago estimated that 5% of Connecticut and 7% of New York state had antibodies. As of today (July 20), 3.8 million Americans have been confirmed infected (1.2%) with 143,000 deaths.
Death certificates also have limitations too, but they are probably the best source on which to estimate future cases. The total death rate in Sweden is .055%, 0.1% if we just look at the more crowded Stockholm County (4 times less than the Diamond Princess, which likely speaks to its skewed passenger demographic). The total death rate in New York state is 0.16%, or 3 times higher than Sweden, 0.27% if we just look at New York City. At least this puts us in an order-of-magnitude range.
Let’s take a middle ground assumption of 0.1%. This means, without a vaccine, 328,000 Americans will die from COVID-19 before this is all over. At 143,000, we are almost halfway there. If we use the New York City death rate, then there will be about 894,000 deaths. There are still no effective treatments for COVID, other than supporting people who are so sick they are on a ventilator with some basic interventions such as steroids and maybe blood thinners.
Now let’s talk about time frames. Sweden and New York ran through most of their deaths in about 3 ½ months, Sweden about 4 ½ months. Let’s compare with Denmark, which is destined to have 5,200 deaths, but has only had 611 so far. It’s been averaging about 24 deaths per day under a stricter lockdown policy. At this rate, Denmark will have finished out its COVID threat by about 6 months. And 6 months is the earliest a vaccine could reasonably be ready for widespread distribution. Just today, there was an announcement that the early phase research of one vaccine showed that it generated an immune response. Vaccine development history is littered with vaccines that generated immune responses, but didn’t protect people from the disease. We cannot assume that any of these COVID vaccines will work until there is proof that they work.
What about the U.S.? It’s been averaging about 750 deaths/day recently, which projects to our COVID pandemic mostly burning out in about 8 months. If we assume a higher death rate closer to New York state, then the number climbs to 17 months, 3 years if we assume the New York City death rate.
Through the examples of Sweden and New York, we can draw conclusions on the natural history of the virus: Sweden through a conscious choice, New York through a lot of cases hitting it at once before it knew what was happening.
The media has looked at countries such as New Zealand and South Korea as exemplars of how the U.S. should have reacted to the virus. New Zealand has had 0-3 deaths per day since May. Both countries are essentially islands that can, to some degree, block themselves from the rest of the world. But now what? Do they remain isolated from the rest of the world as long as COVID is lurking? If their gamble was that a vaccine will be developed by early 2021, then in terms of lives extended, if a successful vaccine is developed, it means they guessed right. If a vaccine is never successfully developed and New Zealand continues to have minimal exposure to the outside world, its COVID pandemic will burn out in about 13 years, 33 years at New York City death rates.
The bummer for Sweden is that even if it is past the COVID crisis within its borders, the depressed worldwide economy will still affect the Swedish economy as companies like H&M, IKEA, Volvo, and Electrolux will likely have decreased sales until the world economy rebounds. Maybe Spotify can prop up their economy in the meantime.
Hi Richard,
totally agree. The assumption that COVID is so different to any other virus is an improbable counterpoint given nature’s emergent processes. Putting all your money on a vaccine sounds like a gambler putting his last penny on the big win – possible, but unlikely. We had viral epidemics before that have gone around, many unnoticed, and herd immunity ultimately has succeeded.
What is rather amasing is the lack of understanding of susceptibility – as we inflammage, some of us at an earlier, others at a later numeric age, we will become prone to any form of perturbation to tip us over into the other worlds. Why is it worse to die from COVID than H.pneumoniae? It is hypocritical to amplify the deaths of every person with a pos COVID swab but not those with influenza or H.pneumoniae all of which are affecting the elderly and the multimorbid out of proportion.
As you say, it may be a long time before we really know how badly, or for that matter who well, Sweden has done with its approach. The true picture may never come up as in 3-13 years no one will still count COVID death.