Sweden, the US, and Covid-19

Sweden has been somewhat slower than other countries to implement social distancing, although looking into it for this post has revealed to me that the difference between Sweden and the US are not as great as is generally believed.

General recommendations and awareness to do social distancing appear to have been happening at more or less the same time in both countries, although that is hard to pin down exactly, because the US is huge and recommendations varied across the states.

In the US, the US federal government officially recommended closing schools and avoiding gatherings of more than 10 people on March 16th. On the same day, the Swedish government recommended that people over 70 should stay home, and that employers should consider letting their employees work from home. At this point, Sweden shut down universities and schools for older kids. But schools for young children remain open, even now. The argument there is that healthcare workers need somewhere for their young kids to go while they are working. Older kids can be home alone and do school online. So there is one real difference. In the US, elementary-aged kids have been out of school since about 3/16, whereas in Sweden they are still in school.

On 3/24, Sweden placed restrictions on restaurants, in which eating at crowded bars was prohibited, but take-out and socially-distanced eat-in tables were allowed. In the US, we shut down all eat-in tables around 3/16. So that's another real difference.

Sweden eventually banned (as opposed to just a recommendation) all gatherings over 50 people (on 3/27). The March 16th recommendation by the White House recommended limiting gatherings to 10. That's another difference.

So far, those are only differences in recommendations and laws, which are not the same as people actually socially distancing. If we look at Google’s cellphone tracking data, the US has a 38% decrease (as of 4/26) in tracking people to workplaces.  In Sweden it’s a 36% decrease, probably not a statistically significant difference. The US has a 14% increase in residential locations, whereas in Sweden it’s 11%. The only major difference in location data for the two countries is for parks (which includes public spaces like beaches).  People in these spaces has decreased 16% in the US, but *increased* 56% in Sweden.  I don’t know enough about public spaces in Sweden (or much of the US, for that matter) to say how difficult it would be to maintain distance there.  The other difference is that retail movement in the US went down 45% and stayed there.  In Sweden it has bounced around a bit, and is now at an 18% decrease.  This gives the impression that people are genuinely spending more time in public spaces in Sweden than in the US. 

Now, if we look at the results (via Worldometer) in terms of Covid-19, the US has 166 deaths per million, whereas Sweden has 217 deaths/million.  We also need to consider testing.  Countries that test more will have more official Covid-19 deaths.  The US is at about 16k tests per million, and Sweden is at 9.3k per million. This suggests that Sweden might be missing more of their Covid deaths than is the US, which would make the difference even greater. 

If you look at the graphs of deaths per day on Worldometer, Sweden appears to be processing tests on a weekly cycle, which makes trends hard to identify.  But to my eye, it looks like the US is leveling out more than is Sweden.

It's at least possible that Sweden's strategy will have an unintended beneficial effect.  There is some evidence that you can re-awaken immunological memory to other coronaviruses to help fight this one.  The idea is that through low-level exposure over time, you will wake up the memory T-cells from prior infections.  But you could get this same benefit without the risk of people actually getting sick by giving everyone low levels of inactivated virus, perhaps by putting it into the water supply.

It's pretty clear that both the people on the left claiming that Sweden is seeing a huge spike, and those on the right claiming that Sweden has managed to avoid the negative consequences without shutting down are both wrong. Sweden is partially shut down, probably not quite as much as the US, and is seeing a moderate increase in deaths relative to the US.