The spread of Covid-19 in Sweden is much discussed. I think that some points are often missed, or not highlighted enough. So here is my two pence on the matter. Partly in reply to posts of @joel_c_miller @AdamJKucharski and @BillHanage

Sweden has been hit relatively hard, with officially almost 600 deaths per million inhabitants, while the hardest hit region (Stockholm) has well over thousand deaths per million. Many deaths were in elderly care homes (almost 70% of the fatalities was in the 80+ age group).

Although the Swedish government relied for a big part on advising the people and trusting their wisdom, they implemented some restrictions, such as no gatherings for group over 50+ and no visits allowed to elderly care homes.

The main point where Sweden stood out compared to the rest of the world is that schools for children up to around 16 years old stayed open, mostly without even the slightest restriction, apart from a stricter rule on sick children to stay home.

The main restrictions were implemented on March 27, around the same time as e.g. The Netherlands and Belgium implemented measures, but 2 weeks after neighboring Denmark and Norway did so.

Already in the first half of April the decline in the (seven day average) number of new IC admissions and new fatalities (both in official numbers and in excess death) was visible. This was most pronounced in the hardest hit Stockholm region.

This is a clear indication that the measures taken in Sweden were enough to bring the reproduction number below 1, perhaps helped by the immunity already built up in Stockholm. Although the region probably has not reached the herd immunity level yet (whatever that level may be)

That immunity in Stockholm might have helped is suggested that the second city of Sweden, Gothenburg only had the peak in IC admissions and deaths in May.

It might be that the Swedish public health agency did not expect that the measures taken would bring R below 1: In April they still predicted the peak of infections in Stockholm to be in the end of April, when the (delayed) number of deaths was already declining for some weeks.

Sweden did worse than many other countries, because the decline in the number of deaths, although clearly present, was slower than in many other countries.

If Sweden would have acted at the same time as Denmark, and would have had the peak of the number of deaths per day around the same time. That peak would probably have been around 50 deaths per day in early April, in stead of around 100 deaths per day around half April.

It is not clear to me whether this would have prevented spreads in elderly cares homes proportionally. Since the elderly home as managed in Sweden seem extremely susceptible to introductions. Just like in many other countries it is hard to keep the virus away from the elderly.

If the (overly) naive approach is taken, by just subtracting the number of deaths which occured between 50 deaths per day was reached “on the way up” until that level was reached “on the way down” about 4000 deaths (2/3 of total) would have been prevented.

Currently Sweden, and especially Stockholm region is doing pretty well. Schools are fully open, universities are partly back to on-campus teaching. Public Transport is less crowded than before pandemic, but there is hardly any mask-wearing going on. Still waiting for future data.

Main statistics on Sweden can be found at
https://www.folkhalsomyndigheten.se/smittskydd-beredskap/utbrott/aktuella-utbrott/covid-19/statistik-och-analyser/bekraftade-fall-i-sverige/
look especially at for the exact numbers
https://www.arcgis.com/sharing/rest/content/items/b5e7488e117749c19881cce45db13f7e/data