There's some new Covid-19 survey data out of Sweden this week: 0.9% of those tested nationwide had an ongoing infection (that is, positive PCR test) as of April 21-24, 2.3% of those in greater Stockholm (1/n) folkhalsomyndigheten.se/nyheter-och-pr…
A similar survey from March 26 through April 3 found a 2.5% infection rate in the Stockholm area (2/n) folkhalsomyndigheten.se/nyheter-och-pr…
It may not be impossible but I think it's now *extremely* unlikely (I'll let @NateSilver538 do the math) that the Public Health Agency of Sweden's projection that 26% of Stockholm's population would have been infected by May 1 was correct folkhalsomyndigheten.se/nyheter-och-pr…
As @AdamJKucharski pointed out a couple weeks ago, the model assumed that people with Covid-19 would only test positive over a five-day period (at least, I think that's what he's saying) 4/n
Meanwhile, a new study out of @IcahnMountSinai finds that people sometimes test positive for SARS-CoV-2 as much as 28 days after their Covid-19 symptoms have gone away 5/n medrxiv.org/content/10.110…
So how many people in the Stockholm region have or have had the coronavirus? If you just take the 1,779 deaths reported so far and assume an 0.5% infection fatality rate, you get 355,800, or 15% of the population. Which seems reasonable enough, give or take 5 percentage points
In other words, the assertions that Stockholm was or is nearing herd immunity seem to have been wrong. On the other hand, infections and deaths are declining. Even a pretty relaxed social distancing policy can keep Covid-19 more or less in check, it seems.
So was the Swedish approach right? The best argument for it always seemed to be that it avoided lockdown fatigue. But neighboring countries are now easing their lockdowns with far lower infection rates than Sweden ... (8?/n; sorry for losing the plot on numbering)
… meaning that they may get back to truly normal(ish) life sooner than Sweden, which will presumably still be dealing with Covid-19 cases and even localized outbreaks for a while to come. Plus, a lot more people have died in Sweden. (9/n)
Still, I also think this means that people who freak out at photos of people sitting in parks in New York or going to the barber in Georgia are maybe overestimating how much behavior like that contributes to transmission. 10/10
