Saturday, April 18, 2020

My letter to John Campbell

John Campbell is a retired Professor of nursing who lives in the UK. I have been watching his youtube site religiouslys it is a wonderful source for COVID-19 information.

Here is the letter I just sent him. Curious if he will reply.

Hi John
I am a huge fan of your youtube videos. I watch them regularly here in Canada.
I was going through the data and I noticed that New York State (based on the Worldometer data) has 901 deaths/1 million people. It seems that the state number could unfortunately hit the 1000 deaths/1 million figure sometime this week which means that 1/1000 of the states's population would have died as a result of this illness. This of course corresponds to 0.1% of the State's population. Unless the entire state is infected (very doubtful especially with herd immunity at 70%) this implies that anything below 0.1% for lethality is way off (ballpark of the Stanford study). In fact if herd immunity is to be considered I would be skeptical of anything below 0.2% (twice the lethality of seasonal flu).
Now if the Dutch study you referenced yesterday is to be believed (and I realize that there is a danger with extrapolations) than that would imply that 100/6.7 = 15% of the population is infected. The Dutch lethality rate is 0.67% or so. At a 0.3% infection rate (somewhere in between the two studies) 33% of New York could be infected...almost halfway to herd immunity.
Here is another take.... Iceland has 9 deaths out of 1760 cases (see worldometer). The country has also tested over 12% of its population. Even in such a biased case sample the death rate for those infected is only 0.51%. Apply said figure to New York would translate to 20-25% of the population infected.
I realize that there are assumptions implicit here (accurate population accounts, regional transference of data, quality of healthcare, accuracy of testing, small number jitter) but there may be something worth thinking about here.
Anyway. Just some food for thought.



No comments: