In this article I utilise data from an unknown NHS Trust and use split-file staged multivariate logistic regression in the prediction of vaccination status prior to death
Good man. Yes indeed, I have a limited dump for one NHS Trust that I'm using for this latest series. There's an awful lot you can deduce when analysing individual records en masse, though it is a tricky business.
1) What is "constant" and why is it so highly significant?
2) I guess the counter argument would be that covid is not so much a respiratory disease. Could you for example find the top 5 supposed presentations of covid death (i.e. respiratory + organ failure + etc...) and see if pooling them together is required to make something pop up?
3) I don't see (total diagnoses*covid),
4) Could likewise do 5 negative control interaction terms (i.e. covid*trauma) and see if they add up to a spurious benefit.
5) Could likewise do all the above interaction terms with just nonfatal covid cases. Can't remember if we've been there already.
Waiting for mainstream media to pick up on this stellar work... Instead, we've got Prof Pantsdown, bought off Spiegelhoff, and double agent Stats Jenkins!
LOL - smack 'em with a kipper (smoked)! More modelling along the same lines is planned for the weekend after I have raided the pantry. Folk will try to squirm out of this by pretending COVID wasn't respiratory, so I'll open up with the gatling gun...
Steve Kirsch has been calling for US States to publish and merge their death and vaccination databases.
https://open.substack.com/pub/stevekirsch/p/the-medicare-records-clearly-show?r=peo1w&utm_medium=ios&utm_campaign=post
Don’t you effectively have a subset of this?
Good man. Yes indeed, I have a limited dump for one NHS Trust that I'm using for this latest series. There's an awful lot you can deduce when analysing individual records en masse, though it is a tricky business.
1) What is "constant" and why is it so highly significant?
2) I guess the counter argument would be that covid is not so much a respiratory disease. Could you for example find the top 5 supposed presentations of covid death (i.e. respiratory + organ failure + etc...) and see if pooling them together is required to make something pop up?
3) I don't see (total diagnoses*covid),
4) Could likewise do 5 negative control interaction terms (i.e. covid*trauma) and see if they add up to a spurious benefit.
5) Could likewise do all the above interaction terms with just nonfatal covid cases. Can't remember if we've been there already.
1) Constant is the base risk of vaccination at death, which will be high owing to the substantial uptake.
2) I've already done preliminary work on this but will be writing this up for substack. This is what is meant by visiting my pantry again.
3) Coming up in the next article since this links to 2).
4) Great idea! I've already established several indicators along these lines. The tricky thing is trauma arising from falls in the elderly.
5) This can be done using the ECDS dataset. Again, prelim work is already reported but not squashed into substack format.
Waiting for mainstream media to pick up on this stellar work... Instead, we've got Prof Pantsdown, bought off Spiegelhoff, and double agent Stats Jenkins!
LOL - smack 'em with a kipper (smoked)! More modelling along the same lines is planned for the weekend after I have raided the pantry. Folk will try to squirm out of this by pretending COVID wasn't respiratory, so I'll open up with the gatling gun...