Two new data points have come in since the last update. 8/18 and 8/19 with 19 and 29 deaths respectively. Both within the prediction bounds, unfortunately.
Somebody has to be your intellectually slowest follower, and I think that would be me. Do you have the patience to help me understand this statement? « However, Walensky noted that data from Israel suggests "increased risk of severe disease amongst those vaccinated early." I think that the group who were vaccinated early was the older, weaker or sicker group. So does it make sense to compare this group with the younger, stronger and healthier group that was vaccinated later and to conclude that the vaccine is losing efficacy? I think it does not, but what do you say?
Second dumb question: I am having trouble keeping in mind exactly what your graph uses as the past which you are now entering data to compare with. Is it 2020 highest spike of disease data? Is it what would be expected if nobody were vaccinated? Sorry. I think I asked before, but something isn’t clicking in my poooooor old brain.
You are right to think of other reasons to explain the data - the issue 'experts' are dealing with right now it's hard for them to keep track of all the age groups, comorbidities, and other influential factors - and they keep communicating vaccines "work" which is a very blunt description. When the data doesn't "look like" working anymore they have to then scramble to look at the data to find out what might be going on - so they wind up being behind in noticing all the details.
It might be true that those who were vaccinated earliest, no matter old or young, are losing the infection protection to some degree. It's hard to know when you also have a different variant that might be more infectious. It could be both are true.
For the predictions I simply looked at cases and deaths before vaccines were widely rolled out. The prediction looks at cases now, and says "what would the deaths look like if they looked like they did before the vaccination rollout." It uses some math that may intimidate you, but it's almost like a mathematical eyeballing of cases/deaths before vaccination and pretending vaccination never happened and guessing deaths.
I had read an article noting that the absolute risk reduction of the vaccines was about 1% reduction --so could this be a reason why the hospitalizations and deaths are tracking similar to previous waves despite hospitalizations?
Tomorrow this experiment will be over. I will rerun the prediction model for the next 14 days, but not update it daily.
I will then begin to investigate the underlying data (it's a bit rough for me because all the pdfs are in hebrew) to try to better see the ages and vaccination status that made up the numbers.
Given unknown ages and risk factors, we observed before vaccination that X cases would have Y deaths approximately 14 days later. Using that observed proportion, we estimate Y* deaths 14 days after observing X* cases.
The exponential nature of the case curve is not a factor in the infection model. Yes, an exponential case curve would yield an exponential death curve if you estimate a scalar proportion of cases result in deaths - but two scalar multiples would still yield distinct curves.
Somebody has to be your intellectually slowest follower, and I think that would be me. Do you have the patience to help me understand this statement? « However, Walensky noted that data from Israel suggests "increased risk of severe disease amongst those vaccinated early." I think that the group who were vaccinated early was the older, weaker or sicker group. So does it make sense to compare this group with the younger, stronger and healthier group that was vaccinated later and to conclude that the vaccine is losing efficacy? I think it does not, but what do you say?
Second dumb question: I am having trouble keeping in mind exactly what your graph uses as the past which you are now entering data to compare with. Is it 2020 highest spike of disease data? Is it what would be expected if nobody were vaccinated? Sorry. I think I asked before, but something isn’t clicking in my poooooor old brain.
You are right to think of other reasons to explain the data - the issue 'experts' are dealing with right now it's hard for them to keep track of all the age groups, comorbidities, and other influential factors - and they keep communicating vaccines "work" which is a very blunt description. When the data doesn't "look like" working anymore they have to then scramble to look at the data to find out what might be going on - so they wind up being behind in noticing all the details.
It might be true that those who were vaccinated earliest, no matter old or young, are losing the infection protection to some degree. It's hard to know when you also have a different variant that might be more infectious. It could be both are true.
For the predictions I simply looked at cases and deaths before vaccines were widely rolled out. The prediction looks at cases now, and says "what would the deaths look like if they looked like they did before the vaccination rollout." It uses some math that may intimidate you, but it's almost like a mathematical eyeballing of cases/deaths before vaccination and pretending vaccination never happened and guessing deaths.
It just happens to be far too accurate.
Thanks. I understand better now.
I expect I will find some of that group unfortunately affected, yes.
See this paper: https://www.medrxiv.org/content/10.1101/2021.07.29.21261317v1.full.pdf
"After adjusting for comorbidities, we found a
statistically significant 53% (CI 40–68%) increased risk for breakthrough infection in early
Vaccinees (P<0.001) (Table 1, Model 1). When stratifying the results by age, we found a
similar trend across all age groups."
Dr RollerGator, would you care to comment on this? Making the rounds as a debunking of the "Israeli misinformation". https://www.covid-datascience.com/post/israeli-data-how-can-efficacy-vs-severe-disease-be-strong-when-60-of-hospitalized-are-vaccinated
That website is not looking at the same things as I am - and is arguing against other people.
I had read an article noting that the absolute risk reduction of the vaccines was about 1% reduction --so could this be a reason why the hospitalizations and deaths are tracking similar to previous waves despite hospitalizations?
So what will you do in a day or 2, recalibrate the formula?
Tomorrow this experiment will be over. I will rerun the prediction model for the next 14 days, but not update it daily.
I will then begin to investigate the underlying data (it's a bit rough for me because all the pdfs are in hebrew) to try to better see the ages and vaccination status that made up the numbers.
I may be misunderstanding your comment.
Given unknown ages and risk factors, we observed before vaccination that X cases would have Y deaths approximately 14 days later. Using that observed proportion, we estimate Y* deaths 14 days after observing X* cases.
The exponential nature of the case curve is not a factor in the infection model. Yes, an exponential case curve would yield an exponential death curve if you estimate a scalar proportion of cases result in deaths - but two scalar multiples would still yield distinct curves.