Israeli physicist Eli Waxmann ("Neil Ferguson from Ali Express"), always been wrong, continues with fear mongering.
Now he predicts many deaths of unvaccinated
But they are mostly children, and healthy not-old adults. And he predicts as many death as pro vax waves... >>
His models are wrong mostly due to two wrong assumptions:
1. Asymptomatic illness is rare (25%), and the authorities know about most of those who have been infected. This way he get a very high death rate, and assume that there remain much more susceptible than the true number.
He gave a lecture in May 2020. Back then the testing in Israel was very limited. There were about 15,000 cases and 222 death, from that he concluded 1.5% death rate ^. And he insisted there is no need of further testing because we know most of them.
A seroprevalence survey conducted in July 2020 (no cases on May - July) revealed that 5.5% of population got infected. That it, 500,000 or 0.5 million.
So death rate is about 0.05%..
Also, the assumption of knowing all infected ppl is blantly falsified. gov.il/en/departments…
Also now, estimating number of not immune people, he ignores the fact that many contracted the virus unknowingly.
For example, a recent seroprevalence survey concluded that 20% of children have antibodies, and 30% of teens (those it was so urgent to vax..) timesofisrael.com/study-finds-1-…
Even the official death rate now
(# deaths / # confirmed cases, Case Fatality Ratio, abbr. CFR)
is much less than the 1.5% he's talking about:
6450 / 850,000 = 0.76%
Now for his second crucial error.
2. He is absolutely clueless about dynamics of pandemics.
In particular, he assumes exponential growth with constant rate of multiplication, and extrapolate from last days.
Take an example.
This is the situation is Sep. 29, 2020.
The wave is in its peak, starting to decline.
(source: worldometers.info/coronavirus/co…)
But Waxmann doesn't know that.
All he knows is the last days' steep incline, what he confuses with "exponential". So take some constant rate, and extrapolate.
In Sep. 29, he predicted 1500 severely ill patients after Sukkot holiday (9 Oct., 10 days later) ynet.co.il/health/article…
But the number of severely ill patients start to decline immediately.
When he predicted 1500 within two weeks, there were 800, and it was almost the peak of the wave
(source: datadashboard.health.gov.il/COVID-19/gener…)
And "after Sukkot", 10.10.20, it was already in decline (and no, no change in behavior can effect in a week. It's less the time frame required to get infected, incubation period, and worsening of the condition).
Actually, as you can see, Israel never exceeded 1200 "severely ill" patients, even in winter wave, and even with their extended definition of severely ill (only 1/3 of them need ICU)
His erroneous assumption is exponential growth, i.e. constant multiplication rate ("R"), that, as such, can be estimated from last days, because it's constant throughout.
In real world, Rt is monotonously decaying, so need all former data to estimate its decay rate and predict.
They keep falling in this error.
For example, from 16 July 2020
What was their mistake?
As I said:
They assume constant R. so they took the last days' R (which, incidentally, was constant for few days, transiently) and extrapolate. Looking on data from whole wave they would have understood that it's transient and will decay.
It doesn't seems so bad, ha?
Now look how multiplication rate is realized into numbers:
constant vs. decay R ~ exponential vs (asymptotic?) number
They started similar, then go different ways...
(see the 3200 they predicted?)
The original @neil_ferguson would never publish such a model.
Erroneous as he is (in 2005, predicted 200M deaths in bird flu), he took Epidemiology 101.
He knows that pandemics have monotonously decaying R, due to people recovered and becoming immune, thus slowing spread.
So he used SIR model, based on Susceptible-Infected-Remove dynamics (with far fetched parameters, of course), modeling a... wave.
He never predicted a spread continuing indefinitely in constant rate. Always wave, amplitude and width can change.
בשדה התעופה בחאניה, 3 דקות מרגע הכניסה ועד לגייט.
בישראל - שעתיים וחצי.
אלפי אנשים תקועים בתורות אינסופיים, התור לזה ששואל אם ארזת לבד ונותן מדבקה צהובה, ואז לזו ששואלת אם יש מצ'טה ונותנת מדבקה כתומה. ועכשיו נוספה גם טפסיאדת הקורונה המיותרת.
אלו התורים בהלוך בנתב"ג.
אף אחד שם לא עושה כלום. זה רק תורות ל"בטחון" ("ארזת לבד? נתנו לך משהו להעביר? עבר לך בראש פעם לרצוח ראש ממשלה?"), ואז לדלפק בו בודקים כל מיני טפסי קורונה, אחד אחד. איפה ההייטק ניישן?
בפעם הקודמת שחזרתי (לפני כחודש) הייתה התקהלות גדולה בתור לבדיקות הקורונה במתחם (הסגור!). אלו היו מסיבות ההדבקה הגדולות של ישראל. אחר כך התפלאו למה הרבה אנשים יוצאים חיוביים שלושה ימים אחר כך, למרות שנבדקו לפני הטיסה ואחריה ונמצאו שליליים
They are measuring their own measures...
bottom left figure (all ages):
%vax in cases very similar to %vax in population while %recovered in cases much lower than in population (835K/9.2M, 9%, but see next) >>
* immunity of recovered is much more stronger, long lasting and robust, than that of vaccination.
* should reduce few cases in unvax group (those unvax & recovered, b/c I compare to unvax & not recovered), thus slightly decrease base rate and VE, but I think it's negligible.
Look at the top right graph.
Does the reason that % of ultra-orthodox people in cases (1%) is much lower than their % in population (about 12%), is the high percent of recovered there?
(officially 30% of them are confirmed cases. The true number can be 50% or more)
Then, vax status.
Go here (data pulled from Israeli MoH "dashboard"). github.com/dancarmoz/isra…
Sum accumulated # 2nd vax number 20 days ago over 20-59 age groups. let's call it A.
Sum pop over 20-59 age groups, let's call it B.
A/B is %vax 2nd dose >= days ago.
Vaccine Efficiency over time in Israel by age groups.
Thanks epidemiologist @prof_shahar for scientific consulting.
Discussion in comments. >>
Source:
Israeli government database: data.gov.il/dataset/covid-…
cases for vax and unvax: אימותים לאחר חיסון
% vax: גילאי המתחסנים
Quick recap of vaccine efficiency:
The true values of # vax, # unvax are known, hence we have odds(vaccination status).
But # infected&vax, infected&unvax for groups whose all parameters being equal is unknown.
We do have number of confirmed cases in each group. Is it a reliable measurement?
what does it imply?
if all other parameters (testing policy and rate, exposure rate - e.g. who travel abroad more, etc.) being equal between the vax & unvax groups, that implies that vaccine does not prevent infection (defined as "being confirmed by PCR"), if not the opposite.
Of course I can't ensure that.
Basically, without equalizing all relevant parameters all this numbers are statistically/scientifically meaningless.
Same is true for the Israeli studies that showed prevention based on those numbers without equalization.
It works both sides...