Main observations so far, excess deaths since vaccination, comments received, more "Why?" questions
Transcript from the Open Canada Data seminar (2022-05-13. Seminar. Part 1)
Video recording on YouTube:
Objectives and means of communicating with you
i am having two streams here: one on zoom and one on facebook live. why i like really have an idea of zoom? because i have a chat window here, and you could ask questions anonymously.
i really want a discussion. in fact, i encourage asking tough questions. as mentioned last time, i don't want to be seen as supporting one group or the other. i'm here to find the truth about what is happening here. so that's what we're trying to do.
if you're watching it on facebook live, then you could still post questions, but you probably know your name would be seen there. alternatively, you could send me email directly to my email, which is email@example.com . it reaches my mailbox, and i will share it with other data scientists, who are helping me to process the the data.
of course, you could also make comments on youtube, because all videos are uploaded on youtube. youtube is a kind of a filter. it would not allow “misinformation”, and we don't have it here, at least i'm aiming to make sure that there is no “misinformation”. it's all about the data, which is the most trusted data, it's government of canada data.
what we do - we just make it more understandable,so we can plot it, because, unfortunately, government of canada sites do not visualize data over time - for example, to show how numbers are changing each week, for example, total number of deaths, covid deaths among vaccinated, and these are a very important metrics…
Following up with @ChiefSciCan
one thing we have just done - we had a follow-up with dr. mona nemer from OttawaU. she was assigned by our prime minister as a chief science advisor to prime minister about four or five years ago. she contributed to developing open canada and scientific integrity guidelines, to make sure we follow the best peer-reviewing process and openness in presenting the data and making use of the data.
you see, i'm wearing again my canadian, proudly canadian, eyeglasses and proudly ukrainian t-shirt, as a part of disclosure that i am ukrainian, i was born in kyiv, ukraine and i grew up there, and i'm canadian.
but also part of that disclosure is that i studied in moscow. and i have to say from the very beginning - i have no expertise in constitutional rights, in legal rights - what anyone is doing in the government to the truckers or other people, leaving them without pay or even without employment insurance after they work for 30 years for the only reason than they follow their beliefs. i cannot comment on that.
Also, I cannot comment on any bio-medical aspects of this phenomenon - vaccines or pandemic. but i could simply just provide tools for visualization of data, so that anyone could see and this decide for themselves, and maybe ask the right questions . that's my objective in all those seminars. again i hope, people will use this opportunity really to challenge my statements here.
now we'll just go quickly one by one through the main results on our main IVIM portal. What’s new?
2:30 IVIM Newsletter, discussion with chief science adviser
i started writting newsletters on substack. you will find the link to it on IVIM portal. we already have there transcripts from our previous seminars. you would also see there discussion on open letter to chief science adviser of canada. - i hope we could have a respectful discussion with her, i'm sure it will be respectful, we have a lot of appreciation for each other's competencies in our domains and support of the democracy… then we had a discussion on when extra dose helps and when it may not help, and we'll talk more about this…
3:00 Why we call it Machine and feedback from alma-mater friends
We decided to call our IVIM portal “independently verified information machine”. why is it called "machine"? because really we're trying to to remove us, myself as a human, from this analysis. we want to consider this discussion in terms of these numbers, just as a machine, as an unbiased machine, which just takes the data and visualizes them, and which allows us - me, you, anyone whatever side you belong to- to look at these graphs, and start asking questions.
again, please don't hesitate to challenge me. and again i have already received some feedback, and i would like to address feedback - from my friends, friends from my university years at university of alberta, where i was doing my PhD studies. We support science we were doing our phds and graduate studies at the universities. so we have a lot of respect for scientists. so and when i received feedback from them, of course, i would like to follow up on it. i will make sure i do mention their comments, related specifically to when boosters help, when they do not help. that will be a little later. but first let's just quickly overview again the main findings we have so far.
Calamity in Atlantic?
The ultimate measure of success of dealing with a pandemic or with any health problem or any major calamity in the country would be, of course, deaths, in particular, by looking at the excess deaths, comparing the deaths this year to the deaths last year when pandemic started, but before vaccination and then before pandemic.
Key question: what were the average number of deaths in each age category before pandemic, before vaccination and after vaccination ?
The most startling or astounding observation we obtained was for age from 0 to 44 years, meaning young children (see Figure).
when we look at excess deaths before vaccination in new brunswick (from 2020 to 2021), we see the red line, which is the total number of deaths, is pretty much the same as average [historical] number of deaths for this young population, which is about 5 deaths per week. and then we would start seeing that right from when vaccination has started in January with high risk population, this number started going up, and then when the vaccination has been extended in may (marked by bold line), to general population., we see how many ? about 20 even 25 deaths or young people weekly. (ignore sharp decrease, because some provinces put zeros when there is no data, there is a delay a few weeks and even months for some provinces).
This is one of the main findings, and i'm inviting my friends, specifically from UofA, from OttawaU, my old friends from my alma mater, to try to explain it to me - how this could be justified or explained ?
we should be expecting less deaths, but see more, why?
we do see that it's not the same in every province. we see that there is increase in alberta, in british columbia in other atlantic province - increase in total deaths this year compared before vaccination.
we know that these variant - OMICRON, which is we have now - it's less lethal, it results in less deaths. for omicron, so we should not be expecting more deaths this year than last year.
of course, if all the measures work, then, of course, we should be seeing a decline in all total deaths. but unfortunately, we don't see that. we see the opposite. except maybe with the exception of quebec. this is the only province where the number of total death after vaccination is less than before vaccination.
we assume that we can trust this data, because this is the only data, in fact, which we have, which comes from official statistics Canada website. this is a regular table from this website.
i invite everyone to look at the transcript and youtube recording from our seminar three weeks ago. it was dedicated specifically to the excess deaths. we talked a lot about how these graphs are obtained, and how you can see those graphs yourselves by using the app - we have a deaths tracker app - it's all available.
Predictions from last week
last week we started talking about predicting the reported number of deaths. we started saying that reported numbers fit a line so nicely with the r square prevision metric - it is the metric which shows you how close your prediction is to your observed measurements - of over 95 percent: from the very beginning, when they started reporting the data (you see the red bar is 1.2% on the 10th of july - that's how they started reporting), and it goes only up, monotonically up, it never goes down.
and we already know why numbers are going only up - because they are also counting all the deaths which happened before people were fully vaccinated
this gives them advantage. it's called bias, mathematically speaking. but you can remove this bias, advantage by subtracting last week from previous week. that gives you weekly count of cases. and then you can compute percentages. and the weekly percentages are kind of stabilized around 75 - 80 % for deaths
we also talked about the fact that these numbers should not be taken literally. what you need to do is to look into the relative percentages within each population - population of vaccinated versus non-vaccinated versus partially vaccinated, and this is what we're doing here. we can plot these graphs covid cases deceased relative to vaccination status population size. blue line is the average for all canadians. then if the bars are above blue line, it means for this population it's worse. if it's lower than blue line for this population it's better. and what we see here ?
we see here that for three doses it's about the same as for average population. and so on. we see for one dose, for zero doses, for two plus doses, for three doses. so this is how you can see how much doses help and when.
Comment on Why more doses show more deaths?
a very important comment received is that we have to understand that doses are given first to those people who are most vulnerable, who are already about to die, so to speak. it's very unfortunate that we have to say it this way, but people do die on a regular basis. and these most vulnerable people, they were given vaccines, were the first, they were the first to be administered the vaccine, and vaccine saved them or at least it delayed, we could say, delayed their death, if the average mortality rate in canada is about 82 years remained about the same.
in either way we can see vaccine efficacy dissipates with time and we really want to know its effect on general, other-wise heathly population.
so we can say that vaccine helped high risk population to delay their deaths for several weeks. but for the younger population, of course, it's a much more challenging question. did it help or did it not ? because we know that after may everyone, including children, were vaccinated or forced to vaccinate not to lose their job.
(because you cannot live without income… you just can try to put yourself in the shoes of anyone who has mortgages, children going to activities, and you're given this ultimatum - whether you just get jab or you will lose your income for your entire family. this is quite a drastic measure, of course. we try to understand where these measures are coming from, and how can we talk about this. why they're doing this? )
we discussed these graphs in last week’s seminar. they show you that the odds of dying from covid (or with covid) dissipates with time for each additional dose.
why don't they talk about this ? we do not know why.
and we know this phenomenon is not only about canada. it's all over the world. that's why i really wanted to think more holistically, or strategically - who is deciding on what to report and what not to report? our governments are not the ones who are making the decisions on which numbers to be reported and when they need to be reported. these recommendations are provided from somewhere else, from a group of experts, let's call them, some organization who recommends how and what to report.
Many questions come up by looking at the data
why do they report counting all deaths even before vaccination has started ?
why they don't report covid deaths following vaccination by age ?
how they define covid deaths?
who are they - who decides on all of that?
this is how i've been thinking about it for quite some time here. and because this is not happening just now in 21st century. it was happening in 20th century as well, well before arrival of pandemic. let me try to show you something here [SHOWING THE SLIDE DECK AND SWITCHING TO PART 2 OF SEMINAR]
that's why i call this part of presentation -
"how to improve the world: first comes the theory and then comes its implementation " and we could compare 20th century to the 21st century