Missing Deaths Exploration (part 4)

Weekly all cause death figures published by the Office for National Statistics do not stand up to scrutiny. The evidence indicates we are missing a bunch of young deaths.

In part 2 and part 3 of this series I treated us to a set of slides depicting excess all cause deaths (date of death) by quinary age group over the period July 2019 – July 2022 using the dataset requested by Joel Smalley under FOI. It is pretty obvious that a whole heap of younger deaths since the beginning of 2022 are missing from the national record and this is likely due to delays incurred by involvement of the coroner. At some point I’m expecting this wrinkle will get ironed out but meanwhile analysts are left with a dirty great hole that prevents robust assessment of vaccine harm.

How Big?

An obvious first question is how big is that dirty great hole? The answer is that we do not know, and cannot possibly know for certain until all the certificates are settled. All I can do at this stage is reveal how many deaths appear to be missing for 2022/w1 – 2022/w30 with reference to the 2015 – 2019 baseline. To this end I’ve produced this summary table:

Overall we are missing 10,799 deaths, this representing a 3.5% loss in comparison to the 2015 – 2019 mean baseline. Of the 20 quinary age groups listed only two (65 – 69y; 75 – 79y) managed to come close to what is expected for this portion of the year. The biggest hole, in terms of percentage missing, is for the 20 – 24y group with only 374 deaths reported for an expected total of 721 (48.1% missing). It should be pretty obvious from figures under the ‘factor’ and ‘%age’ columns that the missing death phenomenon is essentially a youthful issue.

Something else that intrigues me are missing deaths for the 1 – 4y age group, which is 35.1% down on expected. Maybe there are folk out there that might provide some background to this, but it is eminently possible that a dip in birth rate during the pandemic is driving this. If this is the case then it’s an odd situation because locking people away together usually has the opposite effect!

Scraping The Barrel

What should be obvious in all this is that modelling of vaccine benefit/harm for 2022 is pretty pointless. If I were unscrupulous and/or believed in the magic of the experimental gene therapy then I’d have a tremendous opportunity to turn the handle and pretend the jabs are preventing deaths: this would be an inescapable result from so many missing deaths. The public, being as numerically naïve as they are, stand no chance of understanding the level of tripe served by numerically-minded authorities, who must know exactly what they are doing. What puzzles me is how these bods manage to sleep at night.

What I can do, I guess - until the coroner’s reports start flooding the ONS database with the reality of the situation - is to undertake limited statistical modelling using records up to 2021/w52 to see if I can determine a statistically significant early effect for vaccination using ARIMA time series techniques. When I say ‘significant effect’ please note that I am not referring to the direction of that effect (i.e. harm or benefit), for this is something that must be determined empirically within the data despite the mountain of evidence of harm that grows each day.

This may sound unecessarily quaint to folk shouting, “it’s bleeding obvious, mate”, but if I don’t approach this issue rigorously as a statistician doing their geek thing then I might as well jack in analysis and write protest songs for posterity. I shall thus get the biscuit tin open (now brimming with mince pies) and go turn the handle on scraping the barrel…

Kettle On!