Alex is wrong on this one

All cause mortality rates in UK

Preface this post by saying I believe Alex Berenson has been invaluable during this pandemic. We need more folks willing to look skeptically & I believe he has been right far more often than wrong & his warnings often precede the official narrative by months. I also believe he genuinely tries to play it fair and offer analysis that best represents the data. Nevertheless, I believe this recent post has gotten it wrong. I don’t believe he intentionally is meaning to deceive but I do believe he is not as careful as he should have been in implying such a dramatic result.

If you have not seen the post, this chart is the punchline:

I’m confident the above chart is accurate, however it is very misleading. The reason for this is the very wide age range (10-59) and the wide difference in vaccination rates & mortality by age. Here is a simple example to illustrate the point:

• Suppose we have 90 fully vaxxed 50-59 year olds & 10 unvaxxed 50-59 year olds. In both groups 10% die, 9 fully vaxxed and 1 unvaxxed.

• Now suppose there are 100 10-49 year olds, all unvaxxed and no one dies.

• In this (obviously greatly exaggerated scenario), 10% of vaccinated people have died, while only 1/110 or ~1% of unvaccinated people have died.

• While it would be accurate to say vaccinated 10-59 year olds in this group were dying at 10x the rate of unvaccinted, it would be very, very misleading.

The root of the problem, from what I can see, is that Alex is using table 4 from the linked UK data (age specific) vs table 2 (age-standardised). Plotting table 2 gives a different picture:

As you can see, the fully vaxxed (red) have lower age-adjusted mortality than the unvaxed (light blue). However, we do see something very odd… the age-adjusted mortality of the single vaxxed >21 days since 1st dose are highest of all, and hit some alarming rates in May-July. I don’t have an explanation. I’m sure some will jump to saying that dose harmed them. Could be. Could also be that those folks are sicker than average (maybe decided they couldn’t tolerate the 2nd dose). Don’t know, but definitely worth investigating if anyone knows of additional data to look at.

Also of note (harder to see above) is that the rates between unvaxxed and fully vaxxed do look to be converging in some manner:

Again, it’s good that the UK provides some data to be analyzed. Not aware of anything close to this available for US.

None of this is meant to slam Alex Berenson… again, I think he’s been invaluable during this whole thing. But I believe it’s incumbent upon those of us who are skeptical of the general narrative to be very careful in our analysis and not mislead, lest we lose trust and credibility among those we might convince.

PS I have quite a few new folks looking at this. If you are interested in smaller bits of comments and analysis, I’m on Twitter: @VanVoorheesVII