April 8, 2021 - 11:10am

First things first: if you are offered a vaccine, take it.

Second things second: I was wrong, and overconfident. I wrote a piece back when certain European countries suspended the use of the Oxford-AstraZeneca jab, implying that the regulators were innumerate: that they saw 37 blood clots among people who’d had the Ox/AZ vaccine and leapt to ban it, even though 37 was far fewer than you’d expect among a perfectly healthy population.

It’s become clear since then that the clot was of a particular, rare type, cerebral venous sinus thrombosis (CVST), combined with a reduced number of platelets — clotting cells in our blood. I’m told that it’s a particularly unusual presentation, so you can’t easily compare it to the base rate of CVST in the population (between 2 and 5 per million); it looks plausible, though not yet confirmed, that it is a causal relationship.

Since that piece about the vaccine included a “statistical style guide” taken from my new book, and given that the 11th rule of that style guide is “If you get it wrong, admit it,” I want to make that clear now. I still think that the decision was, as I said at the time, stupid and harmful — it has reduced confidence in a very safe, very effective vaccine and I am sure has cost more lives than it saved — but it was not a straightforward numerical error.

The clotting problem seems more common among younger people, and since younger people are also at much lower risk from Covid, the risk-benefit calculation, at an individual level, is quite close. If you are under 30 and if the virus is relatively rare (assuming around two people per 10,000 have it, as was the case last month), then the direct risk to you of ending up in ICU from one of these rare clots appears to be about the same as the risk of ending up in ICU from Covid: about one in 100,000 in 16 weeks, per these very helpful graphics from the Winton Centre for Risk & Evidence Communication in Cambridge.

As a result the Joint Committee on Vaccination and Immunisation (JCVI) has issued guidelines that the Ox/AZ vaccine should only be offered to under-30s if there isn’t another one available. (It has not been “ruled out” for younger people, despite some terrible headlines.)

It’s worth thinking about the risks, though. First, the absolute risk is very tiny. Professor Sir David Spiegelhalter, of the Winton Centre, tells me that if you filled Wembley Stadium with 20-year-olds, and gave them all the vaccine, then about one of them would end up in ICU with a clot, and in the next 16 weeks (given the current prevalence of the disease), about one of them would end up in ICU with Covid. When you take into account the fact that most people who have these clots survive, your risk of death is about the same as that of being killed crossing the road in any given six-month period, and I cross the road quite happily all the time. In essence, as has always been the case, young people are at very low risk from Covid. 

(This is not taking into account the risk of “long Covid”, which does seem to be significant in younger people and would probably change the calculation.)

But the knock-on effects of vaccinating a Wembley Stadium’s-worth of 20-year-olds would be a major reduction in the transmission of the virus to those around them. It would have very little effect on the risk to the young people themselves, but it would reduce the risk for the older people in their lives significantly.

As so often in this pandemic, young people are having to alter their lives in order to protect others from a disease which they are at little risk from themselves. If it is possible to give them a different vaccine, then that makes sense; the Ox/AZ jab will do them little good personally, making them neither safer nor more at risk. But if it means significant delays to the vaccination programme as a whole, then it may mean more deaths among other, more at-risk groups. If you are offered a vaccine, please do take it, whichever one it is.


Tom Chivers is a science writer. His second book, How to Read Numbers, is out now.

TomChivers