« Take 2 on the Johnson & Johnson vaccine trial results | Main | Probabilities of Covid-19 »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Trish Pomroy

Would your recommendation be the same when there was a report of a batch of deaths in the elderly who had received Pfizer’s vaccine?

Michael Droy

Isn't this really about EU politicians avoiding criticism for not having enough of the vaccine in the first place?


TP: Of course, if Pfizer is suspended pending investigation, I'd recommend that they call for a quick investigation, and promise to cooperate fully and provide all necessary data. That's what Dr. Fauci meant when he said "no need for anxiety, but complacency is equally mistaken."

MD: I've seen that argument but I don't see the connection. The worst that could happen here is an advisory for certain subgroups not to use this vaccine. If that is the desired result, it's not clear to me how it excuses slow rollout in general.


Why do you say that "The worst that could happen here is an advisory for certain subgroups not to use this vaccine" after what you say in the last paragraph of your post?

P.S.: Maybe also other vaccines cause blood clots in young people, but it is not apparent since only AstraZeneca is currently administered to young people.


Antonio: I wasn't entirely clear there. I'm envisioning that the "subgroup" will be smaller than "young people". If it's all young people, that would indeed be a big deal. I'm guessing it may be that it could cause some rare type of blood clot for some demographic. That ends up narrowing the side-effect to a small subgroup. That's kind of how statistics work. The signal is stronger if we hone in on a subsegment but then the finding cannot be generalized to everyone. I'm been saying this for a while now: the pharmas can restrict how they count cases, when they count cases, etc. This accentuates the signal but at the same time, the finding applies to a more limited subpopulation/period of time.

Celia S

Hi Kaiser,
Your point about countries and age groups can be expanded on.

Risk of dying with covid over the last year stands around 0.2% in the UK, 0.38% in US. Not sure on age group breakouts, as an example of a subgroup, but obviously these vary significantly.

In Japan and Norway however the rate is 0.01%. ironically, or not, as one sees it, one might be less cautious because one has performed worse

In Norway for example there have been 117 deaths temporarily (in time) connected to around 400,000 vaccinations so one might expect greater caution than in other places if for example a few more came along .

As an aside ifvone were/were able to look at these things from a global perspective I wonder how vaccination policy might differ?



CS: That's right. The cost-benefit analysis should differ by country since the base rates of cases or deaths are different. One challenge is the reliability of estimating rare events when honing in on subgroups. This situation has sorted itself out. The countries that performed the worst at controlling the outbreak have been the most active at pushing the vaccinations.

The comments to this entry are closed.

Get new posts by email:
Kaiser Fung. Business analytics and data visualization expert. Author and Speaker.
Visit my website. Follow my Twitter. See my articles at Daily Beast, 538, HBR, Wired.

See my Youtube and Flickr.


  • only in Big Data
Numbers Rule Your World:
Amazon - Barnes&Noble

Amazon - Barnes&Noble

Junk Charts Blog

Link to junkcharts

Graphics design by Amanda Lee

Next Events

Jan: 10 NYPL Data Science Careers Talk, New York, NY

Past Events

Aug: 15 NYPL Analytics Resume Review Workshop, New York, NY

Apr: 2 Data Visualization Seminar, Pasadena, CA

Mar: 30 ASA DataFest, New York, NY

See more here

Principal Analytics Prep

Link to Principal Analytics Prep