Reports that the AstraZeneca COVID-19 vaccine causes blood clots has been dominating the news.
In mid-March, several European countries paused the distribution of the AstraZeneca COVID-19 vaccine following reports of blood clotting disorders in people who had received the jab.
Update – As of the 20 May, Australia had recorded 21 confirmed cases of blood clots following administration of 2.1 million doses.
The Australian government has now mirrored this restriction, with ATAGI recommending the Pfizer vaccine over the Astrazeneca vaccine in adults under 50 years of age.
In the UK, young people are now being offered an alternative COVID-19 vaccine as a result of 168 people suffering a blood clot shortly following administration of the AstraZeneca jab.
This sounds like a scary number, but what does in mean in context?
What are the stats of blood clots appearing after a COVID-19 vaccine?
Of 21.2 million doses of AstraZeneca given in the UK by April 14th 2021, there were 168 cases of blood clots, and 32 deaths resulting. That’s approximately 8 cases per million, or 0.0008%. Most of the cases were of clots in the brain.
Compared to the number of cerebral (brain) venous sinus thrombosis (clots) normally expected in a general population – five cases per million – the vaccine-related clots are very similar to what is expected.
Deep vein thromboses in general, which commonly occur in the leg and can travel to the lung, are more likely to affect adults (500 cases per million in adults annually).
These stats are not indicative of every country.
For example, by April 9th, Germany had a higher ratio of clots to dosage: there, 31 people developed blood clots out of 2.7 million doses (11 per million).
The causal links between AstraZeneca and the blood clotting side effects are still under investigation, and there currently aren’t hard stats about which demographics are most likely to be affected.
In Australia, three criteria must be observed soon after an AstraZeneca administration, for the clot to be consisered linked to the vaccine:
- Evidence of a thrombosis (blood clot)
- Thrombocytopenia (blood platelet count below a certain threshold)
- Results of blood tests for the D-dimer protein produced by the body to break down clots and anti-PF4 antibodies that activate platelets.
By May 20, the most common clot types were deep vein thrombosis and portal vein thrombosis. Two cases were of clots in the brain. The most common time for onset was 14 days following administration, but had a range of 2-44 days. All but one of these are recovering and stable.
The median age of linked cases in Australia was 66, which makes it sound like the side-effects are more common in the elderly. However, we need to bare in mind that vaccination is still only readily available to the eldery, medical staff and some vulnerable groups, and is not reflective of a normal population
Likewise, inititally women under the age of 60 appeared to make up the largest portion of cases in Germany, because 29 of the 31 cases were women in the early days. However, this doesn’t reflect an average population, because the first people to receive first-stage vaccinations were healthcare workers, most of whom are women.
“If you look at who was being vaccinated in the European Union with the AstraZeneca vaccine back in January, February and early March, there is about a 2-to-1 ratio of women to men,” says Peter Arlett, the European Medical Association’s head of pharmacovigilance and epidemiology.
This is a good lead in the investigations as to whether there is a causal link between AstraZeneca vaccine and blood clots, but it doesn’t portray a true at-risk group.
A helpful example might be the link between sunburn and ice cream. Sunburn may appear to occur after people eat ice cream. Eating ice cream doesn’t cause the sunburn; rather, ice cream is eaten more frequently by people out in the sun. So ice-cream eaters aren’t necessarily a genetically or physiologically ‘at risk’ group when it comes to sunburn, even if they get sunburn more frequently.
It’s always wise to weigh up the risk of taking any medication, including vaccinations during a pandemic, but the numbers require context to understand.
Is the risk worth it?
The risk of contracting COVID-19 is significantly lower in Australia (1729 cases per million) or Vietnam (27/million) than the UK (64,075/million) or Germany (34,856/million). These numbers are further influenced by the state/town/region one lives in, because coronavirus cases are not uniformly spread over a country.
Disregarding the global deathrate of COVID-19 (30,000 per million cases), a study published in JAMA in February showed that the long-term effects of the virus were prevalent even among people who were considered to be fit and healthy.
“The effects of COVID-19 can linger far beyond acute infection, even in individuals who experienced mild illness,” said author Denise McCulloch of University of Washington.
“To our knowledge, this study presents the longest follow-up symptom assessment post-illness, with individuals surveyed out to 9 months after their COVID diagnosis,” she told MedPage Today.
“Our study is unique in characterising a group consisting of mostly outpatients: 90% of our cohort experienced only a mild COVID-19 illness, yet one-third continue to have lingering effects.
“Many of these individuals are young and have no pre-existing medical conditions, indicating that even relatively healthy individuals may face long-term impacts from their illness.”
Ultimately, such findings have led to the European Medical Agency recommending that the risk of COVID-19 is greater than the risk of AstraZeneca vaccine side-effects.
What do the numbers look like?
Note: The following numbers in this article were updated on April 22 2021.
Numbers can feel very big if they’re compared to numbers we’re familiar with. For the following examples, assume that the UK stats are broadly reflective of vaccine-related blood clots.
With an average eight blood-clot cases per million in the UK, the risk of blood clots from the AstraZeneca vaccine is much lower than the risks of blood clot from an oral contraceptive pill (400 per million in Australia), pregnancy (2000 per million) or severe COVID-19 itself (about 31% of people admitted to the ICU, or 310,000 per million ).
So, compared to the AstraZeneca vaccine, the chance of blood clots is about 50 times greater for the pill, 2500 times greater for pregnancy, and 38,750 times greater from COVID-19 infection.
Let’s look at that in a different way by equating these stats to distance instead of volume.
If we say that that one blood clot is equivalent to 1 metre, then the length of AstraZeneca blood-clot cases per 1,000 km would be the equivalent of 8 metres (about two Volkswagen Beetles, or as tall as one and a half giraffes).
For progestin birth-control pills, that would be 400m (a little under four soccer pitches, or the length of a full-lap Olympic sprint, or eight laps of an Olympic swimming pool), and 2km for pregnancy (the north-south length of Adelaide’s CBD).
For COVID-19 related blood clots, it would be 310km – a little more than the road distance from Sydney to Canberra.
Looking at it in yet another way, let’s compare it to the current stats about COVID-19 cases in Australia.
Based on the UK percentages, if all Australians contracted COVID-19, 7.7 million of them would get COVID-related blood clots, and 750,000 would likely die from the disease (based on global averages). On the other hand, if every Australian got the vaccine, 200 would develop blood clots – less than the number of people who’ve already died in Australia from COVID-19 (909 people from 29,000 cases).
Risk of event happening (per million)
|Blood clots from AstraZeneca vaccine (UK stats)
|Anaphylaxis from a vaccine (US stats)
|Struck by lightning this year (US)
|Winning an Olympic gold medal
|People over 110 in Australia (2012)
|Fatal animal related farm injury each year (US)
|Twin birth (ABS, Australia)
|Triplet birth (ABS)
This is ultimately why national health authorities continue to say that the benefits of the AstraZeneca vaccine outweigh its risks.
You may also be interested in: