select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to preprint looking at thromboembolic events and thrombosis with thrombocytopenia after COVID-19 infection and COVID-19 vaccination

A preprint, an unpublished non-peer reviewed study, looks at thromboembolic events and thrombosis with thrombocytopenia after COVID-19 infection and after vaccination with Pfizer-BioNTech or Oxford-AstraZeneca COVID-19 vaccine in Catalonia, Spain.  


Dr Peter English, Retired Consultant in Communicable Disease Control, Former Editor of Vaccines in Practice, Immediate past Chair of the BMA Public Health Medicine Committee, said:

“This is an excellent paper.  It used “standardised incidence ratios” (SIRs) to compare adverse events (“outcomes”) in people with different exposures (vaccination with Pfizer-BioNTech or AstraZeneca Covid vaccines; Covid-19 disease).

“”Standardised” in SIR means that the rates were adjusted so that differences in e.g. age and sex can be accounted for, leaving the exposure itself to be the main cause of any differences in incidence rates.

“A very large number of participants (“945,941 BNT162b2 [Pfizer BioNTech vaccine] (778,534 with 2 doses), 426,272 ChAdOx1 [AstraZeneca vaccine], 222,710 COVID-19, and 4,570,149 background participants”) means that the results are likely to be very robust.

“They found no cause for concern with regard to arterial thromboembolism (ATE) or thrombosis with thrombocytopenia syndrome (TTS) (although this could be because such incidents are extremely rare – which is, in itself, very reassuring), but they did find very slightly increased rates of venous thromboembolism (VTE) in vaccine recipients.  VTE is more common than ATE and TTS, so you would expect to see more of such events; and thus to be more easily able to detect an excess in people who e.g. receive a vaccine.

“The rates of VTE were, however, only slightly raised: “SIRs for VTE were 1.29 [95% CI 1.13-1.48] and 0.90 [0.76-1.07] after first- and second-dose BNT162b2, and 1.15 [0.83-1.58] after first-dose ChAdOx1.”

“So the risk of VTE following vaccination might be slightly raised after the first dose of the Pfizer-BioNTech vaccine (although it is not raised after the second dose, and is not significantly different after a first dose of the AstraZeneca vaccine in this study – although there is a possibility that a larger study might show a slight rise).

“The bottom row of Table 2 on page 27 of the paper seems to show that the expected number of cases of VTE in a population of 938,381 was 157.7.  In other words, the background rate of VTE – the “absolute risk” would be about 16 cases per 100,000; and an SIR of 1.3 means that this would rise to 22 cases per 100,000 – still a very low absolute risk.

“The big question, of course, is “how safe is being vaccinated compared to not being vaccinated”.  And the authors report that “The SIR for VTE in COVID-19 was 8.04 [7.37-8.78].”

“Given that the SIR for VTE following Covid-19 disease is a lot higher than for vaccination with either vaccine, and that the chances of catching Covid-19 if you’re not vaccinated (particularly now that the far more transmissible delta variant is circulating widely), it still seems likely that you are more likely to have a VTE if you choose not to be vaccinated (because you are more likely to catch Covid-19) than you are if you are vaccinated with the Pfizer-BioNTech vaccine.”


Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“As a statistician, I can’t comment on the medical aspects of this work, but there are some cautions about the meaning and interpretation that I can comment on.  First, it’s important to understand that this report doesn’t just concentrate on the specific blood clotting issues that led to the Oxford/AstraZeneca vaccine not being recommended for use in younger people in several countries.  Those issues were about blood clotting (thrombosis) in conjunction with low levels of platelets in the blood (thrombocytopenia), which is referred to as thrombosis with thrombocytopenia syndrome (TTS) in this new report, but is sometimes referred to by other sets of initials (VITT or VIPIT) in other reports.  The new study does report data about this syndrome, and the researchers did not find evidence of increased risk of TTS with people who had had the Pfizer/BioNTech vaccine.  But, because this condition is so rare, it did not arise often enough in their sample of people who had had the AZ vaccine to allow a conclusion.  Therefore, on TTS, the findings of this new study aren’t very conclusive because the condition is very rare, but, taking that into account, they do not go against previous findings on TTS.  Most of the results in the new study are about other types of blood clotting event.

“Second, this is an observational study, as is the case for most studies of vaccine effectiveness and safety after vaccines come into regular use.  There are differences between the people who received the two different vaccines (Pfizer and AstraZeneca), between them and the general population that was used to find the baseline rates of the blood clotting events that were studies, and between all of these and the population of people who got Covid-19, that are used in the report to compare the risks of blood clots after vaccination with the risks after a Covid-19 infection.  Some of those differences are reported in the preprint, and some are pretty large.  That’s particularly true in terms of age (because, as in the UK, generally vaccines went to older people first, but also because the recommendations on which age groups should have the AZ vaccine changed over time and were considerably different from the UK position).  But it also applies to pre-existing medical conditions, where more people on both vaccines had conditions than was the case for the general population, but there were many more such conditions on average for people who had the Pfizer vaccine than the AZ vaccine.  These differences could, in part or in whole, be the cause of any increased risk on blood clotting in vaccinated people, or any differences between the two vaccines.  The researchers did make adjustments to take into account the age (and gender) differences, but not the differences in pre-existing conditions.  If the differences in the patterns of pre-existing conditions existed simply because of the age differences, the adjustment to take account of age might have been enough, but we don’t know that’s the case, and there may well have been other differences between the groups that were also not taken into account.  All this means that the study cannot tell us whether any differences, or lack of differences, between people using the two vaccines or between vaccinated people and the general population, are actually caused by the vaccines.  They might be, but they might not.  The researchers do acknowledge this, by writing, “These safety signals must be interpreted with caution with further investigation required to confirm causality.””



Preprint (not a paper): ‘Thromboembolic Events and Thrombosis With Thrombocytopenia After COVID-19 Infection and Vaccination in Catalonia, Spain’ by Edward Burn et al. was posted online on Tuesday 20 July 2021.  This work is not peer-reviewed.



All our previous output on this subject can be seen at this weblink:



Declared interests

Dr Peter English: “Dr English is on the editorial board of Vaccines Today: an unpaid, voluntary, position. While he is also a member of the BMA’s Public Health Medicine Committee, this comment is made in a personal capacity.”

Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee.  I am also a member of the Public Data Advisory Group, which provides expert advice to the Cabinet Office on aspects of public understanding of data during the pandemic.  My quote above is in my capacity as an independent professional statistician.”

in this section

filter RoundUps by year

search by tag