Thursday 3 March 2022

Official mortality data for England reveal systematic undercounting of deaths occurring within first two weeks of Covid-19 vaccination

Our new research (full paper with first author Clare Craig available on ResearchGate) has discovered over 26,000 covid and non-covid deaths expected to have occurred within two weeks of first dose vaccination have been omitted from the latest UK ONS deaths by vaccination status report:

 Estimated deaths for England that were not included in the ONS dataset


The ONS data also fails to include millions of people categorised in the UK NIMS database as ‘within 21 days vaccination’ status:

Our analysis compares expected deaths in the ONS dataset to those that have occurred in the whole population. We only use data from UK government sources including data from NIMS and UKHSA. We found implausibly low number of non-covid deaths in the ‘within 21 days of first dose’ category:

The ONS dataset only includes the number of non-covid deaths equivalent to the number expected to have occurred in the third week after vaccination only. So the expected deaths for the first two weeks post first dose vaccination are missing. This is the case for all age groups 60+. For example, here is the 70-79 age group:


For covid deaths the same patterns are evident, across all age groups 60+. This is a significant and rather troubling anomaly warranting an explanation. Covid deaths are also implausibly lower than expectations:

Here the expected covid deaths occurring in the 80+ age group during the first two weeks are missing from the ONS data:

Note that the peaks in the graphs are synchronised with the vaccine roll out for each age group and hence are not natural or due to random error. Errors appear systematic and strongly suggest that deaths have been miscategorised as hypothesised in our last research paper.

A variety of factors could have led to deaths being omitted, including miscategorisation, reporting lags and data handling or transcription errors:

Vaccination record data flow highlighting potential sources of error

Hence, we believe that the dataset is corrupted, making any inferences about vaccine efficacy or safety, reliant on the data, moot.

We have presented these findings to the ONS and they have acknowledged in correspondence that our analyses are valuable and important. They have also promised to release all raw data to allow independent analysis and to investigate the serious anomalies reported by our paper.

Clare Craig has published a twitter thread with more insights from the research.

Full details of the new paper:

Craig, C., Neil, M., Fenton, N., McLachlan, S., Smalley, J., Guetzkow, J., Engler, J., Russell, D., Rose, J. (2022). "Official mortality data for England reveal systematic undercounting of deaths occurring within first two weeks of Covid-19 vaccination".



Why are UKHSA obfuscating data on stillbirths by vaccine status: just another statistical illusion?

(See 22 April Update below)

In the latest UKHSA Feb 2022 Covid-19 vaccine surveillance report, Figure 8 suggests that there is no increased risk of stillbirth from vaccination:

This seems like good news but, as pointed out by Hamish Soutar,  women vaccinated prior to pregnancy are included in "no doses in pregnancy".

This is an outrageous piece of obfuscation.  Even though the rest of the report contains quite a lot of detailed raw data, there is no raw data provided to answer the simple question:  

 is the stillbirth rate higher for those unvaccinated than those vaccinated (before or during pregnancy)? 

As implied by Hamish, Fig 8 does not answer this question. To give a feel for how easy it is to arrive at figure like that even if stillbirth rates were significantly higher in the vaccinated, consider the following hypothetical example for women whose pregnancies ended, say, between April 2021 and Feb 2022 (UPDATE see video below for visual explanation):

Suppose 10,000 women received at least one jab pre-pregnancy and 10,000 women received no jabs pre-pregnancy, and that these further break down into the following 4 categories:

  • Cat 1.    (unvaccinated pre pregnancy and unvaccinated during pregnancy): 2000
  • Cat 2.    (unvaccinated pre pregnancy but at least one jab during pregnancy): 8000
  • Cat 3.    (at least one jab pre pregnancy but unvaccinated during pregnancy): 8000
  • Cat 4.    (at least one jab pre pregnancy and at least one jab during pregnancy): 2000

Suppose the stillbirth numbers are:


Total women

Number of stillbirths

Stillbirth rate

Cat 1




Cat 2




Cat 3




Cat 4




So we are assuming in our purely hypothetical example that the stillbirth rate is higher in each of the categories containing vaccinated women.  Then we can present the results as follows:



Number of stillbirths

Stillbirth rate

Never vaccinated (Cat 1)




Vaccinated at least once (Cat 2+Cat 3+Cat 4)




..which answers the earlier question and clearly shows the increased risk of stillbirth in the vaccinated.

...Or, we can do what UKHSA have done in their Figure 8, which ‘shows’ that there is no increased risk of stillbirth from vaccination:



Number of stillbirths

Stillbirth rate

Unvaccinated during pregnancy (C1+C3)




Vaccinated at least once during pregnancy (C2+C4)




All women (C1+C2+C3+C4)




In fact, it is not hard to produce hypothetical examples in which (with the same stillbirth rates assumed above but different numbers in the categories) the vaccine appears to reduce risk of stillbirth using the UKHSA ‘method’. 

UPDATE: Here is a 4 minute video explaining the above:

22 April 2022 Update:

The latest (21 April 2021) UKHSA covid surveillance report  continues with the same obfuscation. As a very timely example of the problems with the UKHSA reporting, today one of Britain’s leading athletes who won gold at the Tokyo Olympics in August 2021 revealed she suffered a miscarriage in November 21 and an ectopic pregnancy in mid January 2022. She says (in the comment attached to her latest photo)

“Since the Olympics we haven’t had much luck and it’s been the hardest few months I’ve ever had to go through. Jason and I fell pregnant immediately after the games and we were absolutely chuffed to bits. But unfortunately in November when commentating at the track champions league I miscarried our baby at 9 weeks. I’ve never felt so lost and sad. It felt like a part of me had been torn away. I grabbed for my safety blanket, bike riding! I found myself back in my happy place training again. I then caught Covid in mid January and found myself feeling really very unwell. I didn’t have typical covid symptoms and I just felt I needed to go to hospital. A day later I found myself in A&E being rushed to theatre because I was having an ectopic pregnancy. Scared doesn’t even come close. I lost a falopion tube that day. I’ve always known I was tough, but sometimes life pushes you to an unbearable limit."

She would have to have got double vaxxed shortly before the games to be allowed to take part. While there may be no reason to suspect the vaxx had anything to do with these two unfortunate outcomes, both would be classified in the UKHSA "no doses in pregnancy" category