UK ONS Admits That They are Not Interested in Exploring Ways to Increase Data Transparency
By STEVE KIRSCH
Basically, they told me to bug off and that the hiding of record level data on the COVID shots will continue indefinitely.
Here’s the full correspondence so you can judge for yourself.
The data that they do expose is inadequate to establish whether or not the vaccine is safe. Their time-series analysis is deliberately artificially truncated to show no harm.
I explained two proven, very easy ways to make the COVID vaccine data transparent (re-running their time series with weekly buckets, and publishing the record level data with obfuscation that is guaranteed to protect privacy) and they didn’t want to talk about it. At all.
Who are they working for? In my view, the UK ONS is not serving the interests of the people of the UK. They seem to be working for the drug companies. They are not interested in exploring ways to make their data more transparent.
From: Steve Kirsch
Sent: Sunday, December 3, 2023 6:35 PM
To: sarah.caul
Subject: I’m the journalist in the United States that got the whistleblower data from New Zealand
I have the original data and I’ve spoken to the leaker who was just arrested.
I need to chat with you about what the data shows.
We did a very extensive time series cohort analysis on it. There is no ambiguity on what it shows. Yale Professor Harvey Risch, one of the world’s top epidemiologists, validated the methodology and conclusions.
We need to talk. Now. Lives are at stake.
From: Caul, Sarah
Sent: Monday, December 11, 2023 6:22 AM
To: Steve Kirsch
Cc: Andrew Bridgen; Norman Fenton; Clare Craig; Health Data
Subject: RE: I’m the journalist in the United States that got the whistleblower data from New Zealand
Dear Steve,
I am unable to comment on New Zealand data, I recommend you speak to those who work with the data.
I have seen through other communication that you are interested in receiving similar record level data for England.
The data is available as part of the ONS’ Secure Research Service (SRS). Information on the SRS and how people can access this data can be found using this link:
https://www.ons.gov.uk/aboutus/whatwedo/statistics/requestingstatistics/secureresearchservice
Hope this helps
Many thanks
Sarah
From: Steve Kirsch
Sent: Tuesday, December 12, 2023 10:34 AM
To: Caul, Sarah
Cc: Andrew Bridgen; Norman Fenton; Clare Craig; Health Data
Subject: RE: I’m the journalist in the United States that got the whistleblower data from New Zealand
Importance: High
Sarah,
Thank you for your response.
1. Regarding the SRS offer, I have chatted with Professor Fenton and Clare Craig, and the problem is that there are so many strings attached with that data. We want to take the ONS data, anonymize it in a way that protects privacy, and make it publicly accessible. Would such a project be approved? Who can I speak with on the phone about that? That would be more efficient than sending in dozens of proposals and getting them rejected each time. That would take years. How about we simply work collaboratively on defining a joint project that WOULD be approved since it is jointly designed?
2. Regarding the New Zealand data, it gives you additional insights into the safety of the vaccine. In 5 countries, we see mortality rise after EACH dose. THIS IS UNPRECEDENTED. The slope in the deaths since dose curve is always negative for any fixed size cohort, not positive. Deaths are always proportional to the # of people available to die, so the slope is always negative unless something highly unusual is going on, like a deadly vaccine. Having it be positive for a year after the dose is EXTREMELY problematic and your health authorities need to explain it or acknowledge the vaccine is killing people. Is there another agency which is looking at that and explaining it? Do you have the contact info for a person? I am not aware of anyone in the world that can explain how this can happen if the vaccine is safe. Are you?
3. Why doesn’t ONS itself make the patient record-level data publicly available just like I did in New Zealand. The technique I used resulted in not a single person finding their record or saying that their privacy was breached. I am happy to share the technique with you. Even if someone knew everything about a person except for one fact, they wouldn’t be able to find the missing information.
4. Why doesn’t the ONS run the time-series cohort analysis again, but with 5 year age buckets, but WEEKLY buckets after a dose is given. So you’d have 100 weekly buckets (instead of the 2 or 3 you use now), and a final bucket for “over 100 weeks”. This would clearly reveal that the vaccines are killing people. Your current choice of buckets today is far too crude and hides the safety signal. This should be easy and would be just a small parameter change. If the vaccine is safe, you have nothing to lose and you would reduce vaccine hesitancy. AFAIK, there is no reason you cannot do this. I’m happy to provide you with a python program that would generate the required stats. Let me know.
Finally, I acknowledge your cohort time-series analysis is an excellent way to convey what is going on, but only if you do #4 above. No other government provides this, so I applaud you for your attempt at transparency and using modern epidemiological techniques. You are way ahead of any other world government. But doing #4 would complete the effort.
I look forward to your response to my four points. Please cc: the folks on the cc: line.
Thank you.
-steve
From: BRIDGEN, Andrew
Sent: 09 January 2024 10:53
To: sarah.caul
Subject: FW: I’m the journalist in the United States that got the whistleblower data from New Zealand
Importance: High
Dear Sarah,
Steve Kirsch sent you an email on the 12th December, which he copied me into. Neither Steve nor myself have received a copy of your response, to each of the four points he raised.
Would you mind sending me a copy of the response you sent him?
Kind regards
Andrew Bridgen MP
From: Health Data
Sent: Monday, January 22, 2024 1:19 AM
To: Steve Kirsch ; BRIDGEN, Andrew
Cc: Norman Fenton; Clare Craig
Subject: RE: I’m the journalist in the United States that got the whistleblower data from New Zealand
Dear Steve,
Thank you for your email of 12 December 2023 regarding the availability of the mortality registration data linked to COVID-19 vaccination records. Please accept our apologies for the delay in responding to you.
In relation to your third point relating to the publication of patient-level record data we would like to emphasise that individual record level patient data held for statistical purposes by the Office for National Statistics (ONS) is personal information and disclosure of this is prohibited by law. Confidentiality and protection of individuals data is paramount throughout the statistical process; however, we do appreciate the need for safely and legally facilitating statistical research that serves the public good.
The individual-level records requested are available for accredited researchers to access the ONS’ Secure Research Service (SRS) and you were previously advised to access the patient-level record data via this route. More information on the SRS and how people can access this data can be found on the ONS website. As you mentioned in your first point, if your proposed project did not result in the publication of disclosive data (following the disclosure guidelines of the Secure Research Service) and could be proved as being beneficial to the public, then there is no reason why your project would not be approved.
On your fourth point, we agree that the approach suggested could have some merits. However, we do believe that the data would also need to be broken down by calendar time in order to adjust for time effects. This would likely lead to some cells having very small counts and could lead to disclosure issues if disaggregating the data by causes of deaths. In any case, if the COVID-19 vaccination was causing many deaths, then the mortality rate within 3 weeks of receiving a COVID-19 vaccination would be elevated. The data we publish do not support that narrative, as the mortality rate for people within 3 weeks of having received a COVID-19 vaccination is lower than people who have not been vaccinated.
We cannot comment on your analysis of the New Zealand data as referenced in your third point, and nor would the Office for National Statistics be the policy lead on the safety of the COVID-19 vaccine. This would sit better with the UK Health Security Agency (UKHSA) or the Medicines & Healthcare products Regulatory Agency (MHRA).
The ONS does not plan on commencing any further work at this point in time. As such, we have nothing further to add to these points.
Many thanks,
Justine Pooley | Head of Data Management, Engineering and Delivery | Data and Analysis for Social Care and Health | Health Analysis and Pandemic Insight
Office for National Statistics | Swyddfa Ystadegau Gwladol
We work flexibly at ONS. I sent this email at a time that suited me, I don’t expect you to read it or reply outside of your working hours.
Contact the DASCH Data Delivery Team at Health.Data@ons.gov.uk
DHSC, ONS, NHS England, UKHSA and NHSBSA are consulting on their health and social care statistical products. Have your say before 5 March 2024.
From: Steve Kirsch
Sent: Monday, January 22, 2024 10:10 PM
Subject: RE: I’m the journalist in the United States that got the whistleblower data from New Zealand
Justine, Thank you for your kind reply.
I’ll respond to each of the points.
- on the SRS request, you said reasonable projects would be approved. I’ll defer to Professor Fenton on this to relate his experience.
- On the NZ data, you said “wrong agency.” Fair point. Thank you.
- The patient level data can be obfuscated so that statistical analyses can be performed without violating any patient confidentiality. I have proved this with the New Zealand record level data. We used an obfuscation process that not even the New Zealand Health authorities could crack. This would provide public transparency of the data without violating UK law. We proved it works. Each and every record is randomly modified so that no data in the record matches the person’s data. So no violation is possible. The New Zealand data is proof that. I can even offer a bounty for anyone who could identify any of the New Zealand records that we obfuscated if that would help. The point is that there is a HUGE public benefit from this disclosure and a zero risk of disclosure since all dates on every record are randomly modified.
- I’m not aware of how having single death counts in a time series analysis can lead to disclosure issues. There are simply too many rows in a given month and too many counts to find out who it is, even with single age ranges. So you can easily mitigate this issue by using 5 year age ranges, limiting the largest bucket to age 95 and beyond and you’re done.
For example, from the New Zealand data which is tiny compared to the UK data, even if you used SINGLE AGE categories, there were 40 people aged exactly 95 who died sometime in Sept 2023. Some had 1 dose, some 2, some 3, some 4, some 5, and some 6, and some had no doses. So you can tell ABSOLUTELY nothing about their vaccine status even if you knew the month they died and their exact age even in the tiny country of New Zealand. Am I missing something? If you can provide me with an example, that would really help. Thanks.
My point is that with 5 year age ranges and limiting the max age to 95, there is no disclosure problem even if a time series row might have a single death count.
-steve
From: Steve Kirsch
Sent: 23 January 2024 07:38
To: Health Data; BRIDGEN, Andrew
Cc: Norman Fenton; Pooley, Justine
Subject: RE: I’m the journalist in the United States that got the whistleblower data from New Zealand (part 2)
Importance: High
I neglected to respond to “In any case, if the COVID-19 vaccination was causing many deaths, then the mortality rate within 3 weeks of receiving a COVID-19 vaccination would be elevated. The data we publish do not support that narrative, as the mortality rate for people within 3 weeks of having received a COVID-19 vaccination is lower than people who have not been vaccinated” in my response.
This statement was made as the excuse for not doing a proper time series analysis disclosure as I suggested.
This is proof that you absolutely need to do the proper time series breakdown as I did with the New Zealand data.
There are two types of Healthy Vaccinee Effect: temporal and general. The temporal HVE lasts for four weeks and artificially depresses mortality rates. So your statement that “the COVID-19 vaccination was causing many deaths, then the mortality rate within 3 weeks of receiving a COVID-19 vaccination would be elevated” is false. Completely false. By your measure, every vaccine is “safe”. The 4 week HVE effect is due to a systemic bias and it’s approximately the same value in all the countries where I have seen data on. It has nothing to do with human judgment and everything to do with policies in hospice and hospitals.
You have to look week by week after the first 4 weeks to see what the trend lines are vs. background mortality of each cohort. It’s the SLOPE differential that matters and the SLOPE tells you everything you need to know.
You don’t have weekly buckets so nobody can see the slope.
But in the New Zealand data, because we had the record level data, we can see the effect.
The vaccine is killing people.
I have a friend, Jay Bonnar. Jay knows 14 people who died unexpectedly since the COVID vax rolled out. 100% were vaccinated. He knows only one person prior to that who died unexpectedly. This is a single anecdote, but it is highly unlikely that Jay is unlucky. 4 of those friends died on the same day as they got the vax and 3 were under 30.
So the UK ONS has failed to find a signal that is clearly there. This is a huge public disservice.
It should only take a few minutes to change the code to re-run your time series with weekly buckets stretching out 100 weeks from time of vaccination and monthly rows. As I wrote below, even with single counts in a row, it is the sum of counts for a given age that matters, and even for 95 year olds in a small country in New Zealand, single counts on a row are NOT a disclosure. So the excuse of data privacy issues is a complete red-herring.
You can prove this without risk. Run the time series with 5 year buckets, weekly buckets to 104 weeks, and monthly observation windows, and a 100 year max age.
Then pay a consultant a big bonus if they can identify any person with a 1 count in the death column in a row. It’s impossible.
Then make the time series data public so the public will finally know the truth.
I know this is worth your time because I did this with the New Zealand data and the mortality signal is huge and nobody can explain it.
-steve
From: Health Data
Sent: Tuesday, January 23, 2024 3:29 AM
To: Steve Kirsch ; Health Data; BRIDGEN, Andrew
Cc: Norman Fenton; Clare Craig
Subject: RE: I’m the journalist in the United States that got the whistleblower data from New Zealand (part 2)
Dear Mr Kirsch,
Thank you for your further emails, the content of which have been noted.
As we stated in our last email, we have nothing further to add on this matter.
Justine Pooley | Head of Data Management, Engineering and Delivery | Data and Analysis for Social Care and Health | Health Analysis and Pandemic Insight
Office for National Statistics | Swyddfa Ystadegau Gwladol
Summary
They don’t want to talk about methods to improve data transparency. They want to keep the data hidden.
Are you surprised? I wasn’t.
They believe the data they have published is dispositive but the New Zealand data (that they don’t want to look at) clearly shows it is not.
They are basically refusing to do the work they should have done in the first place and they should be held accountable for their actions.
