Dear Office for National Statistics,
Please see the response to my FOI to PHE, who have suggested you would be better to answer the second half of this question
Could you provide information relating to the death certificates of all the remaining "official" COVID deaths.
Without providing any personal information such as, name, age sex, please provide all the causes of deaths that were mentioned on the death certificates.
I'd prefer this to be provided as war data per person e.g. person 1, Cause of death 1 - Covid, Cause of death 2 - car accident etc.
For the latest data on the economy and society, consult our
website at http://www.ons.gov.uk
Legal Disclaimer: Any views expressed by the sender of this message are
not necessarily those of the Office for National Statistics
Our Reference: FOI/2020/0835
Thank you for your email requesting data pertaining to list of COVID-19
deaths with all causes of death.
I am writing to confirm that the Office for National Statistics has now
completed its search for the information which you requested and the
response can be found here:
You have the right to have this response to your Freedom of Information
request reviewed internally by an internal review process and, if you
remain unhappy with the decision, by the Information Commissioner. If you
would like to have the decision reviewed please write to The FoI Team,
Room 1100, Office for National Statistics, Segensworth Road, Titchfield
PO15 5RR, and mark your correspondence "Internal Review".
If you have any queries about this email, please contact the Freedom of
Information Team ('[ONS request email]').
Please remember to quote the reference number above in any future
The Freedom of Information Team
Office for National Statistics
For information on Coronavirus, please visit ONS.gov.uk/coronavirus
Dear Office for National Statistics,
Please pass this on to the person who conducts Freedom of Information reviews.
I am writing to request an internal review of Office for National Statistics handling of my FOI request 'Reduction of Covid numbers ONS'.
Thank you for your response. As someone who deals with databases and statistical analysis on a daily basis I would argue that this is not a complex task at all. Depending on the database used it would merely be couple of lines of code to pull the requested data.
Please could you confirm the database used to store this information? I can put this question as a separate freedom of information request if it helps?
A full history of my FOI request and all correspondence is available on the Internet at this address: https://www.whatdotheyknow.com/request/r...
Good Morning Karl
Thank you for your email. I can confirm I have received and I am working on your request for an internal review. I hope to have a response to you within 20 working days of receipt. If our response should exceed this timeline, I will keep you informed.
Good Afternoon Steph,
It has now passed the 20 days in which you said you would have a reply to my internal review or keep me updated if this was to be delayed. Please could you provide your internal review as soon as possible.
As stated, I would also like you to confirm what database is used to store said information due to your reply claiming it would take a lot of work to query said database.
Good Afternoon Karl
Thank you for your email, unfortunately this is taking longer than anticipated mainly due to absences. The draft is almost complete, there are a few details I need to confirm before submitting for sign off. You should have your final response next week.
Thank you for your patience with this matter,
It has been 6 days since you said that the draft is almost complete, how long should I expect to wait for the signed off, finished version to be sent to me?
I will be extremely disappointed if this is a refusal and not all of the information I have requested, due to the fact it has taken you so long to reply, one could only assume that it has taken you longer to come up with an excuse as to why you can't provide the information than it would have taken to actually produce it.
Dear Mr Carpenter
Please accept my apologies for the delay. Attached is our response to your request for internal review.
Thank you for your reply. I appreciate that we are in the midst of a pandemic, and this causes things to take longer.
In your response you state:
"We do hold microdata showing each cause of death related to COVID-19 alongside many other variables pulled from death certificates, however we would be unable to release this in its raw format as it would identify individuals. Even with direct identifiers removed leaving just cause of death, it is still personal information that would identify individuals. Under s.39 of the Statistics and Registration Service Act (SRSA) 2007, it is a criminal offence for the Statistics Board to release information that would directly or indirectly identify an individual. We therefore find that s.44 of the Freedom of
Information Act (FOIA) 2000 applies as the disclosure is prohibited by another enactment. "
It's my understanding that there would be a total of 41,329 remaining deaths classified as an official Covid death and roughly 300,000 deaths in total (sorry I don't have the exact number for august). I'm going to assume, as you've not made it clear, that by this you mean s.39 - 1, 2 & 3:
(2)In this Part “personal information” means information which relates to and identifies a particular person (including a body corporate); but it does not include information about the internal administrative arrangements of the Board (whether relating to its members, employees or other persons).
(3)For the purposes of subsection (2) information identifies a particular person if the identity of that person—
(a)is specified in the information,
(b)can be deduced from the information, or
(c)can be deduced from the information taken together with any other published information.
If personally identifiable information has been released to the public previously along with the each cause of death on the death certificate, then I could see how that information would breach s.39.2.c. However, I do not believe this to be the case, and if it is please point me directly to said information.
Furthermore, if this is the case that the information has already been published publicly then s.39.4."Subsection (1) does not apply to a disclosure which" d "(d)has already lawfully been made available to the public" would mean that your objection under s.39 of the Statistics and Registration Service Act (SRSA) 2007 is not valid and you would have quoted s.21 of the Freedom of Information Act 2000 (FOIA).
Can you please explain how this information would make it possible for anybody to "deduce" the identity of an individual?
You also state "at least two SDC experts to look through more than 50,000 rows to check" can you explain how you have over 41,329 rows to check if each death consist of one row with multiple columns?
Thank you for the link to your website where it explains about the secure research service. I will look into this in due time. Also thanks for the link to the FOI that explain the answer to my question about the database type. which is an Oracle database.
Dear FOI Team,
I've not received a reply since 11/12/20, please could you update me with where you are with FOI.
Dear FOI Team,
Please respond to my request
Good Evening Karl,
I hope this finds you well. Apologies again for the delay, we are completely inundated with different requests at the moment. Hopefully the following information will help to answer your queries.
The reason why we cannot disclose the information you have requested is that it could be used to identify an individual or to learn something new about an identifiable individual, in both instances this would be a breach of the Statistics and Registration Service Act 2007, and therefore exempt from release under FoI. For example a newspaper report could contain the identity of someone who died under unique circumstances - that could allow that person to be identified in our dataset, and our dataset could contain additional information that wasn't in newspaper, such as an underlying cause of death. It's also possible that someone could be identified through information that has been unlawfully placed into the public domain, for example through a data leak, or information published without the proper approvals or consent. In order for ONS to be sure that we weren't breaching section 39 SRSA (which is a criminal offence) we would need to fully investigate all the information publicly available in relation to every death in the dataset to remove all those entries that we couldn't publish, and the time and effort required to do this would be extraordinary.
Apologies for the confusion over the number of rows. The 'more than 50,000 rows' we referred to in our response related to the number of deaths at the time we were responding to you when we should have referred to 41,329 which were the number of deaths involving COVID-19 at the time of your request.
Hopefully answers your questions. Again please accept my apologies for the delays.
Dear FOI Team,
Thanks again for replying. I have passed this onto the ICO and am awaiting a reference number from them.
In your response, you have stated: "For example a newspaper report could contain the identity of someone who died under unique circumstances - that could allow that person to be identified in our dataset, and our dataset could contain additional information that wasn't in newspaper, such as an underlying cause of death."
All I am asking for is a list of causes of death with no personally identifying information. If a newspaper has released the unique way someone has died with a name, and you release only the way the person died you have not released any information that has made this person any more identifiable than he already is, in fact he is still less identifiable than the newspapers own release.
Due to the fact that I am only asking for a subset of the data, being causes of death and the rest of the data you hold will not be released, you are in no way making it possible for the person to be identified. Your argument lacks merit, you should release this information to the public as you are legally bound to under FOI.
There is no conceivable chance that the data I am asking you to release could be used to personally identify someone as it would purely be a list of deaths with no other info.
We work to defend the right to FOI for everyone
Help us protect your right to hold public authorities to account. Donate and support our work.Donate Now