Tuesday, January 24th, 2023, 1:05 pm
Office for National Statistics Responds to My Query About Undercounted Deaths, But Does Not Answer the Question
arlier today I noted that the Office for National Statistics (ONS) hadn’t replied within the time frame or ‘window’ specified in their automated reply. I therefore prodded ONS again, politely, and within hours I received the following response:
Good morning Dr Schestowitz
Thank you for your email. Apologies for the delay in responding, we have a large volume of enquiries at the moment.
Deaths in the last week of December and the first week of January are impacted by bank holidays where registry offices are closed. These deaths are then usually registered in the following week.
This isn’t an undercount, rather we see a lower number of death registrations on weeks with a bank holiday followed by a larger number of death registrations the week after a bank holiday. So the numbers will stay the same, we just ask people to use caution when comparing trends as they need to be aware of this short-term displacement effect that happens across weeks with bank holidays and the week after.
Kind regards
Anne
Anne
Customer Services Team
Health Analysis and Pandemic Insight (HAPI)
Office for National Statistics | Swyddfa Ystadegau Gwladol | www.ons.gov.uk | @ONS
I quickly responded:
Hi Anne,
Thank you for the reply.
The reply does not address my question however. I did not ask about the last week of December or bank holidays.
My query was about the week ending 9th of December. I took a screenshot of your site: (PNG file)
In your own words, there was a “processing issue”, resulting in “undercount of death occurrences”. It said the figures for week 49 would be “published in the next weekly mortality publication”, but in all subsequent weeks after that the number was the same and not corrected. In other words, the number you gave while stating “undercount of death occurrences” is still an undercount almost 1.5 months after the week in question. This needs to be corrected or clarified, otherwise it is a contraction that lessens quality in your data.
Please can you explain what happened to week 49?
Kind regards,
Let’s see if this time they will actually answer the question rather than address something I did not ask at all.