select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
The Office for National Statistics (ONS) have released provisional counts of the number of deaths registered in England and Wales in the week ending 14 January 2022.
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“I’d say the key feature in relation to Covid is that Covid-related deaths are up compared to the previous week, and that Covid deaths form a greater percentage of all deaths, in the latest week, than they did the week before.
“However, these short-term trends are pretty well impossible to make good sense of, given the time of year that the latest figures are for. (And the same goes for all-cause deaths.) They are all based on death registrations, and registration figures are distorted by holidays when registration offices are closed, and to a lesser extent by other things being closed and people generally being on holiday.
“That said, deaths according to the 28-day definition on the government dashboard did also increase between the week ending 7 January and the week ending 14 January, in England, and didn’t change much in Wales. That’s looking at the figures by the date the person died, not when it was reported. So, despite the usual issues about the 28-day definition, that does make me think the numbers or Covid-related deaths did really increase between those two weeks. Also, the fact that the Covid-related death registrations made up a greater percentage of all death registrations in the week ending 14 January than in the week ending 7 January does also make me think the numbers probably really did go up – though that percentage could have gone up because the number of deaths that weren’t caused by Covid went down, rather than Covid deaths going up.
More detail on bank holidays causing registration issues:
“The week before the latest one was the week ending 7 January, the first seven days of the year. That included a bank holiday on Monday 3rd, when the registry offices would have been closed, so reducing the number of registrations that would have happened that week, in relation to the number of people who actually died that week. However, the number of registrations that week would, however, also tend to be increased, compared to the number of actual deaths, because there were also two bank holidays the week before (the week 25-31 December) and registrations of deaths that happened that week would have been moved to the following week (ending 7 January) more than happens in non-holiday weeks. And it’s impossible to tell, so far, whether the increase from late registrations or the decrease from the New Year bank holiday would have been bigger.
“Therefore any comparison between the latest week and the week before is uncertain, because the figure of registrations from the week before is already potentially misleading. But on top of that, the registrations in the most recent week, ending 14 January, would be relatively high in relation to the number of deaths that actually happened that week, because some registrations in the week ending 7 January would have been shifted to the week ending 14 January because of the bank holiday on 3 January. (ONS say in the bulletin that registrations in the first two weeks of January tend to be untypically high.) We will eventually get counts of deaths, based on registrations, but counted according to when the person died, not when the death was registered, but so far those numbers are incomplete because of late registrations.”
Dr James Doidge, Senior Statistician, Intensive Care National Audit & Research Centre (ICNARC); and Honorary Associate Professor, London School of Hygiene and Tropical Medicine, said:
“Weekly deaths in England and Wales (combined) have recently dropped below the five-year average for the first time since June. Combined with the declining number of patients being admitted to ICU with COVID-19 since the emergence of Omicron1, stable total number of adults in hospital2, and declining total number of children in hospital3, these indicators paint a surprisingly positive picture of Omicron in the UK. Could it be that by outcompeting Delta4, the net impact of Omicron has been positive? While the number of patients in hospital with COVID-19 has increased and there are still deaths involving COVID-19 occurring, it can be unclear how many are “with” versus “because of” COVID-19. The total numbers of deaths or patients in hospital provide important indicators of the overall impact of the pandemic that are unaffected by this classification. Currently, these indicators are all pointing in the right direction.”
[1] https://www.icnarc.org/Our-Audit/Audits/Cmp/Reports
[2] https://www.england.nhs.uk/statistics/statistical-work-areas/covid-19-hospital-activity/
[3] https://www.statsjamie.co.uk/children-in-hospital/
Prof Sheila Bird, Formerly Programme Leader, MRC Biostatistics Unit, University of Cambridge, said:
“Tracking COVID-mention deaths by week of occurrence matters. Registration delays are longer in England & Wales than in Scotland. In the vaccine-era, tables on COVID-mention deaths by week of occurrence should heed ONS Infection Survey’s prevalence estimates some 2 weeks earlier. Hence, at England’s Alpha peak (during 27 Dec 2020 to 2 January 2021), 2.08% (95% CI: 2.0% to 2.2%) of the population tested positive for COVID-19; in the Delta-era, 0.61% (95% CI: 0.5% to 0.7%) of the population had COVID during 27 June to 3 July 2021 vs 1.44% (1.3% to 1.5%) during 26 September to 2 October 2021 with Delta peak (during 16 to 30 October 2021) at 2.02% (95% CI: 1.9% to 2.1%). In the Omicron era, England’s prevalence reached 2.83% (95% CI: 2.7% to 3.0%) during 13 to 19 December 2021 and had more than doubled to 6.85% (95% CI: 6.6% to 7.1%) during 31 December 2021 to 6 January 2022.
“In England & Wales, wave 1 accounted for 52,000 COVID-mention deaths (to nearest 1000), the peak weekly toll having been 8,350 deaths (nearest 10). Wave 2 in England & Wales accounted for 69% more COVID-mention deaths (88,000 to nearest 1000) with a weekly peak of 9,060 deaths (nearest 10). Wave 3 for England & Wales dates from mid-June 2021 but is shown below from the week ending 17 September 2021.
“With both pupils and universities returned by 1 November 2021, and as autumn turned to winter, caution was again needed – as COVID-mention deaths which occurred in the last week of October and first week of November will have exceeded 1000 per week.
“Booster vaccinations and Omicron both accelerated thro’ December 2021, with Omicron dominant pre-Xmas with England’s proportion testing positive already higher during 10 to 16 December than the Alpha and Delta peaks. Resurgence in COVID-mention deaths is evident from the week ended 31 December 2021 with a near-doubling in occurrences likely between the week ended 24 December and 7 January 2022 (when fully registered). Continued caution in January 2022 is prudent. When fully registered, the likely number of COVID-mention deaths that occurred in the week ended 14 January 2022 is 1400 to 1600, substantially higher than the weekly-peak in October 2021 of around 1000 deaths.”
All our previous output on this subject can be seen at this weblink:
www.sciencemediacentre.org/tag/covid-19
Declared interests
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.”
None others received.