Hi Chris
I think there it is inevitable that with the way the statistics are counted there will be overreporting AND underreporting.
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/908781/Technical_Summary_PHE_Data_Series_COVID-19_Deaths_20200812.pdf
“The PHE data series does not include deaths in people where COVID-19 is suspected but not confirmed by testing (SARS-CoV-2 PCR either negative or not done).
Furthermore, the PHE data series does not report cause of death, and as such represents deaths in people with COVID-19 and not necessarily caused by COVID-19.”
Personally, if I had drafted this I would have written “In contrast” instead of “Furthermore”, since the first of these paragraphs in the Public Health England text
- does not count deaths where COVID is a suspected cause but has not been confirmed by laboratory test [so probably indicative of underreporting – and given the shortcomings in arranging tests in the UK, particularly in the early months of the pandemic, possibly indicative of very significant underreporting]
- but does include deaths where COVID has been confirmed but may not be the cause [so probably indicative of overreporting, though it must be very difficult for clinicians to exclude COVID as a cause if someone has been in intensive care for say 21 days, and seems to have recovered for the initial reason for being there]
In ye olden days, if you could show that something could be shown to have caused death that occurred with a year and a day, this counted as a fatality in British law, subject to the expert opinion that there is a causal chain all along that time period.
So, as example, if someone sustained an amputation in an accident at work, then an issue in hospital resulted in blood poisoning, and then may be a stroke, you might be able to connect the death from stroke to the initial amputation.
Given the incredible levels of improvements in healthcare, I find the 28 (or 60 day) threshold somewhat arbitrary. Hospitals put people into medically induced comas for well over 28 days when the circumstances dictate.
The good news in the PHE paper (yet to be peer reviewed) is that they quote very high percentages of direct links to back up their statistical approach.
However, I was quite cynical when the method of counting was changed to enable the official total number of COVID deaths in the UK to be slashed – the “precautionary approach” might have suggested changing the approach taken in Scotland, England and Wales to up the numbers, rather than reducing the numbers in England to be compatible with the method of counting elsewhere in the UK.
What will happen when Public Health England is axed?