Excess Deaths Latest

[FabioPorkpie]

I would say it is more likely that people die from complications caused by having had and recovered from covid or beyond 28 days that in some figures are being rejected as not covid deaths.

Yep, I agree it works both ways with the 28 day cut off meaning a fair few won’t be included in the numbers.

ONS wouldn't count car crash victims. Approximately 90% of covid-19 positive deaths had that on their death certificate. Not all <28 day positives are included in death certificates (about 95%).

I hadn’t registered the significance of it the first time you said it
The 5% of positives not included will likely cover the, for want of a better term, car crash etc numbers.

 

[bear66]

Yep, I agree it works both ways with the 28 day cut off meaning a fair few won’t be included in the numbers.


I hadn’t registered the significance of it the first time you said it
The 5% of positives not included will likely cover the, for want of a better term, car crash etc numbers.

Or the doctor didn't believe Covid-19 was the cause of death in spite of positive test.

 

[Billy Horner]

Agree, but has a situation arisen whereby some deaths are being attributed to covid when they clearly shouldn’t be. Is there a manipulation of the figures occurring to some degree? (For why, I do not know.)
If a person is involved in a car crash and comes in with heavy internal bleeding and broken limbs, and tests positive for covid, then subsequently dies as a result of their injuries, are they classed as a covid death or a road traffic accident death? Granted it’s possibly a small number like this, but how blurred are the lines getting?
For a long time during the first wave I wrongly took the covid death figure as meaning these people had died because of covid, when that clearly isn’t always the case. I’m sure I wasn’t alone in that and possibly the majority of the public read it that way. It sensationalises the number and adds to a narrative of fear as well as, most importantly, clouding the discussion about what exactly is happening, in my opinion.
That said I know it’s a very difficult thing to establish what is and is not directly attributable to covid so perhaps this is just the best we can do at the moment.

There are just a lot of uncomfortable questions to be asked around the whole issue.

As Bear has pointed out, the statistically correct way of measuring deaths is by cause of death listed on death certificates. This is what the ONS uses in its weekly publication.

However, there is a delay of at least two weeks between point of death and that death being certified, registered and then included in the ONS publication. For that reason, the deaths within 28 days of a positive test metric is used as a proxy measure of the current situation.

There’s also been a lot of nonsense written about people testing positive for Covid, recovering but then dying from a car crash or heart attack within 28 days of the test. Whilst this might be technically possible (and I’m sure it may have actually happened in a very small number of cases), I’ve seen no evidence that it is a significant problem in the data. Given that the metric is designed to give us a reasonable indication of numbers, trend, etc., even a few examples of this would not fundamentally change the scale of the reported Covid deaths.

The real problem with the 28 days metric is not that it over-reports the number of Covid deaths, but rather that it under-reports them. If we compare the two measures, the 28 days metric currently reports 44,347 deaths in the UK, whilst the latest ONS figures (which are the gold standard) state the number is 58,164 (and that’s as at 9 October).