Comparing Covid-19 Deaths UK and Italy.

Today I heard Keir Starmer making political capital out of the rising death toll in the UK. He is not alone. He has been joined by many in the MSM in telling us (glee barely concealed) that the  UK death toll from CV-19 exceeds that of Italy.

This is not a contest. But it irritated me enough to make me look into the numbers myself.

The data/sources and my reasoning are below for each country. Mostly this is based on discrepancies between total death count this year and the rolling 5 year average.

The main takeaways from this post as as follows:

  • It appears the deaths in Italy from CV-19 are being massively under-reported. The under-reporting is almost by a factor of two. It also appears that no attempt has been made to rectify this problem to-date. So I believe that today (6th May) Italian CV-19 deaths are still heavily under-reported.
  • CV-19 deaths in the UK were originally under-reported as they failed to take into account care-homes and the community. I understand this was corrected in late April. The date range in this article pre-dates this change so I have compensated for care home deaths and community deaths and included them along with hospital deaths. 
As of today (imho) I reckon the numbers stack up more-or-less correctly for the UK. But not for Italy.

I fail to see how the MSM (let alone the Leader of the Opposition) cannot know this.

Incidentally (post posting update!) Reuters newsagency report HERE more-or-less concurs with my figures. In fact (hurts to say!) they beat me to it by a couple of days.

Italy

I am relying on most of the figures for Italy from this BBC ARTICLE. (Yes. The MSM's own data!)

Here the BBC states that from 20th February through to “end of March” (I assume that is and includes 31st March) Italy suffered 90,946 deaths from all causes.

All causes is CV-19, heart attacks, strokes, cancer, accidents and everything else.

The rolling five year average for all-cause deaths during the same period in Italy is 65,592.

Which gives an excess death count of 25,534 (from 20th February up to & including 31st March) this year.

So how many deaths were actually flagged up as being caused by CV-19 during this period?

Sadly the BBC leave that bit out. (And why did they do that I wonder.)

From https://www.worldometers.info/coronavirus/country/italy/  we get this graph:

Incidentally this concurs with the Statista graph HERE:

The number of deaths recorded from CV-19 during 20th February to 31st March in Italy was 12,428

This is less than half the total excess deaths recorded from all causes.

As the Italian health service was at times overwhelmed, it is likely that some of these extra 25,000 mortalities were from causes other than CV-19 (Heart attacks/strokes etc) that went unattended.

But it would seem highly likely that only a small proportion of  these extra 25,000 deaths were from non CV-19 causes.

I would suggest that the excess deaths from non-covid-19 causes would be about at maximum 15%. That would give us an excess death count caused by Covid-19 of about 21,250 over the period 20th February through to 31st March in Italy. Not 12,428 as reported.

So it would look like the Italian death count for Covid-19 from 20th February through to the end of March is under-reported by around 9000 to 10,000.

So do these missing 10,000 deaths get added later on?

That looks impossible. Look at the curve on the worldodometer graph. While a few deaths will always be reported late there is simply no room to accommodate 10,000 pre 31st March deaths in another later time period without significantly changing the graph shape.

In fact the graph shape suggests that this counting scheme used up to 31st March is still in place.

If that is so then the real death toll from Covid-19 in Italy up to today (6th May) is more like 40,000 – 50,000.

UK.

The BBC article unfortunately does not give hard dates for the UK period of “6 weeks” but states that it started from the first UK CV-19 virus death. This death occurred on the 28th February. Six weeks later is 10th April.

(Incidentally this is two days longer than the equivalent Italian period - but never mind)

The BBC informs us that the total number of deaths from all causes during this period was 89,735. The rolling five year average for total deaths during this calender period is 72,812. Which gives us an excess death total of 16,923.

The NHS total of CV-19 deaths (NOT the full total) up to 10 April is given in the graph below as 10,760. (from HERE)



So at this point we are left with 6,000 excess deaths unaccounted for.

This weekly summary from the ONS for deaths during week ending 17th April (HERE) indicates that up to week 16, 77.4% of deaths from CV-19 occurred in hospital. Which would mean that 22.6% occurred in care-homes and the community.

Although the statement is first reported in the PDF for week 16 (17th April) rather than week 15 (10th April) it would seem reasonable that the percentage death by location would be very similar on a weekly basis.

So if 10,760 deaths from CV-19 occurred in hospitals up to 10th April it militates that the real death count from CV-19 (including care homes and the community as well as hospitals)  was 13901.

Which now leaves an excess of just over 3000 unaccounted for.

Although the NHS has not been overwhelmed (unlike the Italian health service was) it has delayed/cancelled/postponed procedures.

People were originally encouraged not to go to hospital. Consequently people were either frightened of catching CV-19 or simply did not wish to burden the NHS further with their CV-19 unrelated problems. Hospital attendances fell off a cliff. Some died as a result.

If (like Italy) we assume 15% excess deaths are due to extra heart attacks/stroke deaths we get down to 915 unaccounted deaths over this 6 week period.

This is small enough to get lost in the noise of rolling 5 years averages. But even if you assume they actually were unreported Covid-19 deaths then that would give an under-reporting in the UK of a mere 6% compared to Italy’s under reporting of 70 – 100%.

Then finally we come to population size and density. The UK has a 10% larger population than Italy. UK population density is 30% higher than that of Italy.

So I think it would be reasonable to say that the deaths from CV-19 per capita in Italy are twice as high as those in the UK.

But this is not a contest.

This is a tragedy.

Maybe Mr Starmer should keep that in mind next time he goes to make political capital out of the misery of his countrymen and women.

No comments: