Agree, testing is the big issue in Australia and only reporting deaths in hospital from Coronavirus. Asymptomatic cases in oz are estimated at 3-5 times higher than known… work colleagues who have family spread across China report that large numbers in regional towns have died of covid19 and infection was and is rampant. Estimates anecdotally for infection across China are up at 3-5 million - accurate or not, it fits in with the rampant spread we are seeing in the US and Europe and is more likely given limited testing.
2 Likes
I confess I assumed that there was not too much delay in collating and publishing the Covid death figures.
Certainly there is under-reporting of deaths because those in the community are not being included - and some of the figures are bound to take a while to reach central government but there is an element of things not adding up.
Their estimate is a true number of deaths of 463 for March 29 - those numbers should have appeared as “new cases” about two weeks ago - but new cases on the 15th were 330. Obviously not every patient is going to be exactly 14 days from diagnosis to death but the offical new case numbers around the 15th were around the 300/day mark and the Guardian article suggests deaths were actually only just under 300 for the 26th, 27th and 28th as well - these numbers just are not there in the new case figures from two weeks ago.
I’m not sure how to explain the discrepancy, one possibility would be that some diagnoses of community cases is being made post mortem - but I am not aware there is any post mortem testing going on - perhaps samples taken just before death coming through with results after death is the most likely explanation.
The official figures might be an underestimate, but let’s assume that they are a consistent underestimate - to get 463 deaths on March 29 you would estimate at least 15,400 cases around 2 weeks prior (based on a true case fatality rate of 3%) - the official No of cases March 15th was 330 but using a small averaging window of +/- 2 days gives 272 new cases - scale that back up to now and it would suggest the true “new case” figure for yesterday should be closer to 250k - pretty frightening.
I don’t think that the Guardian article is necessarily wrong - in fact I think that it is probably correct - but I think it mainly illustrates the fact that it is very difficult to get accurate data in what is still a very fast developing situation.
Just a momentary real example: my 93-year old uncle in law died last week after 4 days in hospital, having been taken in due to breathing difficulties. They posthumously tested, and it turns out he had the coronavirus: I imagine this is the procedure in most hospitals, so he is sadly now one of the many represented in these charts. If he had passed away at home, my understanding is that these figures are now being added to the UK figures at least, and will be back-adjusted when someone has the time to sort it out.
I am really sorry to hear about your uncle.
3 Likes
Likewise - I am sorry for your loss.
There seems to be some opening up about the state of Coronavirus in French care homes - with the admission of 17827 previously uncounted cases and 884 additional deaths (although remove the extra new cases and you are left with 5233 which is still a considerable increase on the day before).
New cases in the UK continue to give cause for cautious optimism - down to 3735 today (with the caveats discussed above as to how accurate that figure might be). This is good because, frankly, I think our “lockdown” is a bit limp but perhaps we are doing enough and we will see a sustained fall. There is worry that good weather at Easter will encourage people to leave their homes when they should be staying in.
Deaths follow the expected trend at 708 - it will be interesting to see if the number of deaths tracks the slight fall in new cases from March 21st to 22nd and 23rd.
It is still unclear whether and how much the Wuhan experience maps to the UK one - I see hints in the press (but can’t find any underlying data) that average UK hospital stays are in the 10-16 days range, rather than the 18-22 day range in China. I suspect that people are further into their illness and sicker before we make the diagnosis hence the shorter period of admission, it also goes with the lag between the shape of the new cases and deaths graphs being 14 days rather than 18-20 days which you’d expect from the Wuhan data.
Comparing the current cumulative deaths with cumulative new cases 14 days ago suggests mortality up to 80% (4313 deaths vs 5108 cumulative cases to 21st March), locally we were reporting1 37 deaths2 to 48 discharges a couple of days ago which is 43.5% mortality.
My suspicion is that daily deaths will continue to rise for the next 14 days at least though if we are over the peak new cases some of the pressure might ease on the NHS.
The suspicion remains, of course that the “confirmed” cases and those in hospital are the tip of a very big iceberg.
1] the data is on the web so I don't think I'm letting out any thing confidential here2] there is slightly more up-to-date info, also online, about deaths but I don't have the number of patients discharged to go with it.
Thank you - so sweet.
Not a nice way to go, but he was probably ready, as he had other problems, (although very
compos mentis to the end.)
Keep safe
Regards
3 Likes
Apparently France has done even less testing than the UK.
This suggests differently…
Slighly more last time I looked at worldometers.
But It has to be said that there’s no obvious correlation between tests per million and deaths per million, despite all the talk about test test test.
The data for testing in France is very sparse. It looks to be in the same order as testing in the UK though whereas Italy & Germany have performed 3-4x as many tests. You can spin the data all sorts of ways. Germany has done lots of testing so shows low mortality because lots of people with milD or no symptoms (the majority) have been confirmed positive whereas in France & UK only sick patients in hospital have been tested so the mortality appears much higher. What are we to make of Italy with lots of testing (much more than France & UK) but much higher mortality?
Older population?
At least that was one explanation doing the rounds.
Italy has the oldest population in Europe and second only to Japan in the world, third though is Germany which is faring much better so who knows whether population age is important.
It’s not clear how the two nations stack up but I believe the approach in both has been to mainly test symptomatic individuals to make a diagnosis rather than undertake massive contact tracing and blanket testing.
France did appear to have a better ratio of deaths to positive cases up to 14 days ago, which suggested a better pick up rate but there was not that nearly perfect visual correlation between the two graphs to follow and work out exactly what the lag time from diagnosis to death is in France. If you are diagnosing more cases earlier than the UK it might be that it will be more like the 18 days or so that was seen in Wuhan.
Unfortunately the sudden injection of that data has made the French numbers very difficult to interpret - I suspect that the UK needs a similar injection but we haven’t had it yet.
At the moment I’m just holding my breath and hoping that new case numbers start to fall in both nations.
It’s the USA which worries me though - their response has been so delayed and haphazard that it makes ours look organised and measured - I think that there will be massive loss of life over there.
The way the country is structured is a hindrance during a ‘national’ crisis, even at local county level there are differences in the restrictions, for example one will say beaches are out of bounds but a neighbouring one will allow access, totally bonkers.
I have at last got round to reading Vera Brittain’s testament of Youth and I am at the point where she, working as a VAD Nurse in London, has had to clear her ward of convalescents and there are rows of empty white beds waiting for the first casualties of the First Battle of the Somme.
So prescient.
1 Like
Rather disappointing figures from the UK today - a further 5903 confirmed cases taking the total to 47,806 - not sure if there is any community data coming through but if that is still just hospitalised cases it is the largest daily jump.
France also reported 7788 new cases on Saturday which is also somewhat up on the very hopeful-looking numbers from the first two days in April.
UK deaths up 619, smaller rise than yesterday as (half) expected - the way deaths continue to track new cases with a 14 day lag is uncanny.
Hate to say it but expect about 31.5k deaths over the next 14 days.
Interesting paper on the UK ITU experience with Covid19 patients - 50% mortality which goes with this graphic I spotted on Twitter explaining the various clinical courses very well - I suspect that we are diagnosing patients around day 7/8 with a heavy bias towards “critical symptoms” (fits with death around day 21/22 and the 14 day lag we see between the new case and deaths curves).
The breakdown of patients from UK Critical Care admissions is interesting.
35% non-white
75% male
75% overweight
1 Like
Well I’m male & overweight so f****d if I get it.
1 Like
I suspect that 75% overweight is just a typical cross section of the UK population. . . .
6 Likes