COVID killed middle aged people in spring. It is not doing that anymore. I have scaled total deaths (by 20) to compare. Thanks to @RuminatorDan for inspiration.
Same chart not scaled.
Same for older people over all time and recently

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Dr Clare Craig FRCPath

Dr Clare Craig FRCPath Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @ClareCraigPath

22 Oct
In spring, the death rate per week was double normal for the over 55s. What is happening now?
Recent COVID deaths appear to have risen since week 36.
It is worth comparing this pattern to the pattern of % positive tests. The rise in positive tests starts in week 36. There has been no lag between this rise in positive tests and resulting rise in % of COVID deaths as proportion of mean 2015-2019 deaths.
Read 5 tweets
20 Oct
ONS data showed that 0.52% of the East Midlands population were COVID positive on 5th Oct that's 24,293 cases. PCR testing is meant to be reliable for 20 days during disease. In the 20 days between 26/09 and 15/10 there have been 23,032 cases diagnosed.
The mean number of daily cases diagnosed was 1152 and 1065 were found on 4th Oct and 1448 on 5th so, although there has been some growth in that 20 day period the 5th is the median amount.
In a false positive pseudo epidemic the data starts to look back to front because the relationships between e.g. predicted cases and diagnosed cases is lost. Both factors become functions of testing variables not disease.
Read 4 tweets
20 Oct
Here is the ONS data showing where people with COVID have died. The disproportionate deaths in hospital recently is in stark contrast to spring. Given we are screening the care home population as well as the hospital population I am surprised it is so stark. Image
Here is same graph for total deaths. It shows that deaths in hospital are still 1000 every week below normal levels. Image
Good to see care home deaths remain below baseline despite it being Autumn and flu season.
Read 4 tweets
17 Oct
We have some excellent laboratories in the UK but they have been under immense pressure to ramp up testing numbers. The more testing the more false positives from volume alone. But people under pressure are more likely to make errors too. 3 ways we have made the situation worse:
1. Testing ramped up from 100,000 in May to 304,415 PCR tests a day today. Even the best labs will have false positive results. Addenbrookes has calculated theirs at 44% by repeat testing. This excludes false positives due to cross contamination, cross reactivity, RNA shedding.
Read 7 tweets
17 Oct
Firstly, I have nothing against @neildotobrien who I am sure is well intentioned but badly advised
The reason for going back to normal is because the epidemic has passed. It’s over. We are now in the midst of a false positive pseudo-epidemic.…
Read 23 tweets
17 Oct
This looks scary. If you look at all deaths and compare to hospital admissions 10 days earlier things look like they're getting worse. There are four possible explanations for this: a) we have got worse at treating COVID over since the summer Image
b) people are dying at home of COVID now whereas before they died in hospital
c) in the summer there were false positive COVID 'cases' but now it's back and it's killing people again
d) something else ?
Let's repeat but differently. Instead of the 10 day lag til deaths let's look at the ratio of admissions and deaths on the same week. That looks much less scary. Image
Read 6 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!