Read on Twitter
Let me tell you about a different test. The COVID antibody test. This has demonstrated that only 7% of the UK population had antibodies to COVID in May (it was 17% in London).
We know that people who had COVID in spring got these antibodies. So does that mean 90% of us remain susceptible as SAGE says. It does not. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/925854/S0769_Summary_of_effectiveness_and_harms_of_NPIs.pdf
This is a measurement error. Let me explain. With every test there will be definite negatives, definite positives and a grey area in between. There is a choice about how to classify the grey area.
You can choose a test that finds every possible positive (sensitive) or one that only finds the definite positives (specific). For a pregnancy test a sensitive test would find all the pregnant women but might find women who were not pregnant and even the odd man (false positives)
A specific test would find only definite pregnant women and would miss women in early pregnancy. The COVID antibody test is a specific test. But it is very good at telling us who had COVID in spring.
However, we want to know who had immunity to COVID in May. That is a totally different question. On this question the test fails us.
It fails because when the test were designed the manufacturer’s had to pick a group to test negative. They picked pre-COVID blood samples. These patients could not have had COVID.
However, by design, any patients who were already immune to COVID (because COVID has similarities to the common cold) would test negative on this test. This is where they put the threshold for positive vs negative.
Thankfully other studies have been done which are better at answering the question – how many of us are immune to COVID? Two studies lowered the threshold. The whole grey area of antibody levels seen in both pre-COVID blood and COVID blood was called positive.
This was the same test – for IgG to the COVID spike protein – but used in a more sensitive way. These studies showed 42% of the population of Ischgl in Austria had antibodies and 45% of 200 members of staff at UCL hospital.
https://www.medrxiv.org/content/10.1101/2020.08.20.20178533v1.full.pdf https://www.medrxiv.org/content/10.1101/2020.06.08.20120584v1
https://www.medrxiv.org/content/10.1101/2020.08.20.20178533v1.full.pdf https://www.medrxiv.org/content/10.1101/2020.06.08.20120584v1
This is fairly high considering that the spike protein is a unique part of COVID but our immune systems are brilliant at learning things from one infection that can be applied to a new infection.
There is an even more sensitive way of testing. Instead of only looking for the red spike protein. Public Health England grew the whole virus in culture and looked for antibodies to all the COVID virus proteins.
They used a threshold that has been used for other antibody testing. This was a very sensitive test. It will exaggerate the answer because it will have false positive results e.g. antibodies that do not help kill the virus. Let’s see what it showed.
Public Health England tested all the staff and residents in six care homes in London. Of the people who remained asymptomatic throughout (i.e. did not have COVID disease) over half had COVID antibodies. https://www.thelancet.com/journals/eclinm/article/PIIS2589-5370(20)30341-2/fulltext
Some tested PCR positive so virus may have attacked them but they had no symptoms. This is immunity. Immunity is when our own defences deal with invaders without us noticing. So half of these people were immune.
This is an interesting figure. The maximum household transmission seen globally was also 50%. So half of contacts living in a house with someone with COVID did not catch it. It also fits with all other evidence of prior immunity.
https://www.medrxiv.org/content/10.1101/2020.07.29.20164590v1.full.pdf
https://www.medrxiv.org/content/10.1101/2020.07.29.20164590v1.full.pdf
If half of us were immune anyway and 7% of us had it then we are not at 57% with immunity. That is a very important figure. 60% approximately is the number predicted to reach herd immunity. 5 mins into this video
Antibody levels by government testing have not shown an increase since May. If we have had real COVID cases since May and if we are not at herd immunity then these levels would have increased.
Why is London at 17% then? COVID resulted in more severe disease and higher antibody levels in non-white people. Those of Asian ethnicity had double the antibody levels of white people and people of black ethnicity had three times the levels.
That is not enough to explain the discrepancy though. There are parts of the country where COVID hit first and we had not all changed our behaviour. These areas seem to have had a sharper rise in COVID deaths. When that happens deaths can overshoot.
This happens because on the day herd immunity is reached lots of people have already caught it. Those susceptible will die. In areas where spread was slowest, at the point herd immunity was reached the minimum number had caught it and deaths were minimised.
A hint of this effect can be seen when looking at deaths in the first 30 hospitals to reach peak deaths and the last 30. But some of this difference may be due to the size and density and other characteristics of the populations those hospitals serve.
COVID was real and lethal and killed many susceptible members of our communities in spring including truly tragic cases. However, we are now at herd immunity. COVID has not gone. It will return in outbreaks in a seasonal way like flu.
But it is not a pandemic anymore. We cannot change our lives forever for it. Plus the changes we have made are causing non-COVID deaths: https://twitter.com/ClareCraigPath/status/1327126149898645506?s=20
There have been lots of concerns about levels dropping. Antibodies rise and fall depending on the job at hand. In addition, we know that these antibodies (IgG to the spike protein) cross react with other human proteins. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7246018/
Our bodies remove such antibodies as they are a danger to us and can cause autoimmune disease – where the immune system attacks healthy tissues.
Finally, the same antibody test the Government are using was used in Wuhan (in a specific way). Only 3% of hospital workers had these antibodies. This is does not mean only 3% of them are immune it means only 3% had COVID. https://www.nature.com/articles/s41591-020-0949-6
We would expect lower levels in South East Asia because of immunity to SARS1 giving the population higher prior immunity than elsewhere.
@threadreaderapp unroll
