I have been looking at data on sequenced variants. I have also been thinking about our vaccination programme and India. The Indian variant needs to trigger surge testing in England.

First published in April 2021.

Some dates: while over 60% of adults have had one dose of vaccine, we have got another 15 weeks before all adults could have immunity from at least one dose. And another 20+ weeks before all adults could be fully vaccinated. And we don’t yet know what uptake will be in younger adults.

So, what can variants do while we are vaccinating people?

We already know that B.1.1.7 can infect some partially vaccinated people and (far, far) fewer – but still some, fully vaccinated.

I am concentrating on five variants in England: our dominant B.1.1.7 (“Kent”) variant, B.1.351 (“South Africa”), B.1.525, P1 (“Brazil”) & B.1.617 (“India”).

The Sun newspaper has a good graphic explaining different variants:

Coronavirus variants. | The Sun

The UK Covid Genetics Consortium (COG) have sequenced about 10%+ of cases a week chosen to be representative. Now they sequence up to 20,000 cases a week. With the current numbers, this means they are sequencing most cases (much more than 10%).

This chart shows the cumulative number of sequenced cases for the five variants. Our B117 (Kent) variant grew fast between September and Christmas, before being controlled by the January lockdown. The fastest growing new variant – especially as it started during the lockdown – is the Indian one (B1617).

Cumulative number of sequenced samples of different variants in England. | Data from COG UK

I have now looked at just the four new variants as a % of overall cases. The South African variant (B1351) has been kept to small numbers by a combo of lockdown and surge testing. It started increasing after schools opened. The most recent spike might be the big South London cluster and associated surge testing.

Proportion of cases in England that each of four new variants (does not include B117 ("Kent") which makes up 98%-99% of cases). | Data from COG UK

The South African variant is thought to be bit less infectious than B117 (good for us) but can infect some vaccinated people, although no indication that it makes them very sick. But it is worrying and warrants the surge testing and close tracking.

B1525 and P1 have struggled to get a foothold thankfully – especially P1 (Brazil variant) which is causing havoc in South America and in British Columbia, in Canada.

However (showing same plot again), B1617 (India) has gone from under 0.2% to over 1% of cases in two weeks. It is doubling in number every week – similar to B117 back when it started growing.

But B1617 is doing it under much tigher restrictions and more vaccinated people.

Proportion of cases in England that each of four new variants (does not include B117 ("Kent") which makes up 98%-99% of cases). | Data from COG UK

Because B1617 (India) is a “variant under investigation” it does NOT trigger surge testing or forensic contact tracing. This is because there is no definitive proof that it is either more infectious or better at immune escape than B117 (Kent).

BUT here is the problem.

We know that it is dominant in some (but not all) of India, going from around 20% to around 60% of cases in the Maharashtra state in a few weeks. This article digs a bit more into what is worrying about B1617.

We definitely know that India is going through a massive surge.

Daily new confirmed Covid-19 cases in India. | OurWorldInData

India sequences fewer than 1% of cases and many who have had Covid-19 never got (or get) a test. Only 8% of its population have had one dose of vaccine. So, definitive evidence of either more infectiousness OR infecting people who have had Covid-19 or a vaccine is likely to take a while.

In fact, because we are very good at sequencing, we might be the first country to provide such definitive evidence. But by the time we do, B1617 is likely to be quite widespread and it might be too late to contain it.

For more stories like this, sign up to our FREE Briefings today.

Note that what follows is based on very low overall numbers and so the estimates of growth rate are very, very uncertain. That said, it is doubling every week or so right now – similar to B117 when it first took off.

At current rate of growth, we could be seeing 1,000 B1617 cases/week in 4-5 weeks. Still not massive – but very hard to contain without more restrictions. It could be dominant by the end of June – 10-12 weeks away. It depends on what is driving its growth.

If many of our cases are imported (implied by Dr Susan Hopkins today), then its growth in India could be causing the growth here. BUT then India should definitely be on the red list! As highlighted in this flight to Singapore:

If it is more infectious than our B117 but susceptible to vaccines then B1617 will likely cause us (maybe big) problems (basically a worse case of the SAGE Spi-M models) but it is self-limiting – by the autumn almost all adults could be fully vaccinated.

Note: If B1617 is susceptible to vaccines, then as we vaccinate more and more people this will limit its growth – but this might be offset by the extent and speed to which we open up and more people mix – at least over the next 8-10 weeks.


So finally – worst case scenario – if B1617 can infect vaccinated people, it could be much worse. If vacinated people can pass it on more but not get too sick, it will mean more unvaccinated people getting sick this summer (because it can spread through more people).

If B1617 can escape immunity more completely (better than the South African variant), then it could cause severe illness even in those who have been partially or fully vaccinated. That means our whole vaccination programme is damaged until booster vaccines are available. It means new lockdowns likely.

Now, this last scenario is definitely the least likely – but it is not impossible. And letting B1617 spread (potentially alongside the South African or other variants) also risks new, potentially more dangerous, variants emerging.

We are (rightly) treating P1 (Brazil) and B1351 (South Africa) variants as real threats to our vaccination programme. We must treat B1617 as a threat too and start surge testing it here and stop new importations from abroad.

Waiting for proof might be waiting too long.

India has now been added to the red list travel ban countries – this is good. Mind you, the whole red list philosophy is very leaky but better than nothing definitely.

We absolutely need to know how many of the most recent data are incoming travellers and how many community transmission. If most are travellers we should see numbers fall after the red list addition (which would be welcome). But why not for another FOUR days?!

Proportion of cases in England that each of four new variants (does not include B117 ("Kent") which makes up 98%-99% of cases). | Data from COG UK

It feels as if everyone is desperate to say that Covid is over.

I get that – I cannot cope with the idea of another lockdown either. But ignoring the very real risks that remain in the UK’s future is the worst way of preventing more lockdowns.

Daily new confirmed Covid-19 cases in the world. | OurWorldInData

Globally this is far from over

  • PS. Last week it was reported that there were 77 cases of B1617 in the UK. The latest COG figures now show 215 cases. 179 new cases in the last 4 weeks.
Covid-19 variants confirmed in the UK, as of 20 April 2021. | COGUK

  • PS2. And this:
  • PS3. In terms of vaccine escape, there is this:

Going Further:

Professor Christina Pagel, Professor of Operational Research, Clinical Operational Research Unit, University College London. Member of Independent SAGE.


[This piece was first published as a Twitter thread and turned into the above article on 20 April 2021 with the purpose of reaching a larger audience. It has been minorly edited and corrected, and published with the author’s consent. | The author of the tweets writes in a personal capacity.]

(Cover: Pixabay. / Licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.)

Creative Commons License