Thread by @SGriffin_Lab

Seeing a lot about how NPIs and policies haven't influenced recent waves of SARS2 infections and it's all down to the new uncontrollable variant that we're powerless to stop...
Sound familiar? It rhymes with malfa...
Promise not to mention endemicity 🤪, but this needs a 🧵

First, a couple of caveats...
1. I'm not in any way a card carrying epidemiologist.
2. See 1.

BUT, whilst I may have slight mathematical issues, I do understand the basics...

So, why can't we control cases of BA2 whilst BA1 is on the way down? Taking the second point, some

have said that BA1 going down means that we were completely justified relaxing all restrictions, that we're probably all immune, and that the fact that even on the way down prevalence eclipses 2021 doesn't matter...
Yes, it's tempting to say this, but association≠causation

Now, I'm not going to show loads of graphs, but it's important to understand why peaks become peaks, or plateaus, or whatever...
On a simple level (my personal favourite), the exponential rise of an infection happens cos the virus is VERY good at spreading, and there are a LOT of

people to infect...we would term these folks/virus factories "susceptibles"...
So, what makes a person susceptible? Well, mainly, it comes down to immunity, resilience (age, other conditions, medicines, etc) and behaviour that affects exposure...as I'm considering infections,

I'll leave the second part...for now.

So, at the start of the pandemic, we were all without immunity, but that's obviously changed because of vaccines. However, it's important to distinguish between immunity vs infection, rather than severe disease, although one does follow the

other, as you might expect...
Anyway, back to peaks...so, whilst susceptibles aren't limiting, SARS2 or any other virus, will spread, its inherent transmissibility being the only limitation on how many become infected. This is the R value from the briefings, and standing alone it

is called R0. But, as things get bad, NPIs kick in, as does behaviour, and this acts to reduce our susceptibility, so R0 becomes Rt, which is the base reproductive value when susceptibles are limiting...
So, ultimately virus spread becomes inherently limited, but this WOULD also

happen if nothing changed as susceptibility isn't uniform either. Highly susceptible populations exhibit behaviour or other factors that make them so, but once the virus burns through these, Rt gets lower than R0, and eventually drops below 1...the peak is reached, decline begins

Some have indeed argued for doing nothing whilst hiding vulnerable people in a box somewhere...problem is that this maximal peak is potentially HUGE, and the lagged severity follows...
Some say that suppression of peaks merely delays rather than prevents infections, which is,

frankly, bollox. All medicine is about delaying illness and death!
We saw this impact of behaviour in July after freedom day (part 1), at Xmas, reducing Rt. But, similarly, EOTHO, July 21, and now, represent increased Rt...weirdly, cases grow...

But we're seeing plateaus and now ANOTHER peak...why?
Well, the answer is because multiple factors can reduce Rt, e.g increasing antibodies, masks, behaviour, environmental controls etc...and they're not there anymore.

However, mitigations, behaviour, all just chip away at Rt

, which is why you need lots of them to bring infections down, and these work together with vaccines.
The problem is that antibodies wane, as does protection from severe disease, and behaviour changes, influenced by policy. As such, susceptibility is dynamic, changing over time.

Now, it gets more fun because the "pool" of susceptibles is different for different viruses. This can be due to transmissibility (⬆️ inherent R0), antibody evasive spike (⬆️ Rt).

So, each time a variant arrives, we see rapid growth as it tears into the highly susceptible folks,

this will eventually start to slow and so you peak, then come down...

Remember, the pool of susceptibles can be reduced by behaviour, boosters etc, BUT, it can also be replenished by waning, exposing new populations, dropping measures etc.
If you get to a point where this balances Rt of the virus, you see an undulating plateau of cases. If, tho, a

New kid arrives on the block, peak 2 grows while peak 1 is declining. This can look like a plateau for a while, but our excellent surveillance means we get warnings of newly introduced strains becoming dominant. We can't afford to lose this, e.g. #ONS 🙏👍👏

Here's the lowdown IRL from @chrischirp , see yesterday's @IndependentSage for a live online breakdown with more details than I have here...

https://twitter.com/chrischirp/status/1504988503104446466?t=ZUCrTsu9aEuCZcb47V9oOQ&s=08

So, back to my point...
The dynamics involved with peaks etc are complex, but anything that increases susceptibles combined with an increased/altered Rt is a bad idea...
We're seeing an extreme example currently, total lack of protections, testing dwindling, and BA2...thankfully

many people continue to behave responsibly and altruistically, but it's not enough.
So, as we've seen before, it's both the nature AND the timing of policies, combined with incumbent viruses, that lead to surges. You cannot say whether changes were a good/bad idea until this

combination of circumstances occurs...
But, we're told this is how things have to be as we live with/ignore COVID. Sure enough, vaccines have reduced the lethality of the virus, so does this matter?
Well, unsurprisingly, yes it does!
First, the link between infection and severity

is weaker, but not broken, and severe disease has it's own pool of susceptibles, which again is dynamic. We are genetically diverse, so don't always make optimal vax responses, we haven’t vaxd enough, incl kids, and the reduced incidence of severe disease is counterbalanced by

there being huge numbers of (re)infections as new variants emerge.
You basically have a similar process going on as for cases, but the scales are smaller. Then, as well, we have the dreadful #LongCovid and #LongCovidKids which continues to build into a vast public health issue.

BA1/2 may be more of a submachine gun than a sniper rifle, but if you're losing your body armour then every bullet counts...

So, when CAN we relax protections? Well, timing is critical, as is flexibility in terms of reintroducing them. Act too late, more severe measures are

required to regain control, relax too soon and you lose what you gain from previous actions. Either way, the combo of extreme freedom with THE most antibody evasive, transmissible variant just can't be sensible, right? Focused protection doesn't work, people aren't free. Yet...

OK, I know I said I wouldn't mention it, but FYI, when these processes are in equilibrium over time, THIS is endemicity...🤪
It does not mean benign.