Big thanks to Michael Story for bringing this to my attention, particularly this tweet and his discussions of herd immunity in the UK. I'd never have written about this if it wasn't for him.
This post is meant to be informative, as a lead in to my next one, on how the UK's pandemic response went wrong - I've tried to keep it as neutral as I can.
There has been a lot of confusion about what exactly the UK's government strategy was for the coronavirus pandemic, at least until the 23rd of March 2020, when the first lockdown was announced. What was the strategy in the weeks and months until then?
The UK followed its influenza pandemic plan
The Scientific Advisory Group for Emergencies (SAGE) provides scientific and technical advice to support government decision makers during emergencies.
SAGE met on the 4th of February to discuss the coronavirus; in that meeting (pdf) they
agreed that HMG should continue to plan using current influenza (aka the flu) pandemic assumptions, which can be modified as data becomes more certain
But what were the "current influenza pandemic assumptions"?
(You might find it odd that SAGE thought that the government should respond to the coronavirus with a plan for influenza, which is a different disease. This can be justified to an extent, since they didn't know anything about it initially, but it will prove important in the future.)
The influenza pandemic plan didn't try to minimise the number of deaths, and said it was impossible to stop a virus from spreading
We can find those assumptions here (pdf), in the "UK Influenza Pandemic Preparedness Strategy 2011". which
describes the Government's strategic approach for responding to an influenza pandemic
First, what did the government want their flu pandemic strategy to achieve? What was their goal?
You might have thought, like I naively did, that it was to minimise the number of deaths from influenza, or at least the number of infections, but you'd be wrong, The UK government's goal was not to minimise the number of deaths from the coronavirus:
The overall objectives of the UK’s approach to planning and preparing for an influenza pandemic are therefore to:
i. Minimise the potential health impact of a future influenza pandemic by:
- Supporting international efforts to detect its emergence, and early assessment of the virus by sharing scientific information.
- Promoting individual responsibility and action to reduce the spread of infection through good hygiene practices and uptake of seasonal influenza vaccination in high-risk groups [me: this wasn't possible, since the first person wasn't vaccinated against the coronavirus into the UK until the 8th of December, 10 months later].
- Ensuring the health and social care systems are ready to provide treatment and support for the large numbers likely to suffer from influenza or its complications whilst maintaining other essential care.
ii. Minimise the potential impact of a pandemic on society and the economy by:
- Supporting the continuity of essential services, including the supply of medicines, and protecting critical national infrastructure as far as possible.
- Supporting the continuation of everyday activities as far as practicable.
- Upholding the rule of law and the democratic process.
- Preparing to cope with the possibility of significant numbers of additional deaths [emphasis mine].
- Promoting a return to normality and the restoration of disrupted services at the earliest opportunity
To me, this sounds like the goal was to live with the virus as much as possible, apart from "Promoting individual responsibility and action to reduce the spread of infection through good hygiene practices".
Note, they specifically mention
Preparing to cope with the possibility of significant numbers of additional deaths
If anything, that is closer to the opposite of "minimise deaths".
But why did they have such an odd goal, and not try to minimise the number of deaths, like you'd expect? Because the government thought it would be impossible to stop it from spreading:
it almost certainly will not be possible to contain or eradicate a new virus in its country of origin or on arrival in the UK. The expectation must be that the virus will inevitably spread and that any local measures taken to disrupt or reduce the spread are likely to have very limited or partial success at a national level and cannot be relied on as a way to ‘buy time’
It will not be possible to halt the spread of a new pandemic influenza virus, and it would be a waste of public health resources and capacity to attempt to do so.
This is the most important point in this entire blogpost - if they did not have this assumption, February to March could have been very different.
This is why the government did not attempt to stop it from spreading. We now know that these assumptions were completely wrong for the coronavirus.
They believed this because
Modern mass global transit also affords opportunities for the virus to be rapidly spread across the world, even before it has been identified. The short incubation period of influenza means that within a relatively short period of time a significant number of cases will appear across the globe. It is likely to take at least four to six months after a novel virus has been identified and isolated for an effective pandemic influenza vaccine to become available from manufacturers
The influenza pandemic plan implied hundreds of thousands of people would die
The UK thought it was inevitable the coronavirus would spread throughout the country, and couldn't be stopped. On the 18th of February, a paper was released saying that the infection fatality rate (not the case fatality rate, which is percentage of confirmed cases that die) was 2.3% - so, of 100 infected people, 2.3 of them would die. We now believe this was an overestimate, and an IFR of 1% is closer to the true value.
The influenza pandemic plan said
For deaths, the analysis of previous influenza pandemics suggests that we should plan for a situation in which up to 2.5% of those with symptoms [me: note, not "of those infected"] would die as a result of influenza [emphasis mine], assuming no effective treatment was available
(the IFR/CFR isn't a hard number that applies to every infection, it doesn't mean if you get infected, you have a 1% chance of dying - CFRs range from close to 0% in the 0-19 age group to ~15% in over 80s, treatments are discovered which reduce the risk of death, etc.).
So, the coronavirus would infect most people in the UK, and kill 1% of them. How many is "most people"?
Viruses stop infecting people (in the absence of measures to prevent spread) when herd immunity is reached, either through infection or vaccination. The virus has infected so many people that it can't find new hosts to infect.
We can calculate the herd immunity threshold - if each infected person on average infects 3 others (which is a reasonable guess for the coronavirus), the virus will in the end infect 1 - 1/4 = 3/4 of the population. 75% of the population will be infected - the UK had ~67 million people in 2020, so it's reasonable to say 50 million people would be infected by the UK's strategy, and 500 thousand of them would die.
Even if the virus infected half as many people as guesstimated there, and killed half as many as it turned out to, that would still be 125 thousand deaths - at the time of writing, there have been at least 121,674 deaths.
The UK government did not realise this until around the 16th of March, when some modelling was released by Imperial College London.
They did not multiply "the number of people that will get infected" by "the percentage of infected people that die".
This is the point Tomas Pueyo is making in this interview with him and John Edmunds, a member of SAGE, on Channel 4 on the 13th of March. The interviewer says "I don't think anyone is saying [we want to kill 200 thousand people in the UK]", but that is what Edmunds implies when he says "the only way to stop this epidemic is indeed to achieve herd immunity", he just doesn't realise it.
Edmunds, along with many others, was at the SAGE meeting on the 4th of February (pdf) when using the pandemic influenza plan was endorsed.
What did the government actually plan to do?
Relating to their above-mentioned goal of "Promoting individual responsibility and action to reduce the spread of infection through good hygiene practices", the measures the government planned to implement (that were relevant to the coronavirus, e.g. "antivirals and antibiotics" were not, since no antiviral or antibiotic treatments were available in March 2020):
Reducing the spread of disease: infection control & respiratory and hand hygiene
To protect others and reduce the spread of infection, anyone ill with pandemic influenza should [emphasis mine]:
- Stay at home.
- Minimise close contacts.
- Adopt thorough respiratory and hand hygiene practices, i.e. covering the nose and mouth with a tissue when coughing and sneezing, disposing immediately of that tissue after use, and washing hands frequently with soap and warm water, or alcohol gel if water is not readily available.
Facemasks and respirators
Facemasks and respirators have a role in providing healthcare worker protection, as long as they are used correctly and in conjunction with other infection control practices, such as appropriate hand hygiene
Although there is a perception that the wearing of facemasks by the public in the community and household setting may be beneficial, there is in fact very little evidence of widespread benefit from their use in this setting
International travel, border restrictions and screening
The Foreign and Commonwealth Office will issue advice regarding travel to affected countries. There are no plans to attempt to close borders in the event of an influenza pandemic. The UK generally has a high level of international connectivity, and so is likely to be one of the earlier countries to receive infectious individuals. Modelling suggests that imposing a 90% restriction on all air travel to the UK at the point a pandemic emerges would only delay the peak of a pandemic wave by one to two weeks. Even a 99.9% travel restriction might delay a pandemic wave by only two months
Passengers should be encouraged to self-report symptoms to crew and ground staff to enable information gathering, investigations and treatment to be undertaken
This was essentially it.
The government wanted to delay infections, not minimise them
On the 3rd of March, the government released a "coronavirus action plan" - in it, they said the UK aimed to
minimise the potential health impact by slowing spread in the UK and overseas, and reducing infection, illness and death
See, the goal has now changed to include "reducing infection, illness and death" - initially, at the start of February, they instead wanted to "[prepare] to cope with the possibility of significant numbers of additional deaths", something very different.
Despite this, it's not right to say the government's goal was to minimise deaths. It's more correct to say it was to minimise deaths, given that the virus would inevitably infect most people in the UK. Essentially, "most people in the UK are going to get infected, what can we do to minimise the deaths that result?"
If the disease becomes established in the UK, we will need to consider further measures to reduce the rate and extent of its spread. Based on experience with previous outbreaks, it may be that widespread exposure in the UK is inevitable; but slowing it down would still nonetheless be beneficial.
For example, health services are less busy in the summer months when flu and other winter bugs are not driving GP consultations and hospital admissions. In the 2009 ‘swine flu’ pandemic school holidays significantly slowed transmission of the virus.
Part of the plan was
Delay: slow the spread in this country, if it does take hold, lowering the peak impact and pushing it away from the winter season
this was because
if the peak of the outbreak can be delayed until the warmer months, we can reduce significantly the risk of overlapping with seasonal flu and other challenges (societal or medical) that the colder months bring.
The Delay phase also buys time for the testing of drugs and initial development of vaccines and/or improved therapies or tests to help reduce the impact of the disease.
It isn't fair to say the government "wanted to infect everyone", or that "herd immunity" was the strategy
The government did not want to infect everyone - they didn't want herd immunity through infection.
Instead, at this point, it was taken as given that most people in the UK would get infected - they assumed that herd immunity through infection was going to happen, regardless of what they did.
"Herd immunity" and "flatten the curve"
Graham Medley is the chairman of the Scientific Pandemic Influenza Group on Modelling (SPI-M), and on the 16th of March he told the Atlantic:
"People have misinterpreted the phrase herd immunity as meaning that we’re going to have an epidemic to get people infected"
He says that the actual goal is the same as that of other countries: flatten the curve by staggering the onset of infections [emphasis mine]. As a consequence, the nation may achieve herd immunity; it’s a side effect, not an aim. Indeed, yesterday, U.K. Health Secretary Matt Hancock stated, “Herd immunity is not our goal or policy.”
On the 13th of March, Patrick Vallance, the government's Chief Scientific Adviser, said
Our aim is to try and reduce the peak, broaden the peak, not suppress it completely; also, because the vast majority of people get a mild illness, to build up some kind of herd immunity so more people are immune to this disease and we reduce the transmission
So, we see the UK's goal at this point was to "flatten the curve" of infections - this means, for example, (using the numbers from my previous example) 50 million people in the UK will get infected not in a month or two, as could happen without government interventions, but instead spreading out the infections so that maybe 5 million people would get infected each month, for 10 months.
The same number of people get infected, just over a longer period of time.
As Story says, it's "[spreading] out the infections up to the herd immunity threshold to keep the peak below hospital capacity".
Story again made a very important point - with the "flatten the curve" strategy, the number of infections per day begins to fall because we've hit the herd immunity threshold.
Remember above when John Edmunds, the SAGE member, said "the only way to stop this epidemic is indeed to achieve herd immunity".
So, "flatten the curve" isn't really different from "herd immunity".
"Protect the NHS" was added as a goal later on
Another reason they wanted to "flatten the curve", aside from pushing the highest number of daily infections into the summer, was to stop the NHS from being "overwhelmed" - the Prime Minister, on the 23rd of March, when announcing the first lockdown, said this:
I want to begin by reminding you why the UK has been taking the approach that we have.
Without a huge national effort to halt the growth of this virus, there will come a moment when no health service in the world could possibly cope; because there won’t be enough ventilators, enough intensive care beds, enough doctors and nurses.
And as we have seen elsewhere, in other countries that also have fantastic health care systems, that is the moment of real danger.
To put it simply, if too many people become seriously unwell at one time, the NHS will be unable to handle it - meaning more people are likely to die, not just from Coronavirus but from other illnesses as well.
So it’s vital to slow the spread of the disease.
Because that is the way we reduce the number of people needing hospital treatment at any one time, so we can protect the NHS’s ability to cope - and save more lives.
Here, you can still see the influence of the "It will not be possible to halt the spread of a new pandemic influenza virus, and it would be a waste of public health resources and capacity to attempt to do so" assumption in his statement - he talks both about slowing the spread and "[halting] the spread of this disease"; they're not the same thing.
"Protect the NHS" isn't the same as "minimise deaths"
The goals of "don't let the NHS be overwhelmed" and "minimise the number of deaths" aren't the same. Even if the NHS isn't overwhelmed, people will still die - roughly 1% of those that get infected - remember, 121,674 people have still died so far, despite the NHS not being "overwhelmed".
If you focus on not overwhelming the NHS, and the NHS isn't overwhelmed, a large portion of the people in hospitals will still die - 39% at the start of the pandemic. "Being admitted to hospital" isn't the same as "in ICU", but 426,243 people have been admitted to hospital since last March - multiplying those 2 together gives you 166,234 deaths, an overcount compared to reality.
Despite the many deaths we have had in the UK, the NHS hasn't officially been overwhelmed - under the "don't overwhelm the NHS" goal, we have succeeded.
So the "don't overwhelm the NHS" goal corrupts the "minimise deaths" goal, succeeding at the former can fail the latter. If you didn't have the first one, only the second, you'd have a lower number of deaths.
(also, whether or not the NHS is "overwhelmed" is not a binary yes/no question, it can be gamed. A sensible definition of "overwhelmed" could be "we can't treat everyone we want to", but if you deny people care in the first place who you otherwise would have treated, you can avoid having to meet that definition.)
More people would have died if the NHS had more capacity
Rory made the very counter-intuitive point that, if the NHS had a higher capacity and could care for more patients, more people would have died, not less, like you'd expect.
Now, with all the previous points, I think I can understand it.
The government believed the virus would only stop once enough people got infected that we had herd immunity, and that they couldn't do anything to stop it from spreading. But they didn't want the NHS to be overwhelmed, so before that happened, they reacted really strongly and restricted everything on the 23rd of March.
But, before the NHS was overwhelmed, people still died. And the longer it takes for the NHS to be overwhelmed, the longer they can wait before locking everything down.
So if the NHS has more capacity, the government can wait longer, and more people will die before they lock everything down.
As evidence for this, the government introduced harsh restrictions when they believed the NHS was close to being overwhelmed - this was 2 days before the lockdown was announced.
Delaying restrictions as long as possible makes sense, if you believe the virus can't be stopped
Under the original "It will not be possible to halt the spread of a new pandemic influenza virus" assumption, it makes sense to delay imposing restrictions as long as possible, rather than introducing them ASAP.
If it's impossible to stop virus spreading, the only way to stop the spread is to reach herd immunity, so you delay restrictions as long as you, can so as many people as possible are infected. You don't want to overwhelm hospitals, so you eventually introduce restrictions just before you "overshoot" and get too many infections for your hospitals to handle.
This is exactly the opposite approach you would take if you want to reduce the number of cases & deaths as much as possible - then, you'd introduce restrictions ASAP, because the smaller the number of infections you start with, the lower they'll be after the same amount of restrictions.. If you want to undo the restrictions when there are less than 10 cases a day, say (maybe because you'll be able to trace and isolate all their contacts), the earlier you introduce restrictions, the earlier you'll get to less than 10 cases a day.
The assumption also explains why the UK stopped tracing contacts of cases on the 12th of March - initially, Public Health England could only trace 5 contacts a week:
Currently PHE can cope with five new cases a week (requiring isolation of 800 contacts). Modelling suggests this capacity could be increased to 50 new cases a week (8,000 contact isolations) but this assumption needs to be stress tested with PHE operational colleagues.
There were far too many cases for PHE to trace properly - from the 9th to 15th of March, there were 2582, more than 50x the total number they could trace. They believed the virus had taken hold in the UK by then, the spread couldn't be stopped, most people would get infected eventually, so there was no point continuing to trace contacts.
(It's interesting to note that each case needed 160 contacts to be traced in March, but in January this year there were < 2 contacts per case (359 thousand contacts / 199 thousand cases). With the first case in the UK, 52 contacts were traced, 45 of them healthcare workers)
So, the UK's strategy was to
- follow their flu plan
- which didn't try to minimise the number of deaths, and said it was impossible to stop a virus from spreading
- and implied hundreds of thousands of people would die
- the government eventually realised this on the 16th of March
- it later added the goals of delaying infections over a longer period of time (not minimising them), and also to stopping the NHS from being "overwhelmed"
- and finally, they locked down the country on the 23rd of March, a week after they realised how many people would soon have died
Thanks to my friend Kat who doesn't have a Twitter account, Luci, Blake and Laura for proofreading.
