Ferguson speaks

In case you can’t read it, here’s the text off the Financial Times’ recent story on Neil Ferguson, including quotes from him. As the FT notes, things are as still as clear as mud.


 

Imperial’s Neil Ferguson: “We don’t have a clear exit strategy”

Financial Times April 7 2020

By Jemima Kelly

 

There’s a stark truth stalking the debate about the effectiveness of lockdowns. While governments can theoretically check out any time they like (and indeed will be inclined to do so quickly), they might not in reality be able to leave.

If restrictions are eased too quickly, we risk a second wave even more deadly than the first.

Yet an apparent lack of awareness about the dangers posed by a secondary wave is adding to a wave of confusion and misinformation around coronavirus that we have noticed spreading across the internet since that report from Imperial College London from March 16th. (The optimism appears, also, to be making markets rather giddy).

But if we aren’t heading towards a clean exit, what are we heading towards? If we have really abandoned herd immunity – the strategy of allowing the virus to spread gradually through about two-thirds of the population in order to let it die out of its own accord – how are we planning on getting back to normal life?

Trying to get to grips with the UK government’s real strategy is problematic, despite the fact that a great deal of virtual ink has been spilt in trying to “explain” it. Indeed it hasn’t seemed totally clear to us whether they are deliberately not giving us the full picture, or whether the government does not actually know what the strategy is beyond this initial curve-flattening stage.

One group of people who might be able to help, we figured, would be the scientists advising government. And while we do not intend to endorse Imperial’s modelling above others’ work – and indeed we realise that doing so at this point constitutes taking a clear side in a debate that has become highly politicised – we do feel that Professor Neil Ferguson, who leads Imperial’s team, appears so far to be carrying the most influence among the government’s advisers.

So we decided to have a chat with Ferguson. What follows here is what we gleaned from our conversation with him (and, crucially, what still remains unclear, which is, we’re afraid… a fair bit).

 

Did the government really U-turn on “herd immunity” after the Imperial report?

On Sunday health secretary Matt Hancock told The Andrew Marr Show that herd immunity was never part of the government’s strategy, adding (somewhat irritatedly, we should add) that the idea they had being pursuing it had “been rubbish from start to finish”, and also that any journalists suggesting otherwise were “talking nonsense”.

But that is somewhat at odds with what we have seen and heard until now.

For a start, Britain’s chief scientific adviser, Sir Patrick Vallance has spoken quite openly about the strategy. He said in a BBC interview on March 13th that “if you suppress something very, very hard, when you release those measures it bounces back and it bounces back at the wrong time”, and that the aim, therefore, was “to try to reduce the peak, broaden the peak, not suppress it completely”, so as “to build up some kind of herd immunity”.

But after causing a considerable amount of controversy and alarm when he told Sky News on the same day that about 40 million Britons would need to catch Covid-19 in order for the nation to build up that herd immunity, there appeared to be a change – in tone, at least.

The Guardian’s front page on March 14th

The very next day, talk of a government “U-turn” was splashed over the front pages, as Downing Street indicated it would be banning mass gatherings from the following weekend. Still no lockdown – that wouldn’t come till ten days later – but it appeared that the government, suddenly, had changed its mind about this herd immunity idea (or had decided that talking about it quite so openly was a bad idea, anyway).

Then on March 16, the “bombshell” Imperial report came out: there would be as many as a quarter of a million deaths in the UK if the government pursued the milder “mitigation” measures that would accompany a herd immunity strategy. That, said the press, must have been the reason that the UK government had abandoned its original strategy.

But was that really what happened?

Here’s what Ferguson told us regarding the rationale for that so-called “U-turn” (all emphasis, here and throughout, ours):

The government were aware of what our results were showing certainly in the previous week, and some of (the results) in the previous two weeks. There was a very active debate within government and within scientific advice bodies as to the timing of interventions and the package that would need to be put in place.

 

The paper came out that day partly because there was pressure on government to be showing the modelling informing policymaking, so we worked very hard to get that paper out at that time… it looked, therefore, like that was the paper that informed policymaking, but in fact a much wider range of scientific advice and modelling advice had gone into government in the previous two weeks, all of which pointed in that direction. 

The above implies the government was aware of the potential death toll – or the one being projected by the scientists on their advisory committee, anyway – but had not considered a drastic lockdown strategy until it became clear that the likely number of deaths from any other strategy would not be seen as politically acceptable. It seems, therefore, that the paper was published at that time partly to help justify a change in the messaging. A “U-turn” doesn’t seem like quite the right term, therefore, for what happened.

 

Was this just a messaging change then?

A few days ago Buzzfeed reported that a person who was familiar with the thinking of both the UK’s chief medical officer Chris Whitty and Sir Patrick believed that herd immunity remained, privately, a long-term objective.

This chimed with a lot of what we had been thinking about the government’s strategy. While the current strategy appears to be geared towards trying to prevent overwhelming the NHS, hence the current lockdown strategy, are we really trying to keep deaths to the lowest number possible?

Some countries brought in lockdowns before suffering even a single death (Politico has a useful list here), so why did we wait so long to begin ours? Was this a case of putting the economy ahead of lives? Was it negligence?

Neil Ferguson seemed to suggest to us it was a bit of both (note the bolded bit in particular):

The government made decisions about the timing of interventions, balancing I would say the impact of the epidemic and impact, therefore, on healthcare demand and mortality, against the staggering economic and social costs of these type of interventions.

I think in some sense, what had happened in China was a long way away, and it takes a certain type of person to take on board that this might actually happen here. And I think that evidence coming out of Italy in the prior two, three weeks to lockdown being imposed here clarified thinking a lot in terms of: this really was in line with the numbers we’ve been producing for some weeks in terms of potential mortality and that this was really likely to happen.

But perhaps there was also another reason. After all, what keeping the country up and running for two extra weeks might have done is infect a good number of people: not enough to overwhelm the NHS, but enough to make a start at building up herd immunity. This, eventually, would provide some kind of exit (and one that was advocated by the chief pandemic modeller, Sir Graham Medley in the Times over the weekend).

Ferguson said that the relatively small proportion of the country that he estimates have been infected so far – between about 3 and 5 per cent – would mean the extra two weeks before lockdown wouldn’t have got us very far with herd immunity. Even after the peak of deaths that he estimates will come in around a week or so’s time, he reckons no more than about 10 per cent of the population will have been infected.

But as we know from the rather controversial Oxford paper that suggested as much as 68 per cent of the population could have contracted the virus, the Imperial researchers are not the only ones trying to work out how many infections there have been so far, and so they could be underestimating the rate. And when we suggested to Ferguson that some in government might be pursuing some kind of watered-down version of herd immunity, or at least might be considering it a back-up option (in case a vaccine is not found in 18 months), he didn’t totally dismiss the idea.

Instead, Ferguson noted that there was actually no definite exit strategy in place at this point, though testing and contact-tracing might help (more on that in a bit).

It doesn’t seem clear to us, though, that we can rule out the idea that at least some in government are still pursuing the idea of herd immunity in the background, even if it is just a fall-back plan.

 

What about this “adaptive policy” Imperial suggested?

The “adaptive policy” outlined in Ferguson’s Imperial report was the basis for the assumption that deaths in the UK could be contained at around the 20,000 level. This strategy, the paper said, would involve relaxing certain measures (namely social distancing and school closures) when ICU case incidence falls below a certain threshold, and tightening them again when that is exceeded, and continuing on like this until a vaccine is found. (Home isolation and household quarantine for symptomatic cases, meanwhile, would be kept in place throughout.)

Ferguson’s paper estimated that the way this would work out would mean that, in the UK, social distancing measures would be in place about two-thirds of the time until we have a vaccine (Ie, about 12 months if the vaccine were to come in 18 months, a timeline which is not at all certain).

But Ferguson’s testimony to the Science and Technology Select Committee on March 25 confused many when he also suggested this strategy would only be used until other “counter-measures” became available, and also, that “we clearly cannot lock down the country for a year” despite forecasting the same numbers of deaths.

In our interview, Ferguson appeared again to distance himself from the version of adaptive policy idea that had been outlined in the report (and that’s despite the fact that because the NHS has now increased its surge capacity, the amount of time that the nation would be in suppression mode would be likely to be less than the two-thirds of the time until a vaccine is found that Imperial had initially estimated).

Instead, Ferguson focused more on the ramping-up of testing.

There are lots of downsides to that strategy: first of all that would mean months and months of the NHS being under high stress. Second you’ll have high mortality even with shielding in place. The third one, and I’m not the one to talk about this, but the population acceptability of such a strategy of relaxing things and then telling people sorry we’re going to have to lock down again – there’s quite a lot of behavioural science underway at the moment suggesting that a lot of people would find that a hard strategy to swallow and accept.

So whilst we are looking at those things, we’re also looking at: can we substitute massively ramped up testing … for some of the more draconian social distancing currently enforced?

 

And what about those death estimates?

There has also been confusion around the fact that Ferguson also noted in his parliamentary testimony, that the reproduction number – i.e. the average number of others that each infected person passes the virus onto – had increased to around 3, or just above, relative to an original forecast of 2 and 2.6, without a commensurate increase in deaths. (He confirmed to us that he now believes that number is between about 2.8 and 3.1.)

So how can we still be talking about the same number of deaths when the variables seem to have changed? (In the Imperial report, the higher the reproduction number, the higher the estimated deaths.) We asked Ferguson about this and he said:

Yes, if you keep the effectiveness of the intervention the same, a higher reproduction number is harder to control and you get a higher number of deaths if you make the same assumptions about interventions. But the thing which is also changing is we have some indication of how effective these interventions actually are, from data collected from the London School (of Hygiene and Tropical Medicine) on how much contact rates have gone down, and also indirect data from people’s mobility patterns, how much people are travelling — there’s a whole range of survey data. And it looks like actually the interventions have been more effective than assumed in that report.

In other words, Ferguson reckons the higher reproduction number will be offset by the fact people are complying to a higher degree than the assumptions made in the report, which were as follows:

In terms of the UK death estimates, Ferguson said:

We will be putting out updated estimates, probably in the next week, both on intervention impact and on growth rate, but we hope that the two cancel out really. We hope we’re in the same sort of regime of mortality — somewhere between about 5,000 and 30,000 deaths, and probably closer to 10,000-20,000.

So what’s the thinking on an exit plan?

Although Ferguson focused on the need to ramp up of testing and to bring in widespread contact-tracing, he said it was not obvious that even this would prevent a second wave:

Obviously what we would like to find is a strategy which allows us to go back to — it won’t be normal life but a bit closer to normal life, and suppresses transmission. That almost will certainly involve something akin to Korea, massively ramping up testing, and contact-tracing. But it’s not clear even in Korea — and I’ve seen some of the analysis done there — that they can really relax all their social distancing and yet keep transmission low.

 

So there is no master plan in the background being followed here. There is a lot of research being done in real time, which is feeding into policy, to try and work out: is there in some sense an optimal strategy which keeps the NHS functioning, allows more economic and social activity to continue than is going on at the moment and gets us through the next, frankly, 18 months? I don’t know quite what that will look like or even if it’s completely feasible.

 

We don’t have a clear exit strategy at the moment.

That’s right. If we’re to trust what one of the government’s top scientific advisers is telling us, there is at this point no master plan in place. Reassuring stuff, we’re sure you’ll agree.

Just over a week ago, we tried to flatten the coronavirus confusion curve. Since then, frankly, the confusion feels like it has intensified. We perhaps just need to accept, uncomfortable as it may be, that we are living in a world in which the “experts” and leaders to whom we normally turn for guidance just don’t have all the answers that we’re looking for. Nobody fully understands this virus yet. Nobody knows where the exit door is.


 

Share this article on social media:

12 thoughts on “Ferguson speaks

  1. How long before the narrative switches from “Listen to the Experts!!” to “I’m just a scientist”

  2. I think it is becoming apparent that a test for antibodies may be misleading. Some people will fight off the virus without developing antibodies. This may be related to the size of the initial ‘inoculum’ they receive.

    In one study of 175 patients with mild symptoms, it was found that 10 did not develop detectable antibodies. https://www.medrxiv.org/content/10.1101/2020.03.30.20047365v1

    In asymptomatic infectees might this be a higher proportion? (Intuitively it would be). Would this go some way to explaining the findings of several studies now showing populations with an apparent limit of 15% immunity according to antibody tests? Might, in fact, more of the population have been exposed but fought off the infection without producing detectable antibodies? They would, at least, have shown ‘natural immunity’ (and may have also produced ‘memory cells’ without antibodies..?). If their immunity is not detectable, however, then that is a bit of a conundrum wrt the ‘exit strategy’.

  3. What we do know, with increasing certainty, is that most people do not have any appreciable risk of dying from Covid-19. The people at risk are those who are elderly and/or medically compromised. UK Gov’s own figures show this. When healthy working age people get exposed to the virus, they generally do not catch it or at worst have minor symptoms.

    It seems the logical, evidence-based Exit Strategy would be to switch most resources to helping the At Risk people isolate themselves from infection, while working-age people are encouraged to get back to work. If some of the working-age people catch C-19 — Great! That will help rapidly to build the desired pool of people with resistance.

    It seems so obvious that one has to wonder why the logical course of action has escaped the attention of our Betters.

  4. Exactly! The only reason I can see that this is not the policy is the extreme fear and pressure the media have created that has meant the only solution is ‘total lockdown’.

    NZ is getting praise for it’s early and extreme lockdown in the international media, yet I strongly suspect when people wake up to the economic catastrophe this has put into motion, unrest will start very rapidly.

  5. If the reproduction number is higher then the total number of infected people is higher but the deaths, being the one thing we can observe relatively accurately, are what they are. So the mortality is actually lower for a higher R. That was, I believe, part of the Oxford contention.

    What part of Ferguson’s argument have I misunderstood?

  6. Xavier, yes, using antibodies to test for prior exposure will give you a falsely low rate of infections. It does however give us a baseline (must be more than X have prior exposure), and for the individuals tells us the likelihood that they could have a future infection. No (detectable) antibodies = higher probability of future infection, irrespective of prior exposure. (Detectable) antibodies = lower but not zero probability of future infection.

    There isn’t really ever binary certainty in medicine (some genetic stuff perhaps). I’m sure most observers here accept that but it’s a constant struggle to get most people to accept it.

  7. Thanks, Bloke. But what if it is the case that in a real Covid-19 epidemic, a population never goes above 15% in terms of the people with antibodies? In other words, the virus has been suppressed such that it cannot spread without the need for antibodies in most of the population; the preliminary non-antibody level of immunity in the general population is sufficient. If a threshold such as 60% antibodies is being used to define herd immunity, it may never be seen, and any models that rely on an equivalence between antibodies and exposure/immunity will be wrong. A typical response to the new Gangelt study is:

    “To me it looks like we don’t yet have a large fraction of the population exposed,” (Nicholas Christakis, doctor and social science researcher at Yale University).

    The experts and politicians will be setting policy on that basis.

    For sure, a lack of antibodies does not guarantee a person cannot get a future infection (as I understand it) if exposed to a stronger source of infection, but it doesn’t mean that practical herd immunity hasn’t been achieved even if it applies to the majority of people. I currently see no attempt to cater for this possibility amongst government, experts and commentariat.

  8. So, it boils down to…

    “We don’t know what percentage of the population is / has been infected”…

    “We don’t know the R0 for the virus”…

    “We are not collecting accurate statistics to show the actual mortality rate of the illness, as we are still combining WITH, FROM and (possibly) BECAUSE OF (ie those who’ve been turned-away from hospital because they have only an ‘ordinary’ illness from which they’ve subsequently died) to give a nice scary number to the BBC”…

    “We’re shit-scared of the MSM”…

    But other than that, the outputs of our shonky models are proof positive that turning the UK into a police state and bankrupting (possibly) millions is the way to go.

    So much like climate modelling it’s (almost) laughable.

  9. Some pretty extraordinary stuff in this. Amazed I’ve not seen it referenced or discussed elsewhere. I follow just about every significant commentator on Covid on Twitter but I didn’t know this interview existed. Eg the claim the delay was deliberate to seed population as a starter towards herd immunity.

  10. The problem politically with that is that not all the deaths are in the aged, or the young but obviously medically compromised (cancer patients for example, people with diabetes etc). There will be a significant number who are medically compromised in ways they are either completely unaware of, or underestimate how at risk they are. And what every politician fears is being put under the spotlight by the media because a 40 yo has left a widow and orphaned his 2 kids after being told to go back to work by the government.

    Incidentally this crisis should be a wakeup call to everyone – its no good being middle aged, fat and unfit and probably pre-diabetic, because if you are this virus could well kill you. You might have thought that Western medicine can save you from your own gluttony and laziness, but it can’t and you need to take action to get yourself in shape. If you don’t then you’re painting a target on your forehead.

  11. The severity of the infection will ultimately depend on the genetics of the individual. Those with capable immune systems will have a less severe infection. Those old and with co-morbidities will most probably die. With no vaccine we must develop a herd immunity to fight off this virus. At present it is not known if this virus COVID-19 confers a seasonal or a longer period of immunity. Once there is a vaccine that works the pressure will ease to a situation we have annually with influenza virus although Corona virus genomes are more stable than ‘flu virus genomes, so mutation risk is lower. Rates of infection should be falling towards the end of April and deaths by mid to late May I expect.

  12. If there is an implication that the flu vaccine may make the outcome of C19 worse for some people due to ADE (there is a paper I have seen making this link), can we afford to give everyone a universal vaccine that could react badly with the next virus to come along? Maybe, in fact, it’s essential that some of the herd do not develop antibodies, and that’s the secret of any species’ survival over the ages.

Comments are closed.