Focused protection

Did Denmark Achieve Focused Protection in the Second Wave?

Before the vaccines arrived, lockdown proponents argued that the only way to prevent large numbers of Covid deaths was by completely suppressing viral transmission. A focused protection strategy, they maintained, was just not workable.

The basic argument is as follows. Because the virus is so transmissible, and society is so interconnected, it would have been impossible to protect vulnerable people if we’d allowed community transmission to proceed unchecked. Without a lockdown, the virus would inevitably have found its way into hospitals and care homes, leading to lots of deaths.

It’s not an unreasonable argument, but I don’t buy it. (And let’s put aside the fact that even if lockdown does prevent more Covid deaths than focused protection, the total costs almost certainly outweigh the benefits.)

We already know that places like Utah, Sweden and South Dakota, which refused to lock down last year, did not do substantially worse than places that did lock down. We can argue about exactly how to do the comparison; the fact is that none of the dire predictions made for these locations actually came to pass.  

But is there an example of a country that achieved focused protection? Denmark might well be the closest. If we zoom-in on the second wave, and compare the country’s infection rate to that of the U.K., it isn’t dramatically lower:

Assuming the numbers are indeed comparable (which I’ll admit is a big assumption), Denmark saw 30% fewer infections between August of 2020 and May of 2021. Denmark did do more testing over this time period, but the U.K. had a higher share of positive tests.

If the lockdowners’ argument against focused protection is right, we’d expect Denmark to have had only 30% fewer deaths than the U.K. during the second wave; or at most, perhaps 50% fewer. After all, the country’s infection rate peaked at over 600 per million.

But this isn’t what we find. According to Karlinsky and Kobak, Denmark has had only 1% excess mortality since the pandemic began; the U.K.’s figure, by contrast, is 20%.

Now, more than half of Britain’s excess mortality was sustained in the first wave (which Denmark managed to avoid). But suppose that eight percentage points of the 20% were sustained in the second wave.

This would mean that Denmark’s deaths were not 30% or 50% lower than the U.K.’s, but almost 90% lower. Despite experiencing a moderately high infection rate in the winter, Denmark managed to keep deaths to a minimum.

Note: I’m not suggesting the country didn’t lock down; it did. (Though there was never a stay-at-home order, and the average stringency index was much lower than in Britain). My point is that some degree of focused protection apparently is achievable. There’s no necessary relationship between the infection rate and the death toll.

It doesn’t follow that Britain could have done as well as Denmark, which tends to finish at the top of every international league table. But with a bit of ingenuity, we could have done better than we did – in terms of both lives saved and collateral damage avoided.

The recent House of Commons report described the U.K.’s initial approach as “fatalistic”. But what was really fatalistic was assuming the only way to stop people dying of Covid was shuttering the economy and throwing civil liberties out the window.    

BMJ Publishes Belated Attack on the Great Barrington Declaration, but It Doesn’t Hit the Target

The Great Barrington Declaration, which advocates a focused protection strategy for dealing with COVID-19, was published in October last year – before many countries around the world imposed their winter lockdowns.   

Recently, The BMJ Opinion – a journalistic offshoot of the well-known medical journal – published a very belated hit piece against the authors. As you might expect, it’s light on scientific arguments and heavy on tactics like ad hominem, guilt by association and appeals to authority.

The authors, David Gorski and Gavin Yamey, really don’t mince words. For example, they describe the Declaration (which has been signed by hundreds of scientists and healthcare professionals) as a “well-funded sophisticated science denialist campaign based on ideological and corporate interests”.

Not exactly a respectful way to talk about your colleagues. But it’s hardly the first time the Declaration’s critics have sunk to this level. Just last month, Jay Bhattacharya became the subject of a censorious petition which claimed that he “sows mistrust of policies designed to protect the public health”.

Gorski and Yamey begin their article by criticising the Declaration’s authors for collaborating with the American Institute for Economic Research, which they claim is a “libertarian, climate-denialist, free market think tank”.

I’m not sure why this is a ‘gotcha’. Lockdown is about as un-libertarian a policy as you could imagine, so it’s not really surprising that a libertarian think tank would oppose it. And in any case, the Declaration’s website clearly states that the document was “was written and signed at the American Institute for Economic Research”.

Martin Kulldorff has since clarified that the AIER president and board did not know about the Declaration until after it was published. But even if they had done, so what? As Kulldorff notes, universities like Duke and Stanford have received money from the Koch brothers. Should we therefore completely disregard what their academics have to say?

Gorski and Yamey’s next move is to cite social media censorship of lockdown sceptics as evidence that their arguments constitute ‘misinformation’. (Incidentally, that term – which basically means ‘information that’s missing from the mainstream narrative’ – appears no fewer than six times in the article.)  

However, this argument relies on circular logic: ‘Something was censored on social media? Therefore, it’s misinformation. How do we know? Well, misinformation is what social media companies censor.’ In reality, of course, the fact that something was censored is no indication whatsoever that it’s factually incorrect.

The authors then allege that when Sunetra Gupta and Carl Heneghan met Boris Johnson in September of last year, they were successful in “persuading him to delay” a ‘circuit breaker’ lockdown, which could have forestalled the second wave of infections.

As historian Phil Magness has already noted, this argument is deficient on two counts. It’s not clear that Gupta and Heneghan did persuade the Prime Minister to shelve the ‘circuit breaker’ idea. But even if they did, there’s no reason to believe that policy would’ve prevented a large number of deaths.

Finally, Gorski and Yamey compare lockdown sceptics to ‘climate science deniers’, insofar as both groups “argue that evidence-based public health measures do not work”. They call for experts to push back against the Great Barrington Declaration by highlighting “scientific consensus”, citing the John Snow Memorandum.

Of course, the pro-lockdown John Snow Memorandum is just another public statement signed by scientists and health professionals. If it constitutes “scientific consensus”, then so does the Great Barrington Declaration. I’m only aware of one attempt to gauge overall expert opinion on focused protection: the survey by Daniele Fanelli.

He asked scientists who’d published at least one relevant paper, “In light of current evidence, to what extent do you support a ‘focused protection’ policy against COVID-19, like that proposed in the Great Barrington Declaration?” Of those who responded, more than 50% said “partially”, “mostly” or “fully”.  

Regardless of the exact number of experts who support focused protection, claiming there is a “scientific consensus” against it is simply false. Long before the Declaration itself was published, many scientists had proposed some version of precision shielding. In fact, this was basically the U.K.’s plan until the middle of March, 2020.

On March 5th, Chris Whitty told the Health and Social Care Committee that we are “very keen” to “minimise economic and social disruption”, and mentioned that “one of the best things we can do” is “isolate older people from the virus”.

Another prominent scientist who has argued in favour of focused protection is Sir David Spiegelhalter. In an article published on May 29th, he and George Davey Smith said that we ought to “stratify shielding according to risk” because lockdown is “seriously damaging many aspects of people’s lives”.

They noted that this would require “a shift away from the notion that we are all seriously threatened by the disease, which has led to levels of personal fear being strikingly mismatched to objective risk of death”.

Among the ad hominems, appeals to authority and repeated uses of ‘misinformation’, finding a scientific argument in Gorski and Yamey’s article is not easy. And given that the content’s almost a year out of date, I’m not sure why the authors felt the need to publish it.

Air Filtration/UV Light Can Remove Airborne SARS-CoV-2 From Hospital Wards, Study Finds

In a study published earlier this year, Paul McKeigue and colleagues analysed data on all diagnosed cases of COVID-19 in Scotland, as well as a large number of matched controls. They found that a staggering 30% of severe cases (those that resulted in critical care admission or death) were linked to a recent hospital visit.

This suggests widespread nosocomial transmission of SARS-CoV-2. In other words, a lot of people caught their infections in hospital, and then became seriously ill.

The fact that such a large portion of severe cases were linked to a recent hospital visit is actually not so surprising. After all, people vulnerable to COVID-19 (the elderly and persons with underlying health conditions) are overrepresented among those who make frequent hospital visits.

Nonetheless, it’s rather concerning that hospitals – places where people are meant to come out healthier than they go in – were a major site of SARS-CoV-2 transmission.  

Given that COVID-19 patients, as well as those vulnerable to COVID-19, tend to be concentrated in hospitals, making efforts to reduce nosocomial transmission would seem like a top priority. Indeed, one would expect interventions that did reduce such transmission to have a large benefit/cost ratio.

Which makes a new preprint so interesting. Andrew Conway-Morris and colleagues investigated whether airborne SARS-CoV-2 could be removed from hospital wards using portable devices that filter and sterilise the air.

Their experiment involved two units within an English hospital: an ordinary Covid ward, and an ICU containing Covid patients. The presence of airborne SARS-CoV-2 was measured during three consecutive weeks: one in which the devices were turned off; one in which they were turned on; and one in which they were turned off again.

In addition to measuring the presence of SARS-CoV-2, the researchers measured the presence of various other microbial bioaerosols, such as E. coli and staphylococcus. Their results for the Covid ward are shown in the figure below.

When the devices were turned off, many microbial bioaerosols (including SARS-CoV-2) were detected. Yet when they were turned on, all of these except candida were undetectable. This means the devices were successful in removing not only SARS-CoV-2, but also other potentially dangerous pathogens.

As the authors note, SARS-CoV-2 was detected on “all five days before activation of air/UV filtration, but on none of the five days when the air/UV filter was operational”. The virus was again detected on “four out of five days when the filter was off”.

Interestingly, SARS-CoV-2 was barely detected in the ICU (regardless of whether the devices were turned on). This may be because viral shedding is lower among critically ill patients, or because ICU staff were wearing proper N95 masks.

It’s important to note: the study didn’t show that the devices actually prevent transmission of SARS-CoV-2 in hospitals. However, the results constitute strong circumstantial evidence that they would reduce transmission.

While attempting to halt transmission of SARS-CoV-2 in the community at large is costly at best and futile at worst, attempting to do so in high-risk hospital environments makes a great deal of sense. Further investigation into the efficacy of these devices is clearly warranted.  

A Question for Chris Whitty

I haven’t watched any of the Government’s COVID-19 press briefings since the early weeks of the pandemic. The scientific parts seemed to be mostly concerned with projections from rather dubious epidemiological models, and the political parts were even less informative.

As I understand it, the Q&A that follows whatever Boris and the boffins have said often involves journalists demanding to know why there aren’t more restrictions in place (more rules, more limits, more penalties).

Ironically, these questions tend to come from people who a few months before the pandemic might have compared Boris Johnson’s Government to certain mid-20th century political movements that we now associate with authoritarianism.

What questions would I ask Boris and the boffins? There are many I’d like to raise, including: “Why hasn’t the government published a cost-benefit analysis of lockdown?” Such analyses are routine in policy-making, and you’d expect that something as far-reaching as a national lockdown would justify one.

Another query I’d like to make is: “What specific evidence led the government to change its advice on masks?” Back on 4th March 2020, Chris Whitty told Sky News that “wearing a mask if you don’t have an infection reduces the risk almost not at all”. And as late as 3rd April, Jonathan Van Tam said “there is no evidence that general wearing of face masks… affects the spread of the disease”.

However, the question I’d most like to ask – of Chris Whitty in particular – is as follows.

Professor Whitty, on 5th March 2020, you told the Health and Social Care Committee that “we will get 50% of all the cases over a three-week period and 95% of the cases over a nine-week period”. You said that we are “very keen” to “minimise economic and social disruption”, and mentioned that “one of the best things we can do” is “isolate older people from the virus”.  

This all sounds rather similar to the Great Barrington Declaration. Why then, in an interview with The BMJ on November 4th, did you describe that document as “wrong scientifically, practically, and probably ethically as well”? You said that the Great Barrington Declaration is “really a pretty minority view”, but it appears to have been your view as recently as eight months earlier.

As I’m sure you’re aware, there is a document titled “UK Influenza Pandemic Preparedness Strategy 2011”, which was published by the Department of Health. It says that attempting to stop the spread of a new pandemic influenza “would be a waste of public health resources and capacity”.

And as late as 2019, the World Health Organisation published a report titled “Non-pharmaceutical public health measures for mitigating the risk and impact of epidemic and pandemic influenza”. This document classifies “quarantine of exposed individuals” as “not recommended in any circumstances”.

Given that the WHO, the Department of Health and you – as recently as March 2020 – have rejected suppression as a strategy for dealing with respiratory pandemics, why did you describe the alternative focused protection strategy as “wrong scientifically”? Thank you for listening, and I look forward to your answer.

COVID-19 Mortality Rate Among Children Is Even Lower Than Previously Thought

We’ve known since the early weeks of the pandemic that age is the single best predictor of COVID-19 mortality, and that the risk of death for young people is vanishingly small.

A letter in the New England Journal of Medicine reported that zero Swedish children aged 1–16 died of COVID-19 up to the end of June 2020. And only 15 were admitted to the ICU, of whom four had a serious underlying health condition.

Of course, England is a much larger country than Sweden, and it’s been a whole other year since those Swedish data were collected. So how many English children have died of COVID-19?

In an unpublished study, Clare Smith and colleagues sought to identify the number of COVID-19 deaths among people aged under 18 between March 2020 and February 2021. They examined data from the National Child Mortality Database, which was linked to testing data from Public Health England and comorbidity data from national hospital admissions.

The structure of their dataset allowed the authors to distinguish deaths that were plausibly from COVID-19 and deaths that were merely with COVID-19.

3,105 under 18s died from all causes in England during the relevant time period. Sixty one of these involved people who had tested positive for the virus. However, the authors determined that only 25 were actually caused by COVID-19. And of the 25, 76% had a serious underlying health condition.

Given that an estimated 469,982 under-18s were infected with the virus up to February of 2021, the survival rate in this age-group (the inverse of the IFR) was 99.995%. What’s more, 99.2% of total deaths were caused by something other than COVID-19.

Smith and colleagues’ findings underline just how small a risk COVID-19 poses to young people, and hence – I would argue – why a focused protection strategy was preferable to blanket lockdowns.  

As early as 10th April 2020, Martin Kulldorff – co-author of the Great Barrington Declaration – published an article on LinkedIn titled ‘COVID-19 Counter Measures Should be Age Specific’.

Based on the data that were then available, he estimated one would need to stop 3.5 million children being exposed in order to prevent the same number of deaths as one could prevent by shielding 1,000 people in their 70s. He argued, therefore, that Covid counter-measures must vary by age.

A similar argument was made by George Davey Smith and David Spiegelhalter in a piece for The BMJ last May. These authors called for “stratified shielding”, while noting that this would “require a shift away from the notion that we are all seriously threatened by the disease”.

According to the medical researcher Russell Viner, who spoke to Nature, “There’s a general feeling among paediatricians that probably too many children were shielded during the first wave.” And the epidemiologist Elizabeth Whittaker said that efforts to shield children “have probably caused more stress and anxiety for families than benefit”.

In addition to “stress and anxiety”, there’s also the learning losses associated with months of online teaching. All this compared to the marginal impact closing schools had on the spread of COVID-19.

When we look back at the response to Covid, serious questions will have to be asked about the costs of lockdown, not only to society in general, but to young people in particular.

Time to Move on From Focused Protection

There follows a guest post from Steve Sieff, creator of GreenBandRedBand.com.

In June 2020 I launched GreenBandRedBand.com. It is a type of what came to be called focused protection. My system proposed a way of people communicating to others if they wanted to be protected from coronavirus or if they were content to run the risk of contracting the virus. Those who were prepared to take the risk would show others how they felt by wearing a green wrist band or some other garment to communicate their position. Those who wanted to be protected but didn’t want to shield at home would wear a red equivalent. Around those requiring protection it was envisaged that all of us would respectfully adopt the measures that were being recommended to help stop the spread of the virus. Although that would still have been disruptive and unwelcome, it would have been far more palatable than being obliged to take measures around those who did not require them, and infinitely more so than laws which criminalised social interaction. So the system seemed to me to strike the right balance between retaining our personal freedoms and respecting the rights and wishes of others.

I’ll take this opportunity to express my thanks to the large numbers of Lockdown Sceptics readers who contacted me to express their support or who purchased bands and to the editorial team for featuring the site on a number of occasions.

A year down the line LS highlighted a Guardian article reporting on a ‘variant’ of my system being used in some places in the U.S., and other readers may recall Freddie Sayers in UnHerd discussing something similar. You might assume that I would welcome news that a similar system is getting some mainstream attention at last.

But times have moved on and in June 2021 I have slightly mixed feelings about it. On the one hand it is great to see that people are realising that they can manage themselves by communicating with each other rather than needing the Government to micromanage their lives. That should have happened from the outset. On the other hand, in places where the vaccines are available to the vulnerable, there is a strong argument that the time for this system is coming to an end because everyone should be ‘green’.

I proposed the system as an alternative to lockdown and restrictions and to recognise that people would rebound from the fear messaging at different rates. It was designed to be sustainable while large numbers of vulnerable people remained. But it was not envisaged to be permanent. As the numbers of vulnerable reduce, so does the need for specific measures to cater for that vulnerability. There comes a point where the position has moved to the extent that it is no longer kind or helpful to continue to indulge fear. Indeed by continuing to do so one risks perpetuating fear unnecessarily.

When GreenBandRedBand.com was conceived, vaccines seemed a long way off. But a year later they are a reality and the rollout in some parts of the world has been rapid. It may be that we manage to improve their efficacy or that we develop more treatments for people who do contract the virus but essentially the vaccines are our best effort. People who are worried about being vulnerable – or who are actually vulnerable – aren’t going to get a better offer than vaccination. So if you aren’t ready to stop asking for protection after vaccination is available to you then it starts to look like you will never be able to be comfortable with normal social interaction. Or in the terms of my system, you absolutely don’t have to be vaccinated to choose green, but if you were red before and being vaccinated isn’t enough to make you choose green, then what will?

Did Care Homes Achieve Focused Protection in the Second Wave?

Contrary to popular understanding, Britain’s second wave of COVID-19 was less deadly than the first: although there were more deaths within 28 days of a positive test, age-adjusted excess mortality was lower. 

One possible explanation is that fewer people were infected in the second wave (even though the infection fatality rate remained constant). However, data from the Coronavirus Infection Survey suggests that roughly the same number of people were infected in the two waves. About 7% of people had antibodies at the end of the first wave, and about 14% had antibodies toward the end of the second wave (before the vaccination program had gotten fully underway). 

Incidentally, some people may have been infected without developing antibodies. I’m using the number who developed antibodies as a proxy for the total number who were infected in each wave.

Another possible explanation is that we became better at treating the illness. Evidence suggests that thousands of lives were saved by corticosteroids like dexamethasone, but these may not have been widely used in the first wave. Yet another explanation is simply that there were fewer frail elderly people alive at the beginning of the second wave, meaning that the average elderly person who became infected was less likely to die from the disease. 

However, there’s possibly a fourth reason why the second wave was less deadly than the first, namely that care homes achieved a degree of focused protection.

In the first wave, a disproportionate number of those who died were care home residents. This is partly because elderly patients who’d caught the virus in hospital were discharged to care homes when they were still infectious, resulting in deadly outbreaks. Hence more effort was made to shield care home residents in the second wave. 

According to the ONS, there were 27,079 excess deaths in care homes during the first wave, but only 1,335 during the second wave:

This finding is supported by two recent academic studies. One study, published in Environmental Research, found that the percentage of COVID-19 deaths among care home residents was lower in the second wave in eight out of 11 countries with available data, including the UK. 

Another, unpublished study observed a major spike in excess mortality among care home residents last spring, but no increase during the final weeks of 2020.

While it’s too early to say exactly which factors explain the reduction in mortality between the two waves, the evidence presented here suggests that effective shielding of care home residents may have been a major contributor. Though it should be noted that care home occupancy was lower in the autumn and winter, which probably accounts for some of the disparity in excess deaths.

Perhaps if more attention had been paid to shielding in the first wave, Britain would have come through the pandemic with a lower death toll. 

This post has been updated.

What Would a Focused Protection Strategy Have Looked Like?

We’re publishing a new piece by Dr Noah Carl today, this time one looking at ‘Focused Protection’, the strategy recommended in the Great Barrington Declaration. Noah thinks it would have resulted in fewer deaths than locking everybody down. Here’s an extract:

At any point during the pandemic, deciding which measures to implement represents a trade-off between their effects on the epidemic’s trajectory and their effects on society at large. Measures that substantially reduce cases or deaths, while having only a small impact on society, are worth putting in place. By contrast, those that barely reduce cases or deaths, while having a large impact on society, are best avoided. Mounting evidence indicates that measures like stay-at-home orders and closures of non-essential businesses are of the latter kind; they have large costs and relatively small benefits. In just the past week, two new studies casting doubt on the efficacy of lockdowns have been published. Vincent Chin and colleagues analysed data from 14 European countries, and found that “lockdown had no consistent impact”. Likewise, Christopher Berry and colleagues examined shelter-in-place orders in the United States, but did “not find detectable effects of these policies on disease spread or deaths”.

However, this is not to say there aren’t any restrictions worth implementing. One measure whose efficacy is supported by a number of studies, and which makes sense intuitively, is restricting large gatherings. For example, Vincent Chin and colleagues found in one of their analyses that “the simple banning of public events was beneficial”. Nicolas Banholzer and colleagues have reported a similar finding. In an unpublished study, they compared the impact of different non-pharmaceutical interventions, and found that “event bans were most effective… whereas stay-at-home orders and work bans were least effective”. Another measure that makes a great deal of sense is telling symptomatic individuals to self-isolate at home.

Worth reading in full.

Up to Two Thirds of Serious Covid Infections are Caught in Hospital – Study

An important pre-print was published last week by Public Health Scotland looking at how all 204,913 people eligible for shielding (the “clinically extremely vulnerable”) – about 3% of the population – have fared during the COVID-19 pandemic. The researchers matched all 160,307 positive cases of COVID-19 (as of January 28th) to individuals in Scotland to form a complete picture of the course of their illness in order to evaluate the effectiveness of the shielding programme.

Their striking conclusion was they found “no evidence that the shielding programme per se reduced COVID-19 rates”, though they allowed for the possibility that “without shielding advice and support the outcome in this group would have been worse”.