Was lockdown necessary?
BMJ 2022; 376 doi: https://doi.org/10.1136/bmj.o776 (Published 23 March 2022) Cite this as: BMJ 2022;376:o776
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Dear Editor,
Kit Yates makes an important contribution to the debates on pandemic response, in particular the discussions around economic versus health costs of different actions, and the shamefully small sums allocated for post-pandemic recovery.
There are, however, points it is important to add about the epistemic situation we found ourselves in early in 2020, and what the actual advice given to Government around pandemic mitigation was. The decision to have secrecy around SAGE discussions and participants led to considerable, and clearly for many distressing, uncertainty about these issues. The aim of such secrecy was to protect SAGE participants from personal abuse, however this has occurred anyway [1]. In fact, the secrecy has led to florid but easily falsifiable accusations that SAGE's scientific advice was motivated by factors from a belief that the economy trumped lives to being motivated by concepts linked to eugenics [2], which in my opinion constitutes its own form of abuse, particularly when originating from people with positions of power and responsibility such as clinicians and academics.
At present, a large number of SAGE papers are available publicly, and it behoves anyone commenting on the early pandemic to read these. The GO-Science document from 3 March 2020 [3] is particularly relevant in this context, stating: (i) that recommendations on policy are not made by SAGE because they involve trade-offs and wider impacts; (ii) that interventions of the kind we would now call "lockdown" do not prevent future waves of cases once lifted; (iii) that there is enormous uncertainty about whether suppression of transmission through "lockdown" is possible, and for how long.
In the original opinion piece, the statement is made that New Zealand “clearly” shows that there is no trade-off of the kind the SAGE advice discusses – but this is the only country in the world where suppression followed by a “true” zero number of infections allowing some release of restrictions, and then comprehensive vaccination before fuller reopening, has been realised. While anyone reasonable will find this an admirable outcome there is likely much to learn from, if there genuinely were a reliable source of knowledge in early 2020 that such a strategy was viable for the UK then it has never been presented.
Most importantly, while many have claimed after safe, effective vaccines became available that they always “knew” they would be, such “knowledge” was never written down before the clinical trials of such vaccines were carried out, at least not in a manner that would allow others to draw the same conclusions. The only reasonable manner to proceed in early 2020 was therefore to make plans that were somewhat robust to the timescale of vaccine development and ultimate vaccine efficacy. Sadly, events in many other countries have shown that every part of New Zealand’s approach needs to come together for it to succeed – the lack of vaccine coverage in Hong Kong at present, for example, or the terrible second and subsequent pandemic waves that affected the majority of the world’s population living in countries unable to achieve a true zero through lockdown, or to lockdown indefinitely.
We should, of course, continue to seek to clarify these and other issues through discussion of pandemic policy (not least since the probability of another one of similar severity is quite likely in most of our lifetimes) in a charitable and professional manner, as the BMJ encourages, and with recognition that none of us has a monopoly on virtue or wisdom.
Yours sincerely,
Thomas House.
[1] Nature 598, 250-253 (2021) https://doi.org/10.1038/d41586-021-02741-x
[2] Medley, Graham F, Herd immunity confusion, The Lancet, Volume 396, Issue 10263, 1634 - 1635 (2020) https://doi.org/10.1016/S0140-6736(20)32167-X
[3] Government Office for Science: Illustrative impact of behavioural and social interventions lasting several months on a reasonable worst-case epidemic, 3 March 2020. https://assets.publishing.service.gov.uk/government/uploads/system/uploa...
Competing interests: Served on the SPI-MO modelling sub-committee of SAGE during the pandemic.
Dear Editor
Dr Yates makes several assumptions in this article.
1. The statement that the economies of Spain and the UK suffered more, assumes that the highest death rate was due to a less effective suppression strategy, rather than geographical or travel hub advantage. That New Zealand locked down 2 days after the UK is conveniently left out.
2. On his criticisms of the post pandemic catch up programme for schools.
There is no cure for the time they lost. Children were denied not only schooling, but the opportunity to mix, exercise: developing physically and emotionally. And with critical periods of psychosocial development missed, how are they to ‘catch up?’. If there is an error here, it was in the insistence from independent sage and the teacher’s unions that schools remain shut during the first lockdown, once the initial data on at risk groups and transmission characteristics of wild type became available.
The rest of this paragraph only goes to reinforce that the author's opinion remains the attempt to control the spread is more important than keeping schools open.
3. “Modelling from Imperial College London suggested at the time that, had the government carried on with their “mitigation” strategy, around 250 000 people in the UK would have died from covid in the first wave.”
This modelling is now known not to have included voluntary changes of behaviour, despite huge and measurable changes occurring at least a week beforehand.
https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-021-10...
4. “But lockdowns are not public health policy. If anything, they represent a failure of public health policy.”
Agree. But any intervention that only modelled the worst case scenario of not taking action without also assessing the worst case scenario of taking that action will only represent the benefits of lockdown. There were always trade offs, and to not consider and model these has led many to believe these harms were insignificant.
5. “all of which would have helped to blunt, if not avoid, the alpha wave we experienced in the winter of 2020.”
We have seen time and time again that suppressing during the summer has led to a higher winter wave. We need only look at the difference between France and the UK in 2021 to aptly demonstrate the point.
6. “Many people would place the lives of the most vulnerable high on their list of priorities.”
Indeed. The restrictions they suffered will undoubtedly have hastened their death. An increase of frailty will reach a point incompatible with life. Not only did some vulnerable die alone, those who lived will have shorter lives due to sarcopenia and a faster rate of immunosenescence.
7. “Many people would value a functioning NHS with equal access for all at the point of need.”
The point of lockdown was to reduce the peak and prevent the overwhelm of critical care. The area under the curve would remain the same, the covid demand would just be spread more thinly. Thus access to non covid healthcare would be restricted for longer as a result of lockdown.
All of the points made suggest the residual belief that suppression would have been successful. I would like to suggest it could only delay, and that everything we have seen since confirms it.
If we had not locked down for as long in the first wave, chasing an impossible zero covid dream, I would suggest that our most vulnerable to covid would have been less affected by the increase of frailty, and some would not have suffered the torture of dying alone. For our children, an earlier return to school may have helped prevent the loss of 100,000 ghost children, the huge rise in mental and physical ill health (as seen in the control group of the CLoCk study) in addition to limiting learning loss.
The largest assumption, that Zero Covid was a valid strategy, at any point, needs to be thoroughly and forensically challenged.
Competing interests: No competing interests
Dear Editor
Kit Yates makes the claim that those that opposed lockdowns did so as a result of a value system that does not prioritise the lives of the most vulnerable, do not support a functioning NHS with equal access at the point of need, and are disinterested in the long term health of their population. The implication of this claim is that those who offered criticisms of the lockdown approach are motivated to do so by a value system that does not consider health of the most vulnerable, or of the general population, to be important.
It is striking that two years into the pandemic, there continues to be a refusal to consider any criticisms of the lockdown approach as being motivated by anything other than an uncaring selfishness.
However there are additional equally important values, rooted in principles of medical ethics, that might motivate a lockdown critical response. This includes having a policy approach that is rooted in patient (and population) autonomy, decisions being made with people rather than imposed upon them, ensuring that policies do not worsen economic and other inequalities, avoiding coercive medical practice (which is undermined by a willingness to invoke the criminal justice system to enforce "stay at home" orders), prioritising psychosocial determinants of health alongside the biological, understanding the limitations of our expertise and not, therefore, seeking to over-ride democratic principles, and a profound respect for civil liberties including freedom of assembly of people and freedom of movement.
Science and medicine thrives by encouraging open debate. There are many aspects of the lockdown policy approach which merit discussion - whether that be examining the efficacy of the lockdown interventions, the ethics of whether it is acceptable to invoke the criminal justice system to enforce one understanding of risk onto others, or how we protect the most vulnerable in our society while still respecting their autonomy and the diversity of opinion within this population group.
Categorising those that offer criticisms as being outside your value system, and implying that they simply do not care about the health of the most vulnerable or of the general population, will have the consequence of making it almost impossible to understand different perspectives. Scientific debate is stifled in such circumstances, learning is lost, which ultimately will have a detrimental impact on our patients and the broader health of our society.
Yours sincerely,
Robert Freudenthal
Competing interests: No competing interests
Dear Editor,
I read with interest 'Was lockdown necessary?' by Kit Yates, and although I found - as is often the case - a lot to agree on, there are two critical aspects to this article where I think he has fundamentally missed the point.
Firstly, the question "What does lockdown mean?" is central, yet is ignored by Yates, who focuses on timing alone. The term 'lockdown' has - I think - become what Laclau describes as an 'empty signifier' - something that means such different things to different people that it's no longer helpful - yet the article implies that lockdown is still a meaningful and well-understood intervention. I think it's vital that as well as timing, we also consider what form measures take, perhaps most obviously what is mandated/legislated versus advised, and who and where is the subject of restrictions. On the latter, to my mind and based on my research work, the way we have disproportionately harmed children through our measures is the most egregious aspect of this, and ought to be central to any discussion about the merits of past types of 'lockdown', and should also inform their future design/implementation. That pubs/bars were opened before schools for many children around the world ought to give us pause to think, and shows how much critical variation is disguised by the term 'lockdown'.
Secondly, Yates says that "Whether you view [lockdowns] as necessary depends on your value system", and implies that those who question the timing, design and merits of this blunt policy instrument do not sufficiently value the lives of the vulnerable, do not believe in the fundamentals of the NHS, nor do they care about long-term population health. To be blunt, these sort of shouts of 'agree with me or you are a nasty, uncaring person' are, I think, part of the root causes of where policy has failed on covid. To question - even politely and drawing on good science - distributional impacts of covid controls has been shouted down as 'enabling eugenics' or state-sponsored mass-manslaughter and to raise the spectre of non-covid harms has been dismissed as 'vacuous'. Nuance has at times been verboten, as it risked undermining trust in 'the science'. Accusations of 'gaslighting' have become so over-used as to also risk becoming as much an empty signifier as 'lockdown', but I think in this case I think it is appropriate.
If all we discuss is the timing, rather than the content, of 'lockdowns' we're going to really limit what we can learn from the last two difficult years. At the same time, while we can agree to disagree on policy prescriptions, to suggest that all those who interpret the science, ethics and, yes, politics of covid differently from you are uncaring people who do not share decent values with you is a dangerous over-simplification.
Yours sincerely,
Rob Hughes
Competing interests: No competing interests
Re: Was lockdown necessary?
Dear Editor,
Dr Yates’ article presents a one-sided case in favour of lockdowns which neglects many issues that represent thorns in the side of his arguments.
1. The biggest omission in Dr Yates’ argument is his lack of discussion of policies in Sweden – a country that did not lock down and yet has suffered fewer cumulative excess deaths than many other countries, including the UK, who imposed much stricter mitigation measures. This is true for both the first wave and the pandemic as a whole. There are many other examples of areas that did not lock down but experienced a plateau of cases well before the herd immunity level, including US states such as Florida in the second wave and afterwards.
2. Modelling figures from Imperial College London are used to make the claim that if a lockdown had not been put in place “around 250 000 people in the UK would have died from Covid-19 in the first wave.” This treats the modelling figures as a valid, realistic counterfactual, whereas time and time again they have proved to be too pessimistic (https://data.spectator.co.uk/category/sage-scenarios). As an example of this, in the summer of 2021, Professor Neil Ferguson who was an author of the Imperial College London model commented that “It’ll almost certainly get to 100,000 cases a day” when in reality UK cases never went above 50,000 during this period (https://www.theguardian.com/world/2021/jul/18/uk-covid-cases-could-hit-2...). Furthermore, the Imperial College London model is entirely unable to explain the dynamics of the pandemic in Sweden as this blog post argues.[1]
3. Dr Yates’ piece does not acknowledge the evidence that cases peaked before the imposition of the national lockdown. Inferring case numbers from data on deaths, I find that cases reached a peak around 3 days before March 23rd when the lockdown was announced.[2] Wood (2021) (https://arxiv.org/abs/2005.02090) also reaches this conclusion. A possible explanation for this is that there was widespread voluntary behavioural change amongst the population which placed downward pressure on the spread of Covid-19. This behavioural change explanation would clearly reduce the benefits of lockdown.
4. Dr Yates claims that there is a “strong argument to say that we should have locked down sooner, saving tens of thousands of lives”. However, there are examples of countries that did lockdown much earlier than the UK did and yet experienced a worse mortality outcome. For example, Peru imposed a strict lockdown 9 days after the first Covid-19 case, at which point no one in the country had died from the disease yet. This is in contrast to the UK, which did not impose a lockdown until 52 days after the first national Covid-19 case. Despite this, Peru experienced a much higher increase in cumulative excess mortality than the UK did during the first wave by a factor of more than five.
5. The costs of lockdown are not given sufficient discussion in Dr Yates’ article. While it is acknowledged that there were economic costs to the lockdown, no mention of any other type of costs is made. For example, there are clear costs to mental health that may be borne in an unequal manner. This paper (https://www.iza.org/publications/dp/14281/mental-health-costs-of-lockdow...) found strong evidence for such detrimental mental health impacts from lockdowns primarily due to heightened social and physical isolation. There is evidence of increased mental health disorders during the first wave in the UK (https://www.cambridge.org/core/services/aop-cambridge-core/content/view/...). There is also evidence of a very strong rise in childhood obesity during the pandemic in the UK, which was more pronounced for those in the most deprived areas (https://digital.nhs.uk/data-and-information/publications/statistical/nat...).
6. Dr Yates argues that “no evidence of the much-discussed tradeoff between protecting people’s health and protecting the economy”. This is highly debatable for many reasons. First, there is evidence that there was a negative correlation between a country’s number of Covid-19 deaths in the first wave and its cumulative fall in GDP in this period.[3] Secondly, the economic impacts of the lockdown may have long-term effects which are experienced far into the future and, as such, cannot yet be reliably measured. For example, there is strong evidence that job loss can have a ‘scarring’ effect, manifesting itself in lower earnings well into the future (https://christopher-huckfeldt.github.io/files/UTSEOR.pdf). Since many workers experienced redundancy in the UK during and immediately after the first wave (https://www.ons.gov.uk/employmentandlabourmarket/peoplenotinwork/redunda...), this is likely to be a relevant factor in the future. Finally, the government’s economic response that accompanied lockdowns resulted in an unprecedented rise in the ratio of government debt to GDP, as this went from 83% in March 2020 to 103.7% in March 2021. An increase in this debt burden brings with it trade-offs in and of itself, as a future decrease in government spending, rise in taxes or higher inflation may be necessary to deal with the issue. Once again, the costs of this will be borne for many years to come.
A balanced assessment of the impact of lockdowns needs to address all of the points raised above. Unfortunately, Dr Yates does not do this in his article. Producing such an assessment is of the utmost importance to inform policymakers in the future when similar pandemics arise.
Yours sincerely,
Andrew Preston
1 Lemoine P. Lockdowns, science and voodoo magic. Nec Pluribus Impar 2020 Dec 4.
2 Preston A. How Did Epidemiologists Get It So Wrong? Part 2 of 3. Andy's Newsletter. andypreston.substack.com 2021 Sep 6
3 Casey B. Covid-19: Is there a trade-off between economic damage and loss of life? 18 December 2020. blogs.lse.ac.uk
Competing interests: No competing interests