Читать книгу The Challenge of Controlling COVID-19 - Lewis Jane - Страница 10

The main debate in the literature so far has centred on the failings of scientific advisers and politicians with the inevitable tendency to blame one party more than the other. But the evidence suggests that there are some difficulties with taking a clear-​cut position on this. There was no one ‘scientific view’ on offer and as Salajan et al (2020) have pointed out, while decision makers may prioritise expert advice, they are challenged by scientific uncertainties. The UK Government did not ignore the views of scientists, but they may have had difficulty using often far from united scientific opinion to reach the policy decisions that they, and not the scientists, were responsible for taking. In addition, there may also have been limitations as to the nature of the advice that was offered by those sitting on the bodies that advised Government, probably due in part to the relatively narrow range of specialities represented.

Historically, epidemics have been controlled by public health doctors and their staff at the local level, exercising the authority provided by law, and using above all the well-​established tools of testing, tracing contacts, and quarantining/​isolating.² The central public health body involved in the control of COVID-​19 was Public Health England (PHE), set up in 2013 by the Conservative/​Liberal Democrat Coalition Government as an arms-​length executive agency with operational autonomy, but responsible directly to the Department of Health and Social Care (DHSC) and the Secretary of State for Health and Social Care. PHE was created by the 2012 Health and Social Care Act following the then Secretary of State’s (Andrew Lansley) major reorganisation of the NHS and has operated alongside a large number of other central health organisations, most importantly NHS England (NHSE), which has much greater autonomy.

However, the main source of scientific advice given to Government was SAGE (the Government’s Scientific Advisory Group for Emergencies), which changes its membership depending on the nature of the emergency. In the event of a pandemic, SAGE was intended to act in collaboration with the Chief Medical Officer (CMO), Professor Chris Whitty, a civil servant and thus statutorily independent, politically impartial and located in PHE, and the Chief Scientific Officer (CSO), Sir Patrick Vallance, who is head of profession for the scientists working in the NHS. Both the CMO and the CSO are members and chairs of SAGE. Advice to Government was filtered through the CSO and the CMO.

SAGE considered a range of options and played a major role in shaping the Government’s response, but it did not make policy recommendations. Its ‘consensus statements’ did not present a single view, but rather put forward assessments as to the nature, degree and range of uncertainty. The CMO explained to the House of Commons’ Science and Technology Committee that he aimed to put forward the Group’s ‘central view’ and to convey the range of uncertainty to Government (Clark, 18 May 2020). But the way in which policy decisions emerged tended to be opaque. Indeed, on 22 May, Sir Paul Nurse, Director of the Francis Crick Institute (a biomedical research institute, established through a partnership between a UK Government funding agency, two charities and three universities) saw fit to ask who was formulating strategy, what was the relationship between SAGE, PHE and politicians, and who was bringing evidence and action together? (Nurse, 22 May 2020, BBC Radio 4 Today interview). He repeated his concern about secrecy in respect of decision-​making in an interview with The Guardian (Sample, 2 August 2020), when he was quoted as saying that ‘it sometimes seems like a “black box made up of scientists, civil servants and politicians are coming up with the decisions”’.

Both ministers and scientists have rehearsed the well-​worn dictum that scientists advise and ministers decide, but the precise way in which decisions on action and possibly strategy (this tended to be particularly opaque) were arrived at has not been clear and may have made the passing of blame onto scientists (and civil servants) easier, not least because scientists often stood alongside politicians at the regular Downing Street Briefings on COVID-​19. However, standing side by side at briefings does not necessarily indicate a shared understanding. Furthermore, the membership of SAGE and its sub-​groups, the papers submitted to it and the minutes of its meetings remained secret until 29 May,³ by which time the role it was playing had become the subject of media speculation, with some scientists objecting to the redactions in the official record of their proceedings.

What is clear from the publication of the membership of SAGE is that the Group was dominated by epidemiologists doing mathematical modelling to determine the course of the disease. Field epidemiologists, public health practitioners and social care experts were notable by their absence. But, while the mathematical modellers on SAGE were aware early on of the risks to elderly people in care homes, they did not know how care homes worked, for example in terms of their reliance on agency staff who were likely to work in several different homes and might thus pass on infection. Lord O’Donnell, Cabinet Secretary between 2005 and 2011, also suggested that given the effects of the pandemic on the economy, more social scientists should also have been members of SAGE and its sub-​groups (O’Donnell, 2020; see also Cairney, 2020), although the views of economists reach the Treasury and the Chancellor and then the Cabinet. Nevertheless, it may be that bringing together scientists and economists in a single forum would have helped to clarify the balance of risks.

SAGE seems to have agreed on 4 February and again on 11 February that the Government should plan by using ‘influenza pandemic assumptions’, which could be modified as the data on the pattern of the disease became more certain. This is significant because the influenza outbreak in 2009 had been successfully handled (Hine, 2010), and planning had taken place for a further influenza pandemic. Indeed, the Government’s Coronavirus Action Plan (DHSC, 3 March 2020) echoed parts of the Influenza Pandemic Preparedness Strategy of 2011 (DH, 2011).⁴ The Coronavirus Action Plan reiterated that the UK ‘is well prepared for disease outbreaks … Our plans have been regularly tested and updated…’ (DHSC, 3 March 2020, para 3.2), which was true in that exercises had been carried out, notably the three day Exercise Cygnus in 2016.⁵ However, the recommendations following this Exercise, particularly regarding the supply of beds, ventilators and masks, were not followed under the conditions of austerity that prevailed after 2010.

In addition, the Government’s strategy for dealing with COVID-​19 outlined by the Plan was also modelled closely on dealing with the expected influenza pandemic. Like the 2011 influenza strategy, the Coronavirus Action Plan described four phases: containment, involving detection of cases and follow-​up of contacts; delay, involving measures to slow the spread of the virus, ‘pushing it away from the winter season’; research; and mitigation by providing the best care for the sick and supporting hospitals and communities (DHSC, 3 March 2020, para 3.9). This Plan did stress the importance of tracing and isolating contacts, but stated that there would be ‘less emphasis’ on measures such as intensive contact tracing in the mitigation phase, because ‘as the disease becomes established these measures may lose their effectiveness…’ (ibid., para 4.48). This echoed the Influenza Strategy, which also confined the importance of ‘test and trace’ to the first phase of the pandemic. Readers were also assured that stockpiles of ‘the most important medicines and protective equipment for healthcare staff … are being monitored daily’ (ibid., para 4.14), which was later shown not to have been the case.

SAGE also commented on issues that were central to the control of COVID-​19 at an early stage. There are examples of advice that seem to back up what the Government chose to do or not to do and also enable us to see how uncertainty played into the decision not to act, for example regarding two of the issues outlined above –​ the decisions not to stop big sporting events and not to impose quarantine regulations. In respect of the first, the Scientific Pandemic Influenza Group on Modelling, Operational sub-​Group (SPI-​M-​O) concluded on 11 February that the direct impact of stopping large public gatherings on the spread of the virus at the population level would be low, but that social interaction in bars and restaurants might ‘slightly accelerate’ the spread. Of course, attendance at large-​scale sporting events was often also accompanied by visits to bars. This ‘consensus view on public gatherings’ was passed on to SAGE (Meeting 12, Minutes 3 March 2020), which added that it would be difficult to stop interaction in bars and restaurants. In the event, Government chose not to cancel large events and nor did it place any limitations on bars and restaurants. It is possible that the necessarily equivocal nature of much of the advice offered by SAGE lulled ministers into a lack of urgency. Or there may have been a lack of capacity on the part of Government to ask the further questions that would have been necessary prior to making difficult policy decisions. Yet again, the Government may have been straightforwardly reluctant to interfere with the everyday lives of citizens.

On quarantine, SAGE considered this early on, before the Prime Minister became actively involved in plans for controlling the virus. On 3 February, SAGE estimated (as it admitted on the basis of ‘limited data’) that if the UK reduced imported infections by 50 per cent this would ‘maybe delay the onset of any epidemic in the UK by about 5 days’. Infections would have to be reduced by 95 per cent to buy a month of time and ‘only a month of additional preparation time for the NHS would be meaningful’. But buying that amount of time would require ‘draconian and coordinated measures, because direct flights from China are not the only route for infected individuals to enter the UK’ (Meeting 3, Minutes 3 February 2020). On 23 March, SAGE minutes recorded the conclusion that closing the UK borders ‘would have a negligible effect on spread’ of the virus, because by then the number of cases arriving was insignificant compared to the domestic infection rate (Meeting 18, Minutes 23 March 2020). However, in response to a Home Office request for advice on what to do about border restrictions at the end of April (Home Office, 2020), SAGE stated firmly that determining ‘a tolerable level of risk from imported cases … is a policy question’, thus excluding it from consideration (Meeting 29, Minutes 28 April 2020).

SAGE’s advice was only as good as the quality of its data, and as Professor Neil Ferguson admitted later, this was poor in the early part of the pandemic: in particular, it was not known how far COVID-​19 was ‘seeded’ from many of the people arriving from Italy and Spain (Ferguson, 10 June 2020). As the House of Commons Home Affairs Select Committee (5 August 2020, p 3) commented: ‘The UK’s approach [to border control] was highly unusual’ compared to other countries, but the advice from scientists at this stage did not support firm action. However, the Home Affairs Select Committee reported that it asked to see the advice underpinning the Government’s decision to impose quarantining for an expanding number of countries up to 12 March and then for abandoning all such measures, but, receiving no answer, suggested that it was reasonable to conclude that such advice did not exist. The Home Office News Team (5 August 2020) denied that this was the case.

SAGE became more explicitly careful about offering advice on policy as a result of its findings as time went on. But politicians wanted it to be thought that they were acting on scientific advice. Thus, when he introduced the Coronavirus Action Plan the Prime Minister said: ‘the plan does not set out what the government will do, it sets out the steps we could take at the right time on the basis of the scientific advice’ (Johnson, 3 March 2020). However, the tensions between political imperatives and scientific understandings (there was never only one) became sharper over time.

SAGE’s main focus was modelling the course of the pandemic and the findings on this seem to have had a major influence on politicians in mid-​March, especially in respect of changing the Government’s belief in a pandemic with moderate effects, to the possibility that it could result in high mortality. There is also evidence to suggest that some toyed with the possibility of achieving herd immunity by allowing a large proportion of the population to become infected, even if this was not a position taken by the vast majority of people in the Group. Building on the experience of controlling influenza may have played a part in any such thinking. On 13 March, the CSO had told an interviewer on BBC Radio 4’s Today programme that the Government’s strategy had in part been ‘to build up some herd immunity’. However, the next day the Secretary of State, Matt Hancock, denied that herd immunity was ever considered as a practical possibility (Hancock, 14 March 2020). In fact, a policy to allow the virus to spread among younger people while attempting to shield the vulnerable⁶ was not only likely to fail, but also to result in the NHS becoming ‘overwhelmed’ and, as Professor Devi Sridhar (24 December 2020) has commented, neglected to take account of possible mutations in the virus. In addition, as Mark Carney (2020), former Governor of the Bank of England and the BBC’s Reith lecturer has remarked, ‘the pathway to herd immunity runs directly through the inequalities in our society’. For as Marmot et al (2020) have documented, inequalities have increased both in terms of extent and depth. People in low-​wage, precarious employment, people in poverty, people in overcrowded accommodation and black and minority ethnic people have all suffered disproportionately high death rates (see also Bambra et al, 2020; Tinson and Clair, 2020).

One of the main problems with the confidence engendered by building on the way in which the 2009 influenza outbreak had been tackled was the extent to which the COVID-​19 virus behaved in a very different way. What turned out to have been important at the beginning of the pandemic was the speed of transmission; however, the number of tests carried out was too small to be a useful tool in analysing its spread. SAGE signalled the need for an early warning surveillance system in mid-​February (Meeting 8, Minutes 18 February 2020), but the Office for National Statistics (ONS) was not asked by the Government to provide this until mid-​April. Data problems may also have affected the early advice SAGE gave on crucial issues such as closing borders.

In the event, the ‘containment’ phase of the pandemic proved relatively short. The main catalyst for the Government’s startling u-​turn on 12 March, which signalled a move to the ‘delay’ phase and with it lockdown, seems to have been the findings by the modelling group at Imperial College under the direction of Professor Neil Ferguson which were submitted to SAGE in early March, warning that up to 250,000 people might die without drastic action and that with no action the death toll could be twice that (Imperial College Response Team, 16 March 2020). However, the 27 February Minutes of SAGE (Meeting 11) show that the Group had already warned that the UK could face an 80 per cent infection rate and a 1 per cent mortality rate (and, importantly, also warned that only a proportion of those infected would manifest symptoms), although the CMO seemed to suggest at the Downing Street Briefing of 3 March that moving to lockdown too early would risk ‘fatigue’ on the part of the public. It is impossible to know at this point whether the Government, the PM included, ignored SAGE’s warnings in late February and during the first two weeks of March, whether SAGE chose not to draw attention to them, or whether the views of SAGE were accurately relayed to ministers. On 16 March, SAGE advised that additional interventions, such as school closures were needed, because the number of cases was thought to be doubling every five to six days (Meeting 16, Minutes 16 March 2020). It was finally decided that doing little or nothing would exact too heavy a price in terms of mortality and, given the extensive spread of the virus in the UK, the only option was complete ‘lockdown’. This was announced a week later on 23 March in the hope of stopping the NHS from being overwhelmed. This was a risk that had been intensified by the way in which austerity had affected hospitals after 2010; for example, occupancy rates for intensive care beds were already running at or just above recognised safe levels before the pandemic started.

The UK was the last country to enter lockdown in Europe; Italy also locked down late on 10 March, but nevertheless had fewer confirmed cases when it did so. As a comparative Imperial College-​led study has shown, the timing of lockdown in relation to when initial infections occurred affected the peak number of people infected, which drives both the number of COVID-​19 deaths and the pressure on the healthcare system, which in turn limits the capacity of the NHS to treat other diseases (Kontis et al, 2020). Peak infection actually occurred in England before lockdown. The delay in locking down, even as little as the week between SAGE’s recommendation for additional measures and the Government announcement on 23 March, resulted in an increased number of deaths and had an adverse effect on the efforts to suppress the virus (Colbourne, 2020), which had become particularly widespread geographically in England compared to other Western European countries (Spiegelhalter, 2 August 2020). The reasons for the delay were possibly related to the Government’s lack of urgency and over-​confidence, but its relationship with its scientific advisory body was also difficult. The modelling done by SAGE required good data (and sound assumptions). Its recommendations regarding the effectiveness of possible measures to control the virus demanded high standards of evidence –​ higher than were often available –​ and its conclusions were thus often equivocal. Above all, the Government was looking for policy solutions which the mathematical modellers who dominated SAGE did not have, and in any case regarded as strictly political territory, while public health practitioners, who might have made a valuable contribution –​ particularly on operational issues at the local level –​ were not consulted. Nor was there any forum in which a balance sheet of the possible impacts on the population in terms of health, but also economic performance, could be drawn up. The Government stressed throughout that it had ‘followed the science’ (although by the end of April, as tensions grew, the phrase was changed to being ‘guided by the science’ [Torjesen, 17 July 2020]), alongside an insistence that decisions had been taken at ‘the right time’.

By May, tensions between SAGE and the politicians were being referred to openly. Jeremy Hunt, Chair of the Select Committee on Health and Social Care (and Conservative Secretary of State for Health from 2012 until 2018), blamed the Group for lack of attention to what had happened in other countries, describing their efforts in terms that echoed those of Richard Horton (2020, p 41), editor of The Lancet, as ‘one of the biggest failures of scientific advice to Ministers in our lifetimes’ (HoC Debates, 11 May 2020, vol 676, col 59), albeit that Horton blamed scientists for colluding with Government rather than misleading it. On 19 May, Therese Coffey, Secretary of State for Work and Pensions, told Sky News that ‘if the science was wrong, advice at the time was wrong, I am not surprised people think we made the wrong decisions’. This prompted Venki Ramakrishnan, then President of the Royal Society, to comment that:

…it is not possible for scientists to give frank advice if they feel that they will be made the scapegoats for difficult policy decisions … Governments not only have to contend with the uncertainty of the science but a host of other practical considerations, including feasibility. In all this they want certainty from scientists –​ and feel or claim that they are “following the science” –​ but wishing something does not make it so (Ramakrishan, 24 May 2020).

A very clear break between scientists and politicians was to come in late September, when SAGE recommended a two-​week ‘circuit breaker’, that is, a short lockdown, alongside a package of further interventions (Meeting 58, Minutes 21 September 2020), but the Government decided not to go this far and instead confined its actions to tightening existing regulations, for example, by ordering the shutdown of bars and restaurants at 10pm.

***

The Government may have had unrealistic expectations of SAGE. The dominant voice among the experts on SAGE belonged to those modelling the virus, with the Group also playing an important role in reporting on whether a particular approach to disease prevention met their high standards of evidence. They relayed their findings via the CMO and the CSO to the Government. They did not make recommendations based on observations, for instance on what was happening in other countries, and they did not engage with operational issues, such as how to set up a test and trace system and make it work. As Nickson et al (2020) have suggested, ministers need to be clearer about SAGE’s remit, the role of science advice and its limitations.

In addition, in the early stages of the pandemic, the data available to influential members of SAGE was inadequate and the problem this posed was compounded by the Government’s tendency throughout the first wave of the virus to postpone taking decisive action. It may also be that the Government should have widened its group of advisers, most obviously to include public health practitioners, and that their means of engaging with scientific advisers and public health and social care leaders required modification, although as the next two chapters show, ministers showed little appetite for consulting with people working on the ground. A Government facing a new disease that was spreading extremely quickly wanted solutions, but SAGE’s advice often spelled out the uncertainties, denying ministers the certainties they desired (for instance on whether to cancel large public sporting events and close borders). This was not the fault of the scientists whose views were sought via SAGE. Nor can the Government necessarily be blamed for hoping for and wanting more, although as the President of the Royal Society intimated, it showed inadequate understanding of the nature of scientific advice.

Inevitably, given the rapid spread and high number of cases requiring hospital treatment by mid-​March, attention became more firmly fixed on treatment and in particular on the danger that the NHS would be ‘overwhelmed’ than on the need for further urgent development of measures to control the pandemic by developing the role of public health.