House of Commons - Scientific advice and evidence in emergencies

Scientific advice and evidence in emergencies - Science and Technology Committee Contents

5 Communication

111. Communication must be considered at all stages of risk assessment and in response to emergencies. A fast moving emergency exacerbates existing communication difficulties even as it multiplies the need for fast and effective information for responders and the public.

112. Previous crises such as the Bovine Spongiform Encephalopathy (BSE) outbreak and foot and mouth disease have highlighted the undisputable importance of good public communication, particularly on risk. The Phillips inquiry into BSE highlighted the following lessons on uncertainty and the communication of risk:

To establish credibility it is necessary to generate trust.
Trust can only be generated by openness.
Openness requires recognition of uncertainty, where it exists.
The importance of precautionary measures should not be played down on the grounds that the risk is unproved.
The public should be trusted to respond rationally to openness.
Scientific investigation of risk should be open and transparent.
The advice and the reasoning of advisory committees should be made public.
The trust that the public has in Chief Medical Officers (CMOs) is precious and should not be put at risk.
Any advice given by a CMO or advisory committee should be, and be seen to be, objective and independent of government.[121]

113. The Government and other organisations have to strike a balance between confidentiality and disclosure when preparing for, and responding to, emergencies: some information may be considered too sensitive for the public domain. One example is in the area of cyber security where specific vulnerabilities are not made publicly known for fear that they would influence the actions of attackers. The question of what information the Government is entitled to keep from the public (and Parliament) is, of course, much wider than our remit. However, we must point out that the Government has a duty to give the public information which they need to safeguard themselves, and in its policies and actions, the Government must be accountable to the public and to Parliament.

Principles of risk communication to the public

114. When a public risk is not communicated effectively by Government it can create mistrust and anxiety. As well as preventing this and raising awareness, the Government and other public bodies may also seek to encourage or discourage certain behaviours that could affect the outcome of the emergency response.

115. There is no lack of guidance to Government Departments on risk communication, including:

Communicating Risk,[122] a toolkit to help with planning communication strategies and developing understanding of risk (Cabinet Office);
Principles of Managing Risks to the public,[123] outlining five key principles applying to the handling of all types of risks to the public (Treasury);
The Orange Book,[124] which establishes the concept of risk management and provides a basic introduction to the concepts, development and implementation of risk management processes in government organisations (Treasury); and
Communicating about risks to public health: pointers to good practice,[125] designed to assist in the identification of public health issues which may create difficulties in communicating health risks and to provide guidance in risk strategies (Department of Health).

116. The Department of Health's guidance Communicating about risks to public health: pointers to good practice is useful in defining factors to consider when communicating risk, many of which can be applied beyond public health risks. The guidance makes the following points:

Crisis conditions—combining time pressure, unexpectedness, and high levels of threat—almost always militate against effective decision-making. A key defence against crisis is to spot possible difficulties in advance although one can never hope to spot all the relevant issues in advance: there will always be a need to "firefight".
A difficulty in risk communication is the difference between a "natural science" perspective and that typically held by a lay audience. Overcoming this is not merely a matter of explaining the science in lay terms—important though this is. An important difference is that scientists usually define risk in terms of effects on populations, while the lay audience is concerned with individuals. In addition, scientists usually will accept the existence of a causal link only once there is good evidence for it. Until then, links are "provisionally rejected". The lay view is much more likely to entertain a link that seems intuitively plausible, and only reject it if there is strong evidence against.
Public perceptions of risk are influenced by "fright factors", meaning that some risks trigger more alarm than others. For example, risks are more worrying if perceived to be involuntary, arising from an unfamiliar or novel source, poorly understood by science and/or subject to contradictory statements from responsible sources (or worse, from the same source). However, despite some common fright factors, "the public" is not a single entity. It is essential to consider different possible ways of seeing risks.
The single most important factor in risk communication is probably openness. This involves not only making information available, but giving a candid account of the evidence underlying decisions. If there are genuine reasons for non-disclosure of data, the reasons need to be given both clearly and early on. There should be a presumption in favour of disclosure.[126]

117. The Government has established the following five principles of risk communication to the public:

i. Openness and transparency: Government will be open and transparent about its understanding of the nature of risks to the public and about the process it is following in handling them;

ii. Involvement: Government will seek wide involvement of those concerned in the decision process;

iii. Proportionality and consistency: Government will act proportionately and consistently in dealing with risks to the public;

iv. Evidence: Government will seek to base decisions on all relevant evidence; and

v. Responsibility: Government will seek to allocate responsibility for managing risks to those best placed to control them.[127]

118. We examine how the Government communicated risks posed by the swine flu pandemic to the public, taking into account the principles outlined above, in the next section. We also kept the principles in mind when looking at the SAGEs set up for swine flu and volcanic ash (chapter 6).

Swine flu

THE "65,000 DEATHS" SCENARIO

119. Of the four case studies we explored, the swine flu pandemic posed the most interesting example of risk communication to the public. The complex and constantly evolving situation posed a number of challenges for Government, particularly communicating scenarios and projections. After explaining the reasonable worst case scenario (covered in chapter 4), Professor Neil Ferguson, Director of the MRC Centre for Outbreak analysis and Modelling, said:

we went from using, right at the beginning of the pandemic, that pre-existing reasonable worst case, to giving, effectively, what was an upper statistical confidence bound on our assessment of what the severity of the current pandemic was. That did not, perhaps, communicate as clearly as it should have done [...], particularly to the NHS. Those estimates got revised really quite rapidly, so within a month we were down from about that 2% level closer to 0.4% case fatality. Six weeks later it was down to below 0.1%-one in a thousand case fatality. So the estimates went downwards over time. [...] that posed significant communication challenges for the Department of Health, the Chief Medical Officer and the NHS.[128]

In addition to the difficulties of communicating changing scenarios, Professor Ferguson noted:

A further problem was that there was about a three to four week lag between the group I was involved in coming up with new reasonable worst cases, and then coming into the public domain in terms of getting through the DH and Cabinet Office approval process. So what was in the public domain as a reasonable worst case was already behind the evidence, given how fast the evidence was building up.[129]

120. On 16 July 2009, the Chief Medical Officer, Sir Liam Donaldson, held a press briefing that led to media reports suggesting up to 65,000 people in the UK could die from swine flu in a worst case scenario.[130] At that time, the number of actual deaths stood at around 30,[131] and by the time the pandemic was over in April 2010, the total number of UK deaths was 460. Dr Justin McCracken, Chief Executive of the Health Protection Agency (HPA), commented that:

it shows how difficult communication is because it was not just the reasonable worst case scenario that was communicated to the press. It was, actually, the range of both the best and the worst. But, inevitably, I think the figure that the press focused on was the worst case scenario.[132]

121. We asked Dr McCracken if, in future, he would recommend giving a mid-range figure instead of a range including extreme scenarios. He replied that "the difficulty of giving even a mid-range figure is the degree of uncertainty that is associated with it, but I think there probably is a case for that". He continued: "I don't think you can escape from communicating a reasonable worst case scenario that you are going to use for your planning in your health care system, but I do think that more emphasis needs to be given to what I would call the more likely expectation."[133] We discussed the press briefing with Sir Liam, who told us:

I spent a long time in that particular press briefing with the journalists, slightly short of pleading with them not to put out misleading information. Apart from one correspondent, they didn't contextualise the figure at all.

The modelling scientists would always say, "Well, even the 65,000 figure or figures like that held scientific water because those were the inputs that we had at the time", and then as they got more and more data, their number would come down and down and down. Unfortunately, that doesn't have much credibility with the public. They can't relate to that at all, understandably. So I think a great deal of care needs to be taken about the use of figures. I certainly felt that at the time. Even a back of the envelope calculation that I did suggested to me that we would get no more than a thousand deaths, but that was not the scientifically agreed figure. So I could hardly dissent from the bigger figure.[134]

122. There are three issues of concern here. First, it appears that, while scenarios were constantly being revised as more data became available, the communication of these updates were subject to delays, resulting in outdated information being provided to the public via government channels. Second, sensationalised media reporting may not simply be due to the press focusing on the worst case; we have misgivings about how clear the concept of the reasonable worst case scenario actually is, particularly as it does not emphasise the most likely situation. Third, the Chief Medical Officer, acting as "the messenger for the 65,000 figure which came from the scientific modellers"[135] was not confident in the figures he was communicating yet felt unable to dissent.

123. If, following the GCSA's Blackett Review, the concept of a reasonable worst case scenario is retained, we recommend that the Government must make continual efforts to establish the concept of "most probable scenarios" with the public. While the Government should be open about the worst case scenarios being used by emergency responders, it should use the experience of the 2009 pandemic to emphasise the range and likelihood of various possibilities. While we do not expect this to remove all the problems associated with communicating risk and uncertainty, we consider that it may provide the public with a better sense of the likely risks.

INFORMATION TO CLINICIANS

124. It is equally, if not more, important for central Government to communicate effectively with emergency responders. In the case of the swine flu pandemic we were alerted to the frustrations of clinicians by the British Medical Association (BMA), which stated that:

Doctors felt overwhelmed by the volume of information about the H1N1 pandemic issued by various bodies [including Government]. Key advice was lost within the large quantity of emails received, which often duplicated information.[136]

125. Dr Peter Holden, giving evidence on behalf of the BMA, told us that the four key sources of information; the Royal College of General Practitioners, the British Medical Association, the Health Protection Agency and Department of Health, cross-linked their websites, but he considered that:

we came unstuck because we were so keen to be up-to-date and offer timely advice, and it was a fast moving scene [...] I think what we should learn from this is that there is a review date on this advice, and you accept that the advice that may be on the website could be a few hours out of date in pure science terms.[137]

126. When communicating information in this situation, the organisations providing information to clinicians clearly had to walk a tight line between under-informing and over-informing clinicians. With hindsight, it is apparent that attempts to provide information in a timely manner were in fact overwhelming to doctors and insufficiently coordinated. We put the issue to the Sir Liam Donaldson, former Chief Medical Officer, who responded:

I think [the BMA] are a little unfair. [...] We also had regular contact with the BMA GP committee. [...] I think the idea that we over-communicated with them is a little unfair because, really, at other times they were saying to us, informally, "We need to know more".[138]

127. We consider that the risk of over-information could be mitigated by a single online portal of information. For example, in the USA, the flu.gov website provides comprehensive government-wide information for members of the public and professionals on seasonal, H1N1 (swine), H5N1 (avian) and pandemic influenza.[139] It also includes links to specific information for families, businesses, and schools.

128. We recommend that there should be a single portal of information for every emergency, along the lines of flu.gov in the USA. This should be of use to members of the public as well as emergency responders and should be the primary source of all information, linking to other websites as necessary. We consider that maintaining this portal should be the responsibility of the Lead Government Department, and should be located within its departmental website.

FROM PANDEMIC TO SEASONAL FLU

129. In paragraph 13 we mentioned the resurgence of swine flu virus during the 2010-11 winter—commonly the season for flu—and that the Government's vaccination strategy differed from when the virus first emerged in the UK. However, even during the pandemic the vaccination strategy changed. The initial strategy, in August 2009, was to vaccinate priority groups including pregnant women, frontline health and social care workers and people in at-risk groups over six months. In November 2009, phase two of the vaccination programme began and expanded to include children over six months and under five years. Professor Neil Ferguson, told us that:

if you have vaccine available really quite early in an epidemic, then targeting the people who transmit the disease, and in this case had we been able to target all school-aged children, for instance, all the way back in August, then we probably wouldn't have had an autumn wave to this epidemic. We would have stopped transmission.[140]

On the decision to expand the vaccination programme, he stated:

I have to say that I was, perhaps, a little surprised by that. That was not something that went to the committee that I sat on, the SAGE Committee. It may have been discussed by other advisory groups in the Department of Health, but it was always going to be of marginal impact given that the epidemic was already largely over. I worried myself that it would lose credibility—that people would already view this not as a threat, so what was the justification for doing it?[141]

130. During the 2010-11 flu season, swine flu was being treated as a seasonal, rather than pandemic flu virus. There was particular media attention on the decision in 2010-11 not to vaccinate healthy children.[142] The Government received advice from the Joint Committee on Vaccination and Immunisation (JCVI). On 30 December 2010 the JCVI met to review its advice on seasonal influenza vaccination. It produced the following statement:

JCVI was presented with data on the current seasonal influenza epidemiology, seroepidemological data collected during the 2009-10 pandemic, modelling of the impact of vaccination strategies during the pandemic, data on the effectiveness of influenza vaccines in the young and vaccine uptake and safety data.

JCVI noted that a large proportion of those individuals with severe disease are in recognised risk groups for influenza but were not vaccinated. JVCI re-iterated its previous advice that all individuals in risk groups should be vaccinated as soon as possible, particularly those aged less than 65 years.

The [JCVI] considered the issue of offering vaccination to healthy children either 0-4 years and/or 5-15 years of age. However, although there is a high incidence of influenza-like illness currently in these age groups, a significant proportion of this is due to other viruses such Respiratory Syncytial Virus. In addition, only a very small proportion of those with severe disease are in these age groups. Based on previous seasonal influenza epidemiology it would be hoped that influenza circulation will have subsided within a month. We do not believe that seasonal or pandemic vaccine should be used for these or other healthy person groups. The greatest gain will be achieved in increasing vaccine uptake in the clinical risk groups.[143]

The Government also stated that:

No projections have been made of the number of deaths from swine influenza infection that may be prevented during the current influenza season by the current vaccination policy or an extension of that policy to include children under five years of age or other healthy age groups. Such projections, if conducted, would be highly uncertain as they would depend on a number of factors that are unknown or uncertain including, the existing immunity to swine influenza infection in different age groups of the population, the vaccination coverage in different groups of the population and how quickly immunity would accrue in these groups, and the effectiveness of vaccination.

As with all vaccination programmes, JCVI will keep this matter under review.[144]

However, the JCVI also noted that the size of the current outbreak was inconsistent with the level of population immunity that had been suggested by research done during the 2009 pandemic, implying that the immunity levels of young children were lower that had been anticipated.[145]

131. In response to criticism about the lack of a national advertising campaign, Rt Hon Andrew Lansley MP, Secretary of State for Health, stated:

We decided not to institute an autumn mass advertising campaign to encourage flu vaccination, because this would have wastefully focused on the entire population when only at-risk groups are being invited for vaccination. This does not mean that there was no campaign; GPs have been inviting those at-risk groups to receive the flu vaccine since October, and the lack of an advertising campaign this year has had no discernible impact on uptake of flu vaccine.[146]

132. Although the Government response to seasonal flu goes beyond our inquiry, we were interested in the ongoing public concern over the risks of swine flu as part of the seasonal flu outbreak. This is unsurprising, given the fresh public memory of the pandemic and the Government's 2009-10 pandemic communication programme, as well as the absence of a seasonal flu information campaign in 2010-11. The Government should carefully consider the public's assumptions about swine flu (or any new flu strain) when communicating the risks of that strain in the context of seasonal, rather than pandemic, outbreak.

133. We have concerns about the evidence on which the JVCI has based its advice to Government in relation to the 2010-11 seasonal flu vaccination programme. There is evidence that vaccinating children creates herd immunity[147] and it appears that in 2010-2011 the immunity levels of young children may not have been as high as originally anticipated. However, we accept that the evidence may not be clear-cut and that factors such as the efficacy of the vaccine in children and cost effectiveness must also be taken into consideration. We recommend that the JCVI conduct a comprehensive review of the benefits and risks of extending influenza vaccination programmes to all children under five, drawing on the experiences of countries, such as the USA, that already have policies of vaccinating under fives.

121 The BSE Inquiry: Findings and conclusions, October 2000, Volume 1, Chapter 14, section 1301; see also para 137. Back

122 Cabinet Office, UK Resilience: Communicating risk, www.cabinetoffice,gov.uk Back

123 HM Treasury and Cabinet Office, Principles of Managing Risks to the public, www.hm-treasury.gov.uk Back

124 HM Treasury, Orange Book: Management of Risks - Principles and Concepts, October 2004 Back

125 Department of Health, Communicating about risks to public health: pointers to good practice, January 1997 Back

126 Department of Health, Communicating about risks to public health: pointers to good practice, January 1997 Back

127 HM Treasury and Cabinet Office, Principles of Managing Risks to the Public, www.hm-treasury.gov.uk Back

128 Q 5 Back

129 As above Back

130 For example, "Swine flu could kill 65,000 in UK, warns chief medical officer", The Guardian, 16 July 2009, www.guardian.co.uk Back

131 "Swine flu could kill 65,000 in UK, warns chief medical officer", The Guardian, 16 July 2009, www.guardian.co.uk Back

132 Q 8 Back