Volume 11: Scientists after Southwood
4. The Spongiform Encephalopathy Advisory Committee (SEAC)
Lessons to be learned from the use of SEAC
Why have an advisory committee?
Conclusion

Why have an advisory committee?

4.730 Since 1996 a great deal of thought has been given, and is still being given, to how government should best make use of scientific experts and, in particular, expert committees. We have looked back on the use that was made of SEAC with this debate in mind. We have decided that there were a number of aspects of the use of SEAC which were unsatisfactory. These are conclusions reached with the benefit of hindsight. They are lessons to be learned from experience. They are not matters which call for criticism of those who were involved at the time.

4.731 Guidelines published by the Office of Science and Technology (OST) ¹ emphasise the importance of identifying as early as possible those issues for which scientific advice will be needed. If a scientific advisory committee is to be set up, it is advisable at the outset to give rigorous thought to those matters on which the committee is likely to be asked to advise. Unless this is done:

1. It will not be possible to ensure that the composition of the committee is tailored to the tasks that it will be required to perform.
2. It will not be possible to draw up terms of reference which restrict the roles of the committee to those that it is competent to perform.
3. It will not be possible to explain to those invited to serve on the committee the nature of the commitment that they are undertaking.
4. It will not be possible to define satisfactorily the relationship between the committee and the commissioning Department or Departments.
5. It will not be possible to identify the role and appropriate source of the secretariat.

4.732 Although there was discussion about the role that SEAC was intended to perform, this was not clearly defined. The Tyrrell Committee had left work in relation to research undone. They had published an Interim Report and drawn attention to the need for peer review of research projects and for a continuing overview of BSE research. It was the need to review the BSE research that was being undertaken and to look out for gaps and overlaps that first prompted the idea of setting up SEAC, but it was then agreed that the new Committee would have a wider brief to advise on 'other aspects'. It should respond to 'specific questions' which the Government might wish to put to it. Plainly it was difficult to foresee what these might be, but there were some aspects of what the Committee would need to consider that were capable of identification at the outset. When the agenda for the first meeting was being considered, Dr Pickles suggested that the committee should be invited to consider whether 'any other new information alters the perception about animal or human health risks as given in the Southwood Report'. This was not just an item for a single agenda. We think that the obvious, and probably the most important, ongoing task for the new committee was to keep under review, in the light of advances in knowledge, the conclusions set out in the Southwood Report.

4.733 That Report had been taken by the Government as the basis for its policy on BSE, and it was vital that the Government should be advised if any of the assumptions or conclusions in that Report proved to be unsound. These included:

1. the scrapie theory as to the cause of BSE;
2. the likely future incidence of the disease;
3. that it was likely that cattle would be a dead-end host for BSE; and
4. that it was most unlikely that BSE would have any implications for human health.

4.734 Had the task of keeping the Southwood conclusions under review been made an express part of SEAC's remit, this would have assisted the identification of the appropriate membership of the committee. It would also probably have led SEAC to produce clearer and more specific risk evaluation than that to be gleaned from their 1990 paper on the safety of beef and their 1994 Report on TSEs. ²

The composition of the Committee

4.735 Dame Deirdre Hine suggested to us that SEAC would have benefited from the inclusion in its membership of a medical epidemiologist and a human public health expert. With hindsight we can see that she was correct. Indeed, when Sir John Pattison took over the chair of the Committee in 1995, his first act was to strengthen the human health representation. By then concerns about the possibility that BSE was transmissible to humans were growing. What might, we think, have been appreciated from the outset was the desirability of having an epidemiologist on the Committee. Concerns about the epidemiology of BSE were repeatedly recorded in the minutes of SEAC meetings. Dr Tyrrell pointed out, correctly, that members of the Committee were familiar with the principles of epidemiology. He commented that to gain added value in veterinary epidemiology a person in the top class was needed and, until Dr Hueston became available in 1994, there was no one qualified to perform this role. That is a sad commentary on the importance attached at the time to epidemiology in the veterinary field. We do not believe, however, that the epidemiological expertise which SEAC lacked could only have been supplied by a veterinarian. The science of epidemiology is the same, whether of humans or animals, and an epidemiologist such as Professor Anderson, who was to join SEAC's epidemiology subcommittee much later, would have significantly strengthened the committee.

The terms of reference

4.736 The terms of reference provided that SEAC was to 'advise MAFF and DH on matters relating to spongiform encephalopathies'. Mr Lowson said that these wide terms of reference had the merit of flexibility. That is true, but the benefit of flexibility carried with it the disadvantage of imprecision.

4.737 When we asked Dr Tyrrell whether it was sensible to ask his Committee for their views on 'slaughterhouse practices', he responded that he felt that they ought to try to answer the questions that government asked and pointed out that the letters inviting new members to join SEAC asked them to advise on all matters relating to spongiform encephalopathies; the committee was not described as a Scientific Advisory Committee.

4.738 In 1997 MAFF and DH commissioned a review of SEAC. ³ This recommended that SEAC's terms of reference should be amended to 'explain the purpose of the Committee' and should provide that they were:

. . . to provide scientifically based advice . . . taking account of the remits of other bodies with related responsibilities.

4.739 It was plainly desirable that SEAC's terms of reference should have at least this degree of precision.

Role in relation to policy

4.740 Particularly careful thought needs to be given by government to the manner in which it wishes an advisory committee to contribute to the taking of policy decisions. Options include the following:

1. The committee advises on those ingredients of a policy decision that fall within its own areas of expertise but does not advise on which of the policy options should be adopted.
2. The committee discusses the pros and cons of each of the policy options but does not recommend which should be adopted.
3. The committee advises which of the policy options should be adopted.

4.741 When SEAC was set up, there was at least a possibility that it would advise on policy. Even so, little thought seems to have been given to how MAFF and DH would interact with SEAC. This was a question on which the Committee itself had something to say at its third and fourth meetings.

4.742 The minutes of SEAC's third meeting record:

It was recognised that [the Committee's] job was to assess scientific data and opinions as objectively as possible, then to set down judgements on these in writing. It was important to communicate the message that science was not absolute and it was for policy-makers to decide what measures to adopt. ⁴

4.743 This was not a message that got across to the Government.

4.744 At its fourth meeting SEAC returned to the theme. The Committee felt that one of its objectives was:

. . . to produce 'opinions' that set out clearly what was implied by scientific knowledge. It was for others to decide what policy decisions should flow from this. ⁵

4.745 In the event SEAC was seldom asked to do this. It was frequently asked not the scientific question 'what does this mean?', but the policy question 'what should we do?':

1. Should we advise farmers not to breed from the offspring of BSE dams?
2. Should we permit brains to be removed before head meat?
3. Should any changes be made to slaughterhouse practices?
4. Should porcine material be fed to cattle?
5. Should we permit animal protein to be fed to farm animals?
6. Should we permit MBM derived from SBO to be included in animal feed?
7. Should we permit tallow to be incorporated in animal feed?
8. Should we permit tallow derived from SBO to be incorporated in animal feed?
9. Should we permit MRM to be extracted from bovine vertebrae?
10. If BSE is transmissible to humans WHAT ACTION MUST WE TAKE?

4.746 On each of these questions SEAC was in a position to give informed scientific advice. Each one called for an evaluation of the likelihood of a risk of transmission of BSE to humans or to animals. But in most cases there were other factors to be weighed in the balance and it should have been government rather than SEAC which took the final policy decision. We can see the attraction of asking SEAC to provide the policy answer. It enabled the Government to say that they were following the best scientific advice. But this disguised the fact that the scientists were being asked to evaluate considerations which were not questions of science, and it raised problems as to whether and how government should provide input to SEAC's decisions.

4.747 The effect of asking SEAC to advise on the adoption of policy options was to put the Government 'in a box'. Once the advice was given, the Government had little option but to adopt it. In those circumstances it was desirable for government to put before SEAC all the relevant matters that needed to be considered when reaching a policy decision.

4.748 On occasion MAFF did more than this and advocated the option which SEAC ought to adopt. Before the experimental transmission of BSE to a pig, MAFF officials prepared a paper in which they invited SEAC to endorse their conclusion that it was satisfactory to include animal protein, even that derived from SBO, in pig and poultry feed. After transmission to the pig, they suggested to SEAC that in their view the preferable option was to ban the incorporation of SBO in animal feed. When in June 1995 the question of MRM came back before SEAC, MAFF gave the Committee a firm steer by stating that instructions to the Meat Hygiene Service and audit checks proposed by the Ministry would ensure that all spinal cord was removed before MRM was extracted.

4.749 We do not suggest that expert committees should never be asked to advise on policy options. What the BSE story shows is that, before asking such a committee to advise on a policy decision, careful thought should be given to the contribution that the committee is in a position to make to it. In most of the examples that we have set out above we feel that it would have been preferable if SEAC had been asked to advise on the risk of transmission involved, reserving to government the decision on the action to take in the light of that advice. If SEAC were to discuss the policy options, this would have been better done by dialogue with MAFF officials speaking to the papers that were put before the Committee.

4.750 The history of the response to BSE leads us to endorse the following OST Guidelines:

Scientific advice is only one element among the considerations which may need to be taken into account by decision makers, which might also include social, political, economic, moral or ethical concerns. Departments will need to judge how and at what stage the scientific and other concerns are to be brought together in the decision making process. Where it is intended that those offering the advice should take such concerns into account, departments should make it clear at the outset that this is the case.

When asking experts to identify policy options or to comment on policy options prepared by others, departments should respect the line between the responsibility of experts to provide advice, and the responsibility of departments for any subsequent policy decisions based on that advice.

4.751 We note the guidance or best practice recommended in the 'May Review' in relation to advisory committees dealing with food safety: ⁶

Advisory committees should, when appropriate, set out a range of risk management options for policy makers, together with their implications, to avoid placing unnecessary constraints upon the decision-making process.

SEAC did not adopt a guideline of this kind when asked to advise on food safety. Looking with hindsight, it would have been better if they had - particularly in March 1996. Indeed, matters would have been improved if the Committee had adopted such a guideline when advising not merely on food safety but on policy measures generally. However, we add a note of caution about the words 'when appropriate' in the quote above: it would not have been appropriate to expect a scientific committee such as SEAC to identify all possible risk management options. That would require input from others.

The secretariat

4.752 In helpful written comments provided to us on the role of advisory committees, Mr Meldrum submitted:

The secretariat of the committee should be totally independent of the commissioning department. It should consist of administrators not scientists. If there is a need for scientific input into the committee's considerations it should be obtained from the members themselves and not from any outside hidden source.

4.753 This is a delicate area, particularly where there are two commissioning Departments that do not always see eye to eye. We have noted the significant role played by Dr Pickles in the preparation of the Southwood and Tyrrell Reports and in SEAC's advice on the safety of beef. We have also expressed concern at the process of editing drafts of that advice, when officials sought to tone down any passages that might have given rise to public concern. Where a Department has a particular viewpoint on a matter that is before an advisory committee, this should be placed before the committee in a manner that is transparent, and the secretariat should not act as the Department's advocate. This does not lead us, however, to support Mr Meldrum's call for a principle that the secretariat should be totally independent of the commissioning Department. Dr Tyrrell told us that it is valuable for the secretariat to be sufficiently expert to be able to draft background papers and to act as a two-way channel of communication between committee and Department. Sir John Pattison said that at times the input from the secretariat was valuable and should be invited. We consider that the role played by the SEAC secretariat as a two-way channel of communication was important and it would have been a mistake to sacrifice this out of concern for the independence of the committee. We would, however, endorse the principle suggested by the OST's 'Consultation Document on a Code of Practice for Scientific Advisory Committees' that the secretariat should respect the independence of the committee's operational role.

Conflicts of interest

4.754 The OST Guidelines advise that:

Departments should ask prospective experts to follow the seven principles of public life as set out by the Committee on Standards in Public Life, which include the obligation to declare any private interests relating to their public duties. Departments should judge whether these interests could undermine the credibility or independence of the advice.

4.755 We do not subscribe to the view that potential conflicts of interest should necessarily disqualify someone from membership of an expert committee. Often the involvement which creates the potential conflict of interest gives the committee member particularly valuable expertise. It is, however, essential that potentially conflicting interests are declared. The interests of SEAC members were published in the form of replies to Parliamentary Questions. The 1997 MAFF and DH review of SEAC sensibly suggested that these should be published in a SEAC Annual Report.

4.756 It is important not only that committee members should register interests which may give rise to conflict, but that if any special circumstances give rise to an appearance of a conflict of interest for a member in respect of any item of business, the member should make that plain. We have noted a failure to observe that principle in relation to the formulation of answers to Mr Hogg's questions in 1995 (see below). ⁷

Remuneration

4.757 Members of SEAC who were not public servants were paid for the time that they spent on the work of the Committee. Having regard to the workload that membership of SEAC involved we think that it was reasonable that members should be paid.

Workload

4.758 Care should be taken that those invited to join a committee are given as realistic an estimate of the amount of work that will be involved as is possible. We believe that there is a tendency of officials to err on the side of optimism - this was certainly so in the case of SEAC. Care must also be taken to see that agendas are realistic having regard to the time available. If agendas are overloaded it is not realistic to expect that all items will receive the rigour of consideration that they deserve. SEAC's agenda was often overloaded. One occasion when this had serious consequences was when advice was given on slaughterhouse practices on the basis of impressions made on a visit to a slaughterhouse, rather than detailed consideration of the lengthy paper prepared by MAFF.

4.759 In order to make the best use of committee members' limited time, it may be necessary to select those questions on which their advice will be of the greatest value. Producing a paper in 1990 confirming advice already given by the CMO about the safety of beef was not the most profitable use of SEAC's time. Nor was the time spent in 1995 attempting to draft answers to the questionnaire that Mr Hogg wished to have answered in order to provide publicity material.

The manner of seeking advice

4.760 Careful consideration should be given to the advice needed from the advisory committee and to ensuring that the advisory committee is the best source of that advice. The advice requested should be targeted so as to avoid the risk of the answers being influenced by considerations falling outside the committee's expertise. We have set out in paragraph 4.738 above examples of failures to adopt this approach.

Application of principles of risk analysis

4.761 When exploring earlier this year the risk procedures used by the Government's advisory committees dealing with food safety, Sir Robert May's Group received answers from SEAC about their role as follows:

Role in risk assessment

Analysis and assessment of the risk to public health from exposure to TSEs, particularly BSE. Whilst much work is done internally SEAC also commissions advice from external contacts on specific issues. Its main area is risk assessment, which is informed by regularly reviewing research and drawing on scientific expertise. Use qualitative and quantitative, systematic risk assessment procedures.

Role in risk management

SEAC informs rather than drives decision-making. Risk management decisions may be influenced by factors that cannot be built into a formal quantitative risk assessment. SEAC offers advice on risk management.

4.762 These answers suggest that today SEAC is consciously applying principles of risk assessment and is not being asked to choose policy options. This is in marked contrast to the position during the period with which our Inquiry is concerned. As MAFF's 1997 Review of SEAC recorded:

SEAC's activities may sometimes seem more like those of a risk management committee than a risk assessment committee.

That Review also observed:

The Committee is conscious of the potential economic impact of its advice, and of the need - on questions where scientific certainty is not possible - to have regard to the practical consequences of possible courses of action.

4.763 This is clear recognition that SEAC was, in effect, choosing the policy options. We have been unable to find from first to last any mechanism for ensuring that SEAC was informed of the potential economic impact and practical consequences of its advice, or any conscious application of the ALARP principle of risk management. ⁸

4.764 During the period covered by our terms of reference, SEAC often did not clearly distinguish between risk assessment and risk management. Our examination of that period provides strong support for the statement of best practice in the 'May Review':

Advisory committees will usually be helped by following a formal structure for the process of risk assessment, even when the scientific facts are cloudy, disputed or even unknown.

The form of advice

4.765 In some cases SEAC went to great lengths to set out the reasoning behind their advice - the explanation of their conclusions on breeding from the offspring of BSE dams and their 'beef is safe' advice are examples. In other cases they gave no formal explanation - their advice on slaughterhouse practices is an example of this approach. It is also an example of the danger of misunderstandings where advisory committees do not explain their reasons in detail. 'It is important that committees make clear to policy maker the assumptions and uncertainties underlying their advice.' ⁹ It is, of course, essential that the advice itself should be clear.

Circulation of advice within government

4.766 It was not clear that SEAC's views - especially when recorded only in minutes of meetings - were always circulated to all those in MAFF and DH who might be affected. The evidence as to circulation outside MAFF and DH, and in particular to the Welsh, Scottish and Northern Ireland Departments, was confused. Our examination of both formal and informal advice given by SEAC suggests a need to identify, either generally or on specific issues, all those within government with responsibility for policy decisions to which the advice is relevant. They can then be sent relevant minutes and more formal 'advices', and any papers necessary to understand them. It was particularly desirable that the Agriculture and Health Departments in Wales, Scotland and Northern Ireland be kept abreast of SEAC's proceedings.

Openness and transparency

4.767 The extent to which the public should be informed of the deliberations and conclusions of advisory committees is part of a wider debate about open government. Mr Gummer introduced the principle that the formal advices of SEAC should be made public, but this was normally delayed until the Government was in a position to announce the measures to be taken in response to the advice. The advices were published by the Government and not by SEAC. It was not, however, the practice to make public the matters on which SEAC had been asked to advise. Thus when UKASTA was seeking to convince MAFF of the merits of an animal SBO ban, and MAFF officials and Ministers were arguing that there was no justification for this, it was not thought appropriate to inform UKASTA that this question had nonetheless been referred to SEAC.

4.768 We have remarked upon the general impression that up to 20 March 1996 the Government was not being open with the public about the risk posed by BSE to humans. We have also, with hindsight, queried the wisdom of Mr Hogg's questions to SEAC about the safety of beef. Had the Committee given the answers he sought, the process of seeking and publishing them would have been likely to damage public faith in SEAC's objectivity. If SEAC had itself adopted a practice of making public details of advices given to government, together with the reasons for them, we consider that this would have gone some way to reassuring the public that information about BSE was not being and had not been concealed from them.

4.769 The OST Guidelines recommend:

Departments should ensure their procedures for obtaining advice are open and transparent. The evidence upon which the advice is based should be published. The analysis and judgement which went into it, and any important omissions in the data, should be clearly documented and identified as such. Any claims for material to be protected, e.g. on grounds of the commercial confidentiality of the information concerned, should be rigorously tested.

Departments should ensure that data relating to the issue are made available as early as possible to the scientific community, and more widely to enable a wide range of research groups to tackle the issue and to provide a check on the advice going to government. This will be particularly important, for example, where the advice will rely on research which has not been peer reviewed, or which has not been previously published.

4.770 The history of scientific advice concerning BSE shows the wisdom of these recommendations, but they leave unanswered some difficult questions:

1. Should the agenda of meetings of advisory committees be published?
2. Should the minutes of meetings be published?
3. Should meetings of advisory committees take place in public?

4.771 There is inevitably a tension between being open about the details of the discussions of advisory committees and maintaining a lack of inhibition on the part of those involved in those discussions. Both Dr Tyrrell and Sir John Pattison believe that there is a need for a degree of confidentiality. They expressed the view to us that meetings should not be held in public, or at least not the whole of them, and minutes should not be published.

4.772 We do not find that the BSE experience provides a clear answer to the question of where precisely freedom of information should give way to the pragmatic requirements of confidentiality.

4.773 Witnesses representing consumer interests made the point that a lay member can play a valuable role on an expert committee, and in particular can ensure that advice given by the committee addresses the concerns of, and is in a form that is intelligible to, the public. There is force in this.

Conclusion

4.774 The reader of this lengthy list of lessons to be learned from the use of SEAC may have reached the end with the impression that little went right about the assistance provided to the Government by SEAC. That would be a false impression. SEAC was a diligent and conscientious body of professionals of high calibre. Whether or not they made use of techniques of risk assessment, whether or not the advice sought from them was satisfactorily targeted, whether or not they were effectively making policy decisions which should have been reserved to government, they made a significant contribution to the protection of human and animal health.