Case number: 020358

Our Reference : 020358                       
1 October 2002
Dear X
I refer to your application under the Freedom of Information (FOI) Act, 1997 for a review of the decision of  the Food Safety Authority of Ireland (FSAI) on your request of 25 February 2002 for records concerning fluoride in water including the FSAI's involvement with the Forum on Water Fluoridation in Ireland (the Forum). In your application of 1 July 2002 to my Office you sought reviews in relation to two separate FOI requests made to the FSAI; however, for reasons which become clear below, it is only necessary now to proceed with a review in one of these cases i.e. in the case of the request made on 25 February 2002.
The FSAI was one of several parties represented in the Forum, a body established in May 2000 by the Minister for Health and Children to review, report and make recommendations to the Minister on the fluoridation of public, piped water supplies. As you know, the FSAI responded to your request for access to particular records relating to its involvement with the Forum by granting access to some records and withholding access in full or in part to the remainder. Following discussion with Mr  Fintan Butler, Senior Investigator in my Office, you agreed on 15 August 2002 to narrow considerably the range of records you wished to have included within the review. In fact, you agreed that the review should be confined to the question of your right of access to the original draft of the document entitled "Risk assessment of the fluoride intake of Irish infants under 4 months of age as a result of the consumption of infant formula re-constituted with fluoridated water" (the "Risk Assessment Report"). You informed my Office that this document was initially referred to in the minutes of the fourth meeting of the FSAI sub-committee on Additives, Chemical Contaminants and Residues in May 2001 and circulated to the members of that sub-committee in October 2001. The FSAI subsequently explained that there are, in fact, two versions of the draft Risk Assessment Report; identified them as items in File 20 (under FSAI reference FOI 018) and File 23 (under FSAI reference FOI 019) in the schedules that it sent to you with its decisions on your original requests. These two records are amongst those which the FSAI provided to my Office for the purposes of this review. I have proceeded on the basis that your revised review application covers both drafts of the Risk Assessment Report.
In view of some of the arguments considered later in this decision, it is important to be clear as to the context in which the FSAI decided to undertake the Risk Assessment Report. I understand that concerns were raised at the Forum regarding the possibility of a risk to infants whose infant formula is re-constituted with fluoridated tap water. In the light of these concerns, I understand the FSAI asked its Scientific Committee to conduct a risk assessment to determine if such a risk exists. Once this risk assessment was completed, the FSAI made it available to the Forum. However, as I understand it, the risk assessment was not in any sense commissioned by the Forum or by the Minister for Health and Children, under whose aegis the Forum was convened. 
At the time of the internal review decision on your original request, on 24 April 2002, it appears that the Risk Assessment Report had been completed by the FSAI's sub-committee and was before the Forum. Since then, as recently as 10 September 2002, the Forum's final report has been published by the Minister. I understand that the findings and recommendations of the FSAI, in relation to infant formula and fluoridated water, are incorporated into the Forum report and that the FSAI's Risk Assessment Report is reproduced in full at Appendix 18 of the published Forum report.
I have now completed my review of the FSAI's decision. In carrying out this review, I have had regard to your letter of application to my Office as well as to subsequent contacts and discussions between yourself and members of my staff. I have also taken account of the submissions of the FSAI as well as of a number of contacts between the FSAI and members of my staff; in particular, I have taken account of a telephone contact on 28 August 2002, between Fintan Butler of my Office and [  ] the FSAI, in which Mr. Butler outlined his preliminary views on the applicability of the various exemptions being claimed by the FSAI. Finally, I have also taken into account the content of the records and the provisions of the FOI Act generally.
Scope of the Review
This review is concerned solely with the question of whether the FSAI's decision, to refuse access to the two drafts of the Risk Assessment Report, is justified in accordance with the provisions of the FOI Act. 
The FSAI relied in its decision in relation to the two records on the exemptions in sections 20(1), 21(1)(a) and 30(1)(a) of the FOI Act. In addition to these provisions of the Act, which I deal with below, it is important to note that, in accordance with section 34(12)(b), a decision to refuse to grant a request is presumed not to have been justified unless the head of the public body shows to my satisfaction that the decision was justified.
Section 20
Section 20(1) provides that a head may refuse to grant a request 
" (a) if the record concerned contains matter relating to the deliberative processes of the public body concerned (including opinions, advice, recommendations, and the results of consultations, considered by the body.......), and 
(b) the granting of the request would, in the opinion of the head, be contrary to the public interest".
Section 20 (2) provides, inter alia, that "subsection (1) does not apply to a record if and in so far as it contains-
"(e) a report, study or analysis of a scientific or technical expert relating to the subject of his or her expertise or a report containing opinions or advice of such an expert and not being a report used  or commissioned for the purposes of a decision of a public body made pursuant to any enactment or scheme".
Section 20(2)(e)
The records in this case comprise the drafts of a risk assessment undertaken by scientists in which they reviewed and drew conclusions from data and research available in relation to the effects of fluoridation. Before dealing further with the application of section 20(1), I should first say it is my finding that these records comprise "a report, study or analysis of a scientific or technical expert" and contain "opinions and advice" of such experts within the meaning of section 20(2)(e). Therefore, these records do not fall to be exempted under section 20 unless they are held to be "a report used or commissioned for the purposes of a decision of a public body made under any enactment or scheme." 
It is important to recognise that the two draft reports at issue here amount to records (as defined in the FOI Act) in their own right. In the normal course, it is fair to say that a draft report will always constitute a record which is separate from the record which comprises the eventual, completed report. In this present case, the separate existence of the drafts is emphasised even further by virtue of the significant difference in the conclusions reached as between the drafts and the eventual completed, report. The purpose for which these drafts was used was that of producing a final Risk Assessment Report. However, it may also be reasonable to take account of the purpose for which the Risk Assessment Report itself was being produced. This purpose, as I understand it, was to enable the FSAI advise the Forum on a particular aspect of food safety, i.e. the use of fluoridated water in infant formula. (According to the FSAI, it is "not responsible for the legislation on drinking water" but it "does have statutory responsibility for the safety of water as a food ingredient".)  I cannot see any case that the drafts were "used or commissioned for the purposes of a decision of a public body made under any enactment or scheme." 
I consider that the type of "decision" envisaged by section 20(2)(e) is one which a public body is authorised to make on foot of a specific provision where, for example, an application is made for a grant or a licence and scientific or technical advice is sought from experts to assist in the decision making process. I am not satisfied that the "decision" of the FSAI, to provide the Forum with scientific advice on a particular issue of concern to the Forum, falls into the category of "decision" contained at section 20(2)(e) of the FOI Act. While the FSAI may have a a general mandate to ensure the safety of food produced and sold in Ireland, any decision to issue advice (which, in essence, is what is involved in the provision of the Risk Assessment Report to the Forum) can hardly be seen as the type of "decision" envisaged by section 20(2)(e) of the FOI Act. 
The Forum, for its part, is not a public body as defined in the FOI Act; and, in any event, the Forum is not involved in the making of decisions "pursuant to any enactment or scheme". Even were it to be argued that the Risk Assessment Report is intended ultimately for the benefit of the Department of Health and Children, and this has not been argued, there is no case that it is intended to be "used" for the purposes of any decision of the Department, being a decision of the type envisaged in section 20(2)(e). In any event, the Department has already published the completed FSAI Risk Assessment Report so it would be difficult, if not impossible, to sustain the argument that the drafts of the Report should be withheld by reference to section 20(2)(e) where the Report itself has been published.
In short, I am satisfied that section 20(2)(e) does apply and I find accordingly. Whereas it is not strictly necessary to do so, I will now deal with the other arguments raised in relation to the applicability of section 20.
Section 20(1)
I accept  that the draft reports contain matter relating to the deliberative processes of the FSAI and, to that extent, are records which fall to be considered under section 20 of the Act. However, the section 20 exemption will only apply where the release of the draft reports would be contrary to the public interest. 
As I have indicated in other decisions, I do not accept that section 20 requires, as a matter of principle, that material should be withheld until the conclusion of a deliberative process. However, it is clear the Act envisages the possibility that there will be cases in which release of details of a public body's deliberations, even after a decision based on those deliberations has been made, would be against the public interest. The present position of the FSAI, notwithstanding that its own deliberative process has been completed and its recommendations made, is that section 20(1) applies to the draft reports; its position, in effect, is that release of these records would be contrary to the public interest.
The FSAI has made a number of arguments in relation to specific harms which, it contends, would flow from the release of the draft reports. It contends that these specific harms, were they to occur, would be contrary to the public interest. The harms identified by the FSAI can be summarised as follows:
release could have pre-empted and prejudiced the publication of the final report of the Forum on Fluoridation by the Minister;
release could give rise to unnecessary confusion, concern or distress to members of the public;
allowing into the public domain the early deliberations of the scientists who assisted in the risk assessment may be misleading and may damage the reputations of individual scientists and jeopardise their scientific publications;
having draft reports in the public domain could negate the impact of health promotion campaigns because such reports, containing incomplete advice, are likely to get equal media coverage with the final and fully considered report. 
Taking these arguments in turn, it is self evident that, since both the FSAI's Risk Assessment Report and the overall report of the Forum have now been finalised and published, release of the two draft reports cannot now pre-empt or prejudice such publication. Therefore, it is not necessary that I consider the first argument further. However, were it necessary to consider this further, I would expect to be provided with compelling, supporting evidence. On the face of it, the evidence adduced seems far from compelling. 
In discussing any effect that release of the records might have, I must have regard to the requirement of section 43(3)  that I take all reasonable precautions not to disclose information contained in an exempt record. The fact is, however, that the media has already disclosed to the public, and you as the requester are aware from the release of other records under the FOI Act, that differences arose as to the recommendation the FSAI should make in relation to the level of fluoride which might safely obtain in the case of tap water used to re-constitute infant formula. Clearly, there was a debate within the Scientific Committee as to whether the precautionary principle should apply and whether infant milk formula should be re-constituted with fluoridated tap water. I believe it is already in the public domain that expert scientific opinion differs as to the desirability of water fluoridation; this is acknowledged in the published report of the Forum. The FSAI is arguing that allowing access to draft or preliminary reports which may contain findings that differ from those of the final, published report would be likely to confuse and upset the public and parents in particular.
The various harms identified overlap to some extent in that they envisage a negative affect on the public if the draft reports are released. I have already indicated in other decisions that I do not accept that the fact that the contents of records might mislead means that release is contrary to the public interest. My views on the potential harms of  "confusion" and "upset" are very similar. The argument that releasing certain information may confuse or mislead has to be based on either of two assumptions: (1) that the public is not capable of properly understanding the information (and may be influenced by partisan or inaccurate accounts particularly from vested interests), or (2) that the information itself is inaccurate or poorly presented. What is at issue here is the degree of upset (if any) that members of the public might suffer if certain expert advice were disclosed to them. Although I find it difficult to imagine what the nature of such material might be outside of the area of personal information, I accept that there is a possibility that some information may exist in records of the deliberations of public bodies the release of which would be likely to cause such distress that it would be contrary to the public interest. I do not see, however, that release of the draft reports in this case would have such grave and harmful consequences. 
I have already made it clear in other decisions that I am not impressed by arguments based on the need to protect the public from possible "misuse" of the information; nor is there any provisions in the Act to exempt the release of information on the grounds that it is factually inaccurate. In this case the FSAI's concern seems to be, not that the content of the draft reports is incorrect, but that the recommendations were revised as a result of further discussion and review of the data before the final report was adopted. I believe that the FSAI has not taken sufficient account of the ability of the public to assess conflicting information and to put it in context. My view is that the public would be quite familiar, in the area of public health and elsewhere, with the concept of experts differing and conclusions being re-evaluated in the light of additional research. One would imagine that this would be relatively easy to explain to anyone who might query the information, particularly since the final Risk Assessment Report as well as the report of the Forum are now available and can be read in conjunction with the drafts. It is my understanding that the issues in question have, during the life of the Forum, been the subject of debate not alone in media reports and on the internet but also in Seanad Éireann and in Dáil Éireann. Disclosure of the draft reports might give rise to further public debate or criticism, which process could well be a positive aspect having regard to the public interest in facilitating more widespread knowledge of, and participation in, matters affecting public policy.
The FSAI further submits that allowing the early deliberations of its scientific advisors into the public domain, where those advisors had not had the benefit of input from their peers, might damage the reputations of individual scientists and jeopardise their scientific publications. I think the point being made here is that scientific research papers would generally be published only through peer reviewed scientific journals and that release of these drafts under FOI would damage the public interest. 
I accept the submission of the FSAI that the advisory structure to which it has access is an important national resource; and there can be no doubt that it is in the public interest that decisions involving food safety are based on the best scientific advice. While the submissions of the FSAI have not actually stated that release of the drafts would prejudice the willingness of participants in its committees to contribute in an open and frank manner, I consider that such a fear may be implied from its concerns about the effects on scientists of release of the drafts. In my decision in Case Number 98127 (Ms Eithne Fitzgerald and the Department of the Taoiseach), I dealt in some detail with "the candour and frankness argument" in the context of section 20. I found that, while an open and frank discussion is often required in order to evaluate and assess options in the deliberative process, a specific harm to the public interest flowing from the inhibition of such frankness and candour must be identified if the records at issue are to be exempted under section 20. I do not accept that scientists would express views that were contrary to their research findings, or fail to carry out their role in contributing expert advice to the debate in the FSAI's committees, merely because the records of their deliberations are potentially releasable under the FOI Act.
It seems to me, given the availability of the completed Risk Assessment Report, the general understanding which exists in relation to the status of draft reports and the acknowledged controversial nature of the subject matter, the scientists' colleagues would recognise that the draft conclusions and recommendations had undergone rigorous reassessment and amendment. The fact that the drafts represented no more than work in progress is, I believe, widely accepted and understood. I am aware that there is information already in the public domain to the effect that the draft of the risk assessment, in October 2001, contained recommendations which differed from those in the final report. I take the view that participation by external scientific experts in the work of the FSAI would be more likely to enhance their reputations than to damage their reputations. I find that the FSAI has not provided evidence that the reputations of individual scientists, or the standing of any of their scientific publications, would be jeopardised by the release of the draft reports.
The last of the public interest arguments put forward by the FSAI is that release into the public domain of draft positions or recommendations, in the context of its involvement in a health promotion campaign, would undermine the impact of that campaign. The FSAI contends that it acts in an open and transparent manner as an advocate for Irish consumers in the area of food safety and provides information to the public on request and by way of publishing on its website the agenda and minutes of its scientific committees and subcommittees as well as its final reports. It takes the view that a draft report would be likely to get as much media coverage as the final report. This may well be the case but it is not a factor which can influence my decision since, as I have found in previous decisions, the use to which the record might be put after its release is not normally a relevant consideration in deciding whether an exemption applies under the FOI Act. In any event what is at issue here is not whether, in general terms, release of positions considered at the draft stage of a report would undermine the impact of the subsequent finalised position; what is at issue is whether, in relation to the specific Risk Assessment Report of the FSAI, release of the drafts sought by you would undermine any health promotion campaign in which the FSAI might be involved. I am not aware of any proposal to mount a health promotion campaign in relation to the use of fluoride and it seems unlikely that any such campaign will take place. Given that water supplies are fluoridated at source, such a campaign would not be necessary. As set out above, I do not think that the public would find it distressing, confusing or even unusual to find that there are views within the scientific community on the use of fluoridated water which differ from those in the final report of the Forum or the final version of the FSAI's Risk Assessment Report. In all the circumstances, I cannot see that release of the draft reports sought by you would damage any public interest in maximising the impact of a health promotion campaign. 
 As a general principle, I consider that it is in the public interest that views and representations which influence the deliberative process should be open to public scrutiny. As I have said before, the very enactment of the FOI Act suggests that significant weight should be attached to the public interest in an open and transparent process of government. 
 I am not satisfied that the FSAI has presented convincing arguments that the release of the draft reports would be contrary to the public interest under section 20 of the Act and I find that the two records at issue are not exempt under that section.
Section 21
The FSAI has also relied on section 21(1)(a) in refusing you access to these records. Section 21(1)(a) of the FOI Act allows a head, subject to consideration of the public interest, to refuse to grant a request for information made under the Act if access to the record concerned could, in the opinion of the head, reasonably be expected to -
"(a)  prejudice the effectiveness of tests, examinations, investigations, inquiries or audits conducted by or on behalf of the public body concerned or the procedures or methods employed for the conduct thereof,".
As a preliminary point, I should say that I accept that a risk assessment of the type in the draft reports can properly be described as an investigation or inquiry and the procedures and methods employed would include the practice of the FSAI having external, expert scientists contribute to the overall process.
In the case of the Sunday Times Newspaper & Others and the Department of Education and Science (case number 98104) I explained the approach which I adopt to the application of this exemption. In summary, the exemption is concerned with whether or not the decision maker's expectation is reasonable. It seems to me that in arriving at a decision to claim a section 21 exemption, a decision maker must firstly identify the potential harm to the functions covered by the exemption which might arise from disclosure and, having identified that harm, consider the reasonableness of any expectation that the harm will occur.
Section 21(1)(a) envisages two potential types of "prejudice" which must be considered by a decision maker in terms of his or her expectations.  The decision maker must hold the view that the release of the records will prejudice the "effectiveness" of the investigation or inquiry or that release will prejudice the "procedures or methods employed for the conduct thereof".  
In this case the FSAI originally argued that release of the records would be likely to prejudice the final version of the Risk Assessment Report. Since both the FSAI's and the Forum's reports have now been published, it is not necessary for me to consider in my review whether the publication of the final Report would be prejudiced. I will, however, consider whether there is a reasonable expectation that the effectiveness of the FSAI's investigation, or the procedures and methods which it employed, would be prejudiced. 
In arguing that it is important that it retains the co-operation and expertise of the best available scientists, the FSAI appears to be making the case that release of the draft reports might, in some way, deter external scientists from continuing their voluntary involvement with its Scientific Committee and subcommittees. This, the FSAI believes, would lead to difficulties for the committees in making judgements on the basis of the best scientific information, including new information emerging on an ongoing basis from sources including EU scientific committees. This is the potential harm to the functions covered by the exemption which has been identified in the FSAI's submissions. For section 21(1)(a) to apply, the decision maker must also form the opinion that disclosure could reasonably be expected to lead to the harm identified.  Therefore, the question which I have to address is whether the expectations of prejudice in this case are reasonable. 
As I have explained above when dealing with similar arguments advanced by the FSAI in relation to section 20, I am not satisfied that participation by external scientists in the work of the FSAI would be likely to damage their reputations even where it is clear that draft recommendations have been reviewed and amended in the period between preparation of draft reports and release of the final report. I think that the experienced experts involved would be more than capable of explaining and defending their positions in the circumstances of the records under review and would also be able to do so if a similar situation were to arise in the future. I would be surprised if scientists were not accustomed to controversy, especially in the case of the debate on fluoridation. I consider that participation in assessments for the FSAI allows experts to have a platform on which to discuss their research and that of their peers, to air their views in an important forum and to have an influence on the making of public policy in the important field of food safety. When advising the FSAI, I have to assume that the scientists are aware that they are contributing to the working of a public body and that it is not reasonable to expect that views, which support positions other than those expressed in the final report, can be withheld from public scrutiny even after the completion and publication of that report.
Having regard to the requirements of section 34(12) and the submissions of the FSAI, I find that it has not made out a sufficiently strong case and is not entitled to refuse access to the records concerned under section 21(1)(a) of the FOI Act.
Section 21(2) permits the granting of access to a record which otherwise would be exempt by virtue of section 21(1)(a), where the public interest would, on balance, be better served by doing so. In the circumstances, I wish to state that even if I were to accept that the requirements of section 21(1)(a) were met (which I do not), I would find, in this case, that the public interest would, on balance, be better served by granting than by refusing the request. I acknowledge, of course, that there is a public interest in ensuring that the FSAI is able to conduct its business in an effective manner. However I consider that, having regard to the fact that the completed Risk Assessment Report has been published and that the fluoridation controversy is already in the public domain, the public interest in the public knowing how the risk assessment process was conducted in this case outweighs the other public interest considerations.
Section 30
In its decision, the FSAI also sought to rely on Section 30 in refusing access to the draft reports. This section provides that :
"(1) A head may refuse to grant a request under Section 7 if, in the opinion of the head -
(a) the record concerned contains information in relation to research being or to be carried out by or on behalf of a  public body and disclosure of the information or its  disclosure before the completion of the research would be likely to expose the body, any person who is or will be  carrying out the research on behalf of the body or the subject matter of the research to serious disadvantage, ...........
(2) Subsection (1) does not apply in relation to a case in which, in the opinion of the head concerned, the public interest would, on balance, be better served by granting than by refusing to grant the request under section 7 concerned."
This is an important exemption designed to protect the legitimate interests of public bodies and researchers. I accept that there will be cases where the release of incomplete research may, legitimately, be denied by a public body. However, I am  not at all sure that section 30 is even potentially relevant in the present case. The process undertaken by the FSAI was one of risk assessment taking account of the findings of research already conducted and of the expertise and knowledge of external scientists. This was not a situation in which the FSAI conducted or commissioned its own research for the purposes of reaching a conclusion; rather it relied on findings from existing research and on the opinions of recognised experts. 
In any event, even if one were to accept that the risk assessment process constituted "research" for the purposes of section 30, that process has been completed and a final report produced. Given that the completed Risk Assessment Report has now been published, it is hardly conceivable that release of drafts of the Report would in any sense prejudice the position of the FSAI. Furthermore, the exemption at section 30(1) is subject to a public interest test and, for all the reasons already set out above, I would in any event conclude that, on balance, the public interest is better served by release of these records than by their being refused.
Accordingly, I find that the FSAI has failed to sustain a claim for exemption under Section 30.
Having considered carefully each of the exemptions claimed by the FSAI, I am satisfied that none of them applies to the two draft reports now at issue in this review. I am conscious that I am giving this decision in a context in which the completed Risk Assessment Report has been published (on 10 September 2002) as part of the Forum Report. There may be a perception that the publication of the completed Risk Assessment Report is a defining consideration in my overall decision. While the publication of the Report is certainly a consideration, it may be helpful to clarify that in my view the completion by the FSAI of the risk assessment process is a much more significant consideration. Whereas it may have been legitimate to refuse access to the drafts of the Report for as long as the Report itself was being compiled, I cannot see any legitimate basis for refusing access to the drafts under the FOI Act once the Report had been completed. I note that at the time of the internal review decision, on 24 April 2002, the Risk Assessment Report had already been completed.
Having carried out a review under section 34(2) of the Freedom of Information Act, 1997  I hereby annul the  decision of the FSAI in relation to the two records identified in the "Scope of the Review" section above and direct that they be released to you.
A party to a review, or any other person affected by a decision of the Information Commissioner following a review, may appeal to the High Court on a point of law arising from the decision. Such an appeal must be initiated not later than four weeks from the date of this letter.       
Yours sincerely
Kevin Murphy
Office of the Information Commissioner