(Parliamentary Commissioner for Administration)

1   Re: Participation of members with possible conflicts of interest

in meetings of the Committee on Safety of Medicines

2   Re: Non-disclosure of information relating to possible conflicts of  interests of

members of the Committee on Safety of Medicines

3    Re: Non-disclosure of expurgated minutes of CSM meetings 
and dates of CSM meetings in 1998


1. Outline of Complaint

This is a complaint of maladministration by the Medicines Control Agency (MCA). It relates to formal requests to the MCA, made under the Code of Practice on Access to Government Information, to which answers could and should have been supplied within 20 working days. After 100 working days, no answers had been provided, nor credible explanations given for the delay. Complainant (CM) believes no proper justification exists for withholding the requested information sine die, and claims his rights under the Code have been abused. He seeks disclosure of the information and investigation of the reasons and justification for the delay.

By implication, criticism extends to the Committee on Safety of Medicines, an advisory body, intimately linked to the MCA. However, the CSM is believed to be beyond the PCA's jurisdiction. Therefore the CSM is not the subject of this complaint but, because it is centrally involved, its role needs to be explained.


2. Background to complaint

[a] Eight months ago, a report by the complainant was published in the International Journal of Risk & Safety in Medicine; this was a long, critical review relating to the safety and effectiveness of certain antidepressant drugs. The paper noted that, at the time such drugs were being licensed, the CSM's leading expert in this field had personal and other interests with most of the manufacturers of these drugs.

[b] The complainant's paper also criticised the methodology, design and interpretation of a scientific study on one antidepressant drug (paroxetine), conducted jointly by that CSM member and an employee of the manufacturers of this drug. Their study was subsequently published in a journal edited by the CSM member. Complainant alleged in his report (and in subsequent correspondence with the MCA/CSM) that the authorities were mistaken in accepting as valid the main conclusions of this research.

[c] If the authorities are mistaken, as alleged, it would have had far-reaching consequences relating to the safety, effectiveness and monetary value of this and other antidepressant drugs. In addition, questions would arise about the competence of the authorities in relation to assessment of benefits and risks of other medicinal drugs.

[d] On 3 March 1998, Professor Michael Rawlins, Chairman of the CSM, was engaged (with complainant) in a debate organised by the Kings Fund Centre. On this occasion, he emphasised:

[i] that there had existed "over many years strict arrangements for members to disclose their interests with individual companies";

[ii] that "members are required to disclose any interest with the company concerned at the start of each agenda item at CSM meetings. Members with personal or specific interests take no part in the discussion and leave the room"; and

[iii] "The data collected by the Committee itself is publicly available and, contrary to what Charles Medawar may tell you, has been so for many years".

[e] On 12 March, complainant wrote to Prof Rawlins seeking to verify these remarks, and to establish relevance of such procedures in the case cited in [a] and [b] above. Clarification was sought also because the procedures described appeared to conflict with CSM standing orders. Its Code on Declaration of Interests states: "The member shall take no part in the proceedings as they relate to the product except, at the Chairman's discretion, to answer questions from other members" (section 14), and "the Chairman has the power to determine whether a member with an interest shall take part in the proceedings" (s 18). It also requires that a record be kept indicating "whether the member took part in the proceedings" (s. 20b).

[f] Complainant's letter was acknowledged (26 March) by the MCA. They confirmed that the questions were "being handled under the Open Government Code of Practice" and had been passed to the CSM. On 22 April, the MCA again wrote, to apologise for CSM Secretary's failure to acknowledge receipt, and "your requests are being considered and replies will be sent as soon as possible". However, on 8 May, Prof Rawlins wrote to complainant, refusing to respond: "You ask for certain information relating to the CSM be provided to you under the Open Government Code. This was discussed at the last meeting of the Committee. Members decided that I should not provide you with answers you pose. If you wish to take the matter further then I suggest you should write to the MCA." No explanation was offered for the refusal or delay, as the Code requires; later Prof Rawlins wrote (3 July) to say that "The Committee's arrangements for disclosure are determined by s118 of the Medicines Act and are not covered by the Code of Practice for Open Government."

[g] Several expressions of regret later, the MCA indicated (14 July) that the reason it had not answered complainant's questions was that "important questions of operational practice" (IQOPs) were involved. On request, the MCA explained (23 July) what was meant by this, but without specifying which of the listed excuses applied to any particular question. Later (29 July), MCA said that they regretted the delay and emphasised that this reflected the care they were taking in addressing complicated issues.

[h] Complainant maintains that full disclosure of information would be appropriate in view of the extent of possible conflicts of interest that might arise. Analysis of interests declared by CSM members in the period 1992-1997 indicates that:

  • Over half of all CSM members have personal interests with regulated companies
  • Three out of four CSM members have personal or non-personal interests with regulated companies
  • About half have personal or non-personal interests with companies that market SSRI/SRNI antidepressants

See tables.

Re: CSM members' declarations of interest, 1992-1997

Number and percentage of CSM members who disclosed: 1997 1996 1995 1994 1993 1992
personal or non-personal interest in any company 23/31 74% 24/30 80% 19/21 90% 18/21 86% 19/22 86% 16/21    76%
personal interest in any company 16/31 52% 19/30 63% 12/21 57% 11/21 52% 15/22 68% 13/21    62%
personal or non-personal interest in any company marketing SSRIs etc 14/31 45% 17/30 57% 12/21 57% 9/21 43% 10/22 45% 8/21   38%
personal interest in any company marketing SSRIs etc 6/31 19% 9/30 30% 2/21 10% 2/21 10% 2/22 9% 3/21    14%


Re: CSM members' declarations of interest, 1997 (1996)

Number and percentage of CSM members who disclosed, in 1997: Personal interests with any company marketing SSRIs (listed on left) Personal/non-personal interests with company marketing SSRIs Either personal or non-personal interests with at least one company
SmithKline Beecham 2/31        6% 7/31      23%   7/31      23%
Eli Lilly 4/31      13% 4/31      13%   6/31      19%
Pfizer   9/31      29%   9/31      29%
Bristol Myers Squibb        5/31      16%   5/31      16%
Wyeth 1/31        3% 1/31        3%   2/31       6%
Lundbeck 1/31        3%   1/31       3%
Solvay   0/31       


3. Basis of complaint

One question, central to this complaint, is whether and to what extent the Code of Practice on Access to Government Information would apply. Complainant claims that, if the CSM is not bound by the Code, then neither can it claim any rights, privileges or protections under the Code - and nor can the MCA credibly do so on its behalf. How can the MCA handle such a request on behalf of a different organisation, not bound by the Code, which has already emphatically refused to answer?  Complainant believes that independent adjudication is needed in this case, because the MCA itself faces difficult conflicts of interest. These are believed to have led to the present impasse.

On the one hand, the MCA needs to maintain its normally very close relationship with the CSM, and to avoid the friction that would result if it made recommendations for disclosure that threatened the autonomy or authority of the CSM. On the other hand, the MCA's would have been embarrassed by the initial confusion over the CSM's responsibilities under the Code, later compounded by the CSM's brusque dismissal of Social Audit's requests for information. In doing so, the CSM demonstrated that it had considered the relevance of the Code in terms of self-interest, whereas the MCA recognised that the Code also gives rights to requesters.

The dilemma for the MCA seems intractable. It is in no position to order the CSM to disclose, yet non-disclosure of simple information about CSM's members' possible conflicts of interest would seem indefensible. In the circumstances, it seems doubtful whether the MCA is capable of fairly and/or effectively handling such a Code request.

On this basis, complainant argues that the Code would be relevant only in relation to requester's rights to information, but not in providing justifications for non-disclosure. There appears to be implicit acceptance of this argument in the MCA's explanation for non-disclosure. The MCA's objections have been expressed in terms of IQOPs, without reference to provisions of the Code. To this extent, the complaint would seem either to stand or fall on a straightforward test of public interest.


4. Re: first request for disclosure

This complaint relates to the MCA's refusal to respond to the following question, addressed to Prof Rawlins on 12 March:

On how many occasions in meetings of the CSM held during both 1991 and 1997, was there recorded in the minutes or elsewhere: [a] any formal vote on any matter directly connected with the licensing of any particular drug; and [b] any instance when a member absented him/herself from the room to avoid some possible conflict?

This question asks only for simple, aggregated numerical information. A satisfactory answer might, for instance, have been: "[a] On 3/88 occasions in 1991, and 6/109 occasions in 1997; and [b] On one occasion in 1991, none in 1997."

Complainant contends that no important questions of operational practice arise sufficient to justify protracted non-disclosure in response to a Code request. The refusal to disclose seems groundless to the point of being ridiculous. Complainant will not labour the point but would rely, if necessary, essentially on the arguments below.


5. Re: second request for disclosure

This complaint relates to the MCA's refusal to respond to a second question raised in the letter to Prof Rawlins of 12 March:

Please supply the following from the record (referred to in section 20 of the Code of Declaration of Interests) of names of CSM members "who have declared interests at meetings … giving dates, names of relevant products and companies, details of the interest declared and whether the member took part in the proceedings":

[i] copies of the full entries made, from 1987 to date, relating to any personal specific interests disclosed during CSM meetings, when the product under consideration was fluvoxamine, fluoxetine, sertraline, paroxetine, venlafaxine, citalopram or nefazodone.

[ii] copies of the full entries made, relating to any personal specific interests disclosed during the last five CSM meetings, for any "relevant product".

On the following grounds, complainant doubts the relevance of the "important questions of organisational practice" said to arise, and again maintains they could not justify a sine die refusal to disclose:

[i] " … current policy on disclosure of information of confidential advice between Ministers, the MCA and the CSM and whether disclosure would harm the frankness and candour of advice and discussion. The guidance in the Code envisages limited circumstances in which a refusal to disclose expert advice may be justified …". Comment: This request does not concern "advice", nor does it involve ministers or the MCA, and there is no reason otherwise to suppose it would inhibit effective communication.

[ii] " … The minutes of CSM and the details they contain are marked "not for publication" …Comment: The request relates to information not from CSM minutes, but from "a record … kept in the Department". The label, "not for publication", is in itself irrelevant: the question is whether and to what extent the restriction is justified. The restriction would clearly not apply to all details (eg date of meeting, or the election of a new chair), nor would it necessarily override the imperative to respond to a reasonable request for information.

[iii] "… explicit undertakings of strict confidentiality have been required and given at CSM meetings and members have proceeded on this understanding …" Comment: In relation to disclosure of information by CSM members about possible conflicts of interest, this proposed exemption appears irrelevant if not spurious. It contradicts what the CSM Chairman said about the Committee's policies on disclosure, and is also at variance with guidance in the Code on members' Declaration (sic) of Interests: "Ministers have decided that the arrangements which govern relationships between members and the pharmaceutical industry and information on significant and relevant interests should be on the public record". Members' interests with companies (if not individual products) have been published for some years.

[iv] "… It is against this background that we are currently considering your requests. We also note that information supplied in confidence by advisory bodies is one of the specified interests set out in the White Paper at para 3.11. In particular, it refers on page 18 to the need to protect information "given freely on the understanding of confidentiality" and on page 19 to communications "covered by explicit undertakings of confidentiality, or at a least a reasonable expectation that the law of confidentiality applied to them". Comment: The MCA does not say it relies on statements the White Paper, only that it "notes" them. It seems inconceivable that CSM members "declared" conflicts of interest with any expectation of confidentiality but, if such undertakings were either given or implied, the MCA should produce evidence to this effect.

In short, there appear to be no substantial reasons for refusing to disclose. Complainant also asserts that disclosure of the information would be in the public interest on the following grounds:

[i] Reasonable doubts and concerns exist about the adequacy of present arrangements relating to possible conflicts of interest in the CSM, as outlined in 2[a] and 2[b] above;

[ii] Reasonable doubts also exist about the accuracy of statements made by the CSM Chairman (3 March 1998), as cited in 2[d] above.

[iii] The unjustified refusal to disclose possible conflicts of interest would amount to conduct unbecoming in any organisation with important public responsibilities - let alone one which is monopolised by doctors, secretive to an extent considered excessive even by its Chairman, close to the pharmaceutical industry but remote from consumer interests, where members walk a tightrope of temptation and then make decisions on matters of life and death.


6. Re: third request for disclosure

This complaint relates to two Code requests originally sent to the Secretary of the CSM on 20 March. It seems convenient to consider these with the above complaints though, in this case, the above paragraphs 2[a] to 2[e] do not apply.

The first request was for facsimile copies of minutes of CSM meetings held in 1998 from which all non-disclosable and confidential information had been deleted. Complainant asserts that, given the terms of the request, there can be no justification under the Code for refusing disclosure.

In refusing this request, the MCA is effectively arguing that the public interest is best served when requesters may not know "how much of the Committee's business is typically devoted to secret discussion of licensing matters", and how much attention is "devoted in meetings to questions of policy, finance, image, procedure and other matters not specifically related to licensing of individual drug products". Complainant believes this to be quite unacceptable and asserts that, in refusing disclosure, the MCA/CSM has been excessively guided by considerations of self-interest.

The second request made on 20 March was for a list of dates of forthcoming CSM meetings. This list was eventually supplied by the MCA, with regrets, four months after request was made (23 July).

Complaint alleges delay was unacceptable and unjustifiable, and that an explanation should have been given for it. Complainant speculates that the reason for ultimately disclosing was that MCA recognised the futility of seeking to justify non-disclosure of a list of dates on grounds that 'important questions of operational practice' were involved. Complainant has wondered if it was just coincidence that MCA eventually supplied the list of dates on the same day as the last CSM meeting before their summer break.


7. Re: adjudication

Complainant requests prompt adjudication. Delay would be unwelcome in this case, both because the agency has already delayed for several months, apparently for no good reason; also because this might in other cases reduce the Agency's incentive to disclose. The requester/complainant has been professionally frustrated. He is a specialist journalist. His work is published mainly on the Social Audit website, and referred to and relied on by others: this website now receives well over 3,000 visits per month.


Charles Medawar
10 August 1998
Contents page
What's New?