Social Audit Ltd
P O Box 111 London NW1 8XG
Telephone/Fax 44 (0)171 586 7771


M.Fernand Sauer, Director,
European Medicines Evaluation Agency
One Canada Square
Canary Wharf

2 December 1997


Dear M. Sauer,

I am enclosing a copy of my paper, The Antidepressant Web, to be published this week in the International Journal of Risk and Safety in Medicine. I believe many of the issues discussed in this paper have important health and regulatory implications.

The immediate cause for concern is the strength of the evidence that (in particular) SSRI antidepressants are as much drugs of dependence as benzodiazepines - though they fall outside current definitions of 'dependence' in ICD-10 and DSM-IV (as indeed the benzodiazepines arguably do too). However, the BDZs were classified as drugs of dependence essentially on the strength of evidence of normal dose dependence, especially after long-term use - in the days when DSM-III determined that "The diagnosis of all of the Substance Dependence categories requires only evidence of tolerance or withdrawal …". The same point is underlined in the 1990 report of the American Psychiatric Association's Task Force report on Benzodiazepine Dependency:

"The presence of a predictable abstinence syndrome following abrupt discontinuance of benzodiazepines is evidence of the development of physiological dependence…"

"Historically, long-term, high-dose, physiological dependence has been called addiction, a term that implies recreational use. In recent years, however, it has become apparent that physiological adaptation develops and discontinuance symptoms can appear after regular daily therapeutic dose administration ... in some cases after a few days or weeks of administration. Since therapeutic prescribing is clearly not recreational abuse, the term dependence is preferred to addiction, and the abstinence syndrome is called a discontinuance syndrome." (APA, 1990)

The enormous scope for confusion is underlined by data in the following table, based on 'Yellow Card' reports of suspected adverse reactions submitted to the UK Medicines Control Agency/Committee on Safety of Medicines:

Introduced Ballpark No. Yellow Card reports of:
of NHS scripts withdrawal reactions dependence- all forms
Temazepam 1977 50m 5 3
Diazepam 1963 180m 20 16
Fluoxetine 1989 9m 59 10
Paroxetine 1991 6m 802 9

On the face of it, it seems absurd to classify benzodiazepines as drugs of dependence but SSRIs not: either they both are or neither is. In the meantime, the public is receiving extremely confusing advice, being told by the authorities that BDZs are drugs of dependence but antidepressants not. Moreover, because the authorities believe this, I fear that sweepingly optimistic, but quite mistaken, assumptions have been made about the long-term effectiveness of antidepressants.

The picture today seems to be alarmingly reminiscent of the position around 15 years ago, when the conventional wisdom was that barbiturates were drugs of dependence but benzodiazepines not. Indeed, similar patterns of behaviour by the experts go back to the last century, as described in the book Power & Dependence (1992). However, the present problem may be more serious, because of the very bullish statements made by some authorities about the lack of any dependence risk with SSRIs, together with advice about the need for long-term, if not indefinite treatment. The effect of this would of course be to compound the risk of dependence.

The MCA/CSM have shown no real signs of concern about these matters, though the evidence in my paper suggests to me they have comprehensively failed to assess the true risk, in concluding that withdrawal reactions from SSRIs may be so low as to be "undetectable except through spontaneous reporting where drug exposure is high". (Price et al., 1996). This prompts me to ask whether the EMEA shares the view of the MCA/CSM and, if not, what its estimate of the risks might be?

I would welcome your comments also on the evidence in this paper (notably in sections 2.7 to 2.9 and 3.7 to 3.8) that the 'scientific' assessments on which the authorities have relied are so flawed as to leave open the possibility that even basic assumptions made about safety and effectiveness (and therefore benefit and risk) may be quite wrong.

You may not be surprised by my views on the main underlying causes of this apparently serious problem, among many other examples of ill-health resulting from drug treatment in therapeutic settings. They include [a] lack of transparency and related failures of understanding and communication; [b] the lack of any significant active participation by consumer and patient representatives in regulatory matters; and [c] conflicts of interest within regulatory systems, including many individual expert advisors. By the nature of its operation, which involves a high degree of dependence on national regulatory functions, I believe these are serious shortcomings also in the EMEA/CPMP system.

I would welcome your comments on these points, together with any indication you might give of the role the EMEA/CPMP might play in helping to sort out this problem. Please note that I intend shortly to post on the Internet both the paper I am sending you, and correspondence on relevant matters arising. I hope this will encourage rather than deter any response from you because of the opportunity it will provide to all concerned to consider all points of view. Thank you for your attention and I shall look forward to hearing from you.

Yours sincerely,

Charles Medawar


Reference: J.S. Price, P.C.Waller, S.M. Wood, A.V.P. Mackay, A comparison of the post-marketing safety of four selective serotonin reuptake inhibitors including the investigation of symptoms occurring on withdrawal, Br. J. Clin. Pharmacol., 1996, 42, 757-763.


Contents page
Correspondence with government