Social Audit Ltd
P O Box 111 London NW1 8XG
Telephone/Fax 44 (0)171 586 7771
Dr. Robert Kendell CBE, MD, FRCP, PRCPsych
President, Royal College of Psychiatrists
17 Belgrave Square
London SW1X 8PG

30 October 1997

Dear Dr Kendell

I am writing to outline a number of concerns about the conduct of the College's "Defeat Depression Campaign" over the past five years, and to suggest the need for a fundamental reappraisal of some of its key messages and recommendations. I would appreciate your comments, also your response to some specific requests for information. Briefly, I am concerned on three main fronts:

1. Throughout the Defeat Depression Campaign, the RCP has emphasised that, unlike benzodiazepine and other tranquillisers, antidepressants do not cause dependence. The evidence for this claim was always weak and I think it now clear that the risk of dependence with SSRI antidepressants, in particular, is at least as great as with benzodiazepines.

2. The RCP has endorsed the 1992 consensus statement treatment guidelines which emphasise the need for long-term drug use, though these were not based on reliable evidence of long-term drug effectiveness. Taking into account also the College's advocacy for definitions of "depression" which have greatly expanded the numbers of 'eligible' patients, any risk of dependence has been increased; and

3. The independence of the College might be called into question, since it has done much to promote the use of, in particular, SSRI antidepressants, while receiving major financial contributions from the manufacturers of all such products. Some of these companies have also supported the work of leading figures in the Defeat Depression Campaign.

I hope you will accept that I appreciate the importance of the general objectives of the Campaign. I know depression can be severely disabling - all the more so when sufferers are stigmatised - and I do appreciate the value of effective treatment. The reservations I have about the Defeat Depression Campaign arise in spite and because of this.

1. Risk of dependence

From the outset, in 1992, the Campaign has emphasised that there is no risk of dependence with SSRI and other antidepressants. The Campaign’s first press release was headlined, "Antidepressants not addictive..." and went on to express concern "that people may fail to take the medicine in the mistaken belief that it can cause dependence". In revised treatment guidelines, published soon after, doctors were advised to emphasise to patients "the facts that antidepressants are not habit-forming or addictive…" (Paykel & Priest, 1992). The same message emerged on conclusion of the Campaign: "In particular patients should know that dependence is not a problem with antidepressants". (Priest et al., 1996).

In such statements, the Campaign has always sought to distinguish between tranquillisers (where dependence problems are acknowledged) and antidepressants (where not). However, because two fundamentally different definitions of "dependence" have been used - one for tranquillisers and another for antidepressants - I believe the Campaign has consistently given confusing and misleading advice.

Benzodiazepines were classified as drugs of dependence, in the early to mid 1980s, essentially because of their tendency to cause withdrawal reactions, even when taken at the usual recommended dosage, though especially after long-term use. This 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)

Essentially the same point is made in the College's report, Benzodiazepines: risks, benefits or dependence (1997): "Dependence on benzodiazepines is mainly manifest by withdrawal symptoms on cessation", and "Dependence is now recognised as a significant risk in patients receiving treatment for longer than one month…"

Using this traditional and familiar definition - in which evidence of withdrawal problems alone provides sufficient evidence of dependence - SSRIs would certainly have to be classified as drugs of dependence. The College's insistence that they are not is presumably based on its use of the fundamentally different definitions of "dependence" in ICD-10 and DSM-IV. I imagine this would have caused much confusion even if the double standard had been explained. But I have seen no evidence that the College has tried to explain that it has two quite different notions of "dependence" in mind. Have I missed something? If not, is this simply an oversight, or was it knowingly done in the belief that it would help to convince people of the benefits of treatment for depression?

The risk is such that, under the ICD-10 and DSM-IV definitions, neither benzodiazepines nor barbiturate sedatives would ever have been classified as drugs of dependence, on the basis of experience in routine therapeutic use. The extent of the confusion can be seen by comparing the numbers of Yellow Card reports of suspected adverse drug reactions sent to the Committee on Safety of Medicines/ Medicines Control Agency for both benzodiazepines (BDZs) and SSRIs. These can only be rough and ready comparisons but, even allowing for confounding factors, the present situation seems untenable. The table below compares withdrawal reactions reported (to March 1997) for the most prescribed BDZs, temazepam and diazepam, with those reported for paroxetine:

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
Paroxetine 1991 6m 802 9

It seems to me incomprehensible to proceed on the basis that diazepam and temazepam drugs are of dependence, but paroxetine not. Either they both are or neither is. To suggest that one is, the other not, surely cannot be justified?

I appreciate that reports of withdrawal reactions to paroxetine far outnumber those for all other SSRIs (all BDZs too). Nevertheless, the presumption must be that some risk of dependence attaches to other SSRIs if only because:

1. the numbers of reports of withdrawal reactions to fluoxetine and sertraline are at least double those for diazepam; and

2. the essence of the benzodiazepine problem is not that some people experience sometime severe symptoms on withdrawal, but that many more continue to take drugs in order to avoid them.

2. Evidence of long-term effectiveness

Against this background, also taking into account that it took 20+ years to understand the risks of long term use of BDZs, I think there must be serious doubts about the long-term effectiveness of SSRI antidepressants. I therefore question the Campaign's recommendations relating to the need for more aggressive treatment and extended use. I realise that the (consensus statement) guidelines have not been formally adopted by the College. Still, they embody Campaign objectives and were developed by leading figures in the Campaign, and the College has done much to promote them.

There are several grounds for concern. One relates to the adoption of very broad spectrum definitions of "depression". Over 300 disease entities are now listed in DSM-IV, many relating to symptoms of anxiety. In marketing speak, "anxiety" has, in effect, now been 'repositioned' as "depression"; many people who would in the recent past have been treated with BDZs are now being treated with antidepressants instead. At the start of the Defeat Depression Campaign, the College called for a 70% increase in treatment of depression, and prescriptions for antidepressants have indeed increased by almost as much since then.

The broad-spectrum definition of "depression" would also increase the risk that antidepressant withdrawal symptoms were mistaken for reappearance of the original condition. With the BDZs, it took over 20 years to appreciate that doctors and patients tended to mistake withdrawal symptoms as signs of relapse, and the evidence suggests they are still often likely to be confused. I believe there is a substantial risk the same is now happening with antidepressants. The risk would be all the greater because of the Campaign's advice to doctors and patients that dependence isn't a problem - with the clear implication that withdrawal reactions barely exist.

Most worrying are the consensus guidelines, which promote the Campaign's recommendations for higher-dose, longer-term antidepressant prescribing. The increased drug exposure increases any risk of dependence. In addition, these recommendations are based on trials of long-term effectiveness in which no account was taken of the possibility of withdrawal symptoms. This means it would not have occurred to the investigators that withdrawal reactions might be mistaken for relapse; therefore all symptoms of withdrawal would have added to the impression of long-term effectiveness. This is exactly what happened with the BDZs.

An example of the genre is the paper by Montgomery & Dunbar, "Paroxetine is better than placebo in relapse prevention and the prophylaxis of recurrent depression", (1992). I mention this particular paper because: [a] withdrawal symptoms with paroxetine are particularly conspicuous; [b] the principal author, Dr Montgomery, has published probably more papers than anyone which conclude with recommendations for long-term use; [c] At the time the manufacturers of the main SSRIs were seeking Product Licenses, Dr Montgomery was the leading expert in the field with the Committee on Safety of Medicines, and he declared 'personal interests' with four manufacturers of SSRIs now on the market; and [d] he is not only a leading figure in the College but also works alongside the Chairman of the Campaign at St Mary's Hospital, Paddington. I will not bother you with a detailed critical analysis of this paper, but I can assure you it deserves it. In the light of the evidence in the enclosed book, Power & Dependence, I find it lamentable.

Finally, I do not think that the best evidence supports the preference expressed for SSRIs in the treatment guidelines; their benefits seem exaggerated. Surely, the most striking things about antidepressants, new and old, is their lack of specificity. None of the 70-odd antidepressants used in the past 40 years has ever proved to have a greater effect on depression than any other. Antidepressants are usually found to work more often than placebos, though qualitatively there seems to be little difference in a positive response to either. It is as if all antidepressants work like strong placebos; the SSRIs certainly represent no great advance on previous models.

3. Commercial impact on the Campaign

In the College's 1996 Annual Report, all the major manufacturers of SSRI antidepressants are listed as major contributors to the Defeat Depression Campaign. The absence of the major producers of tricyclic and other non-SSRI antidepressants is notable, in the light of the Campaign's implicit endorsement of SSRIs and its emphasis on higher than ever success rates in treating depression.

I am worried by all this and shall soon be reporting publicly on the subject, with a view to establishing if other health professionals are concerned too. Could you therefore please let me know what the College's policy is on accepting financial support for a Campaign from sources which have a direct pecuniary interest in the outcomes? As a Registered Charity, I imagine the College will have developed a policy statement on this; I would welcome a copy, with an indication of when the policy was first formulated, and when last revised. I should be grateful if you would also let me know the total contributions made by each of the listed SSRI manufacturers who sponsored the College's campaign. Finally, please could you please arrange for me to be sent a copy of the Audit of the Defeat Depression Campaign carried out by the College Research Unit, for which funding of 65,000 was secured from the Department of Health.

I shall be grateful if you would consider these points. I realise that it could take a little time to investigate them fully, though I think the essence of it will be immediately clear. In the circumstances, I hope you will feel able to get back to me with your comments in the near future.

I am sending a copy of this letter to the President of the Royal College of General Practitioners, as co-sponsors of the Defeat Depression Campaign.

Yours sincerely,

Charles Medawar


S.A. Montgomery, G. Dunbar, Paroxetine is better than placebo in relapse prevention and the prophylaxis of recurrent depression, Int. J. Clin. Psychopharmacol., 1993, 8, 189-195.
E.S. Paykel, R.G. Priest, Recognition and management of depression in general practice: consensus statement, Br. Med. J., 14 November 1992, 305, 1198-1202.
R.G. Priest, C. Vize, A. Roberts, M. Roberts, A. Tylee, Lay people’s attitudes to treatment of depression: results of opinion poll for Defeat Depression Campaign just before its launch, Brit. Med. J., 5 October 1996, 313, 858-859.
Royal College of Psychiatrists, Twenty-Third Annual Report, London, 1996
Royal College of Psychiatrists, Royal College of General Practitioners, Defeat Depression Campaign press packs, 1992 and 1994.


A copy of the paper, The Antidepressant Web, was sent to the RCPsych on 2nd December 1997



Contents page
Correspondence: matters arising