28th November 1997

Mr Charles Medawar Social Audit LimitedP.O. Box 111London NW1 8XG

Telephone: 0171 235 2351
Facsimile: 0171 245 1231

Dear Mr. Medawar,

I am writing in response to your letter of October 30th about this College's "Defeat Depression Campaign". I must say at the outset that I do not see any need for the "fundamental reappraisal of some of its key messages and recommendations" which you are advocating.

You suggest that the SSRI antidepressants have now been shown to produce dependence and that the risk of their doing so is "at least as great as with benzodiazepines." I do not accept this. You are confusing dependence with a withdrawal state. The International Classification (ICD10) distinguishes clearly between the two and provides a definition for each. Three criteria must be fulfilled for diagnosis of a withdrawal state: (a) clear evidence of recent cessation or reduction of substance use, (b) symptoms and signs that are compatible with the known features of withdrawal from that particular substance and (c) those symptoms and signs are not accounted for by any unrelated medical disorder. The dependence syndrome, on the other hand, is defined by six characteristics of which at least three must be present simultaneously. The six are: (a) a strong desire or sense of compulsion to take the substance, (b) impaired capacity to control substance taking behaviour, (c) a physiological withdrawal state when substance use is reduced or ceases, (d) evidence of tolerance, (e) preoccupation with substance use (i.e. giving up alternative pleasures or interests; devoting much time to obtaining, taking or recovering from the effects of the substance) and (f) persistent use of the substance despite clear evidence of harmful consequences.

You will see from this that the presence of a withdrawal state is neither necessary nor sufficient for a diagnosis of dependence. Nor do the two have a comparable clinical significance. Dependence, particularly on a drug commonly prescribed by doctors, is always a matter for serious concern, as in the case of the benzodiazepines. A withdrawal state, on the other hand, is in itself simply an adverse effect to be weighed in the balance with other adverse effects against the drug's benefits.  

As you probably know, a very wide range of drugs produce withdrawal syndromes. Even peripherally acting preparations like beta-blockers and H2 antagonists do so. Indeed, tricyclic antidepressants commonly do so but have never, despite being extensively used for nearly forty years, been shown to produce dependence. I agree that in recent years there have been an increasing number of reports of withdrawal states after stopping SSRI antidepressants, particularly paroxetine. The risk of a withdrawal state is a good reason for recommending gradual withdrawal from these drugs, but not for alarm and certainly not for questioning their therapeutic value. I agree, though, that the existence of these withdrawal states, although they do not at present seem significantly different from those associated with the old tricyclic drugs, is good reason for keeping a careful watch for dosage escalation, the development of a black market and other possible evidence of patients developing dependence. So ,far, though, I am not aware of any evidence of this.

Your second claim is that the College has been advocating the prescribing of antidepressants to too wide a group of patients over too long a period. You also talk almost exclusively about the SSRI antidepressants, although all our therapeutic advice dealt even handedly with these and the older tricyclic drugs.

I suspect that you may not appreciate how devastating depressive illnesses can be and how serious a burden they impose both on affected individuals and the community as a whole. People who have had the misfortune to suffer from a depressive illness and also, at some other time in their lives, a painful physical illness, frequently insist that the former was far harder to bear. In many patients, too, depressive episodes recur many times in the course of their lives with catastrophic consequences for their careers, their marriages and their self esteem. The Global Burden of Disease study mounted by the Harvard School of Public Health suggested that the burden (measured in disability adjusted life years) imposed by unipolar major depression alone was, world-wide, more important than malaria, diabetes, asthma, HIV, tuberculosis or road traffic accidents. And in what they called the "established market economies and formerly socialist economies of Europe" it was one of the three most important burdens, exceeded only by ischaemic heart disease and cerebrovascular disease.

I am confident that our campaign did not advocate any therapeutic policies that were not supported by random allocation trial evidence. In particular, there is strong evidence that the relapse rate after recovery from a depressive illness is substantially reduced by remaining on an antidepressant drug for four to six months after full recovery. Similarly, there is substantial evidence that long term treatment reduces the risk of further episodes in patients who have already experienced at least two episodes. The strength of this evidence fails off after three years or so, but this is because of the difficulty of maintaining a random allocation trial over very long periods of time rather than because of any evidence that antidepressant drugs eventually lose their therapeutic powers.

The College was also careful throughout the campaign never to advocate the prescribing of particular classes of antidepressants, still less the products of any individual manufacturer. The consensus statements to which you refer are not, as I think you realise, statements of the formal views either of this College or of the College of General Practitioners. They reflect the views of the psychiatrists and general practitioners invited to participate in the conferences because of their expertise. I am also assured by Professor Paykel, who acted as chairman and rapporteur for the first conference and was responsible for the final draft of the consensus statement, that no pharmaceutical company attempted to influence the content of that statement. Had they done so they would have failed.

Incidentally, Dr. Montgomery is not, as you suggest, "a leading figure" in this College. Although he was once chairman of our Programmes and Meetings Committee he was not a member of the Defeat Depression Campaign's management committee, nor did he attend the consensus conference on management. His close relationship with the pharmaceutical industry is also well known, and his views are generally treated with caution for that reason.

Finally, you make a number of assertions and ask a number of questions about the contribution of the pharmaceutical industry to the Defeat Depression Campaign.

The funds which the College received for this campaign came from many different sources, including the Department of Health, the BBC's This week's good cause', sales of publications and donations from the general public. The campaign's total income amounted to 449,800 of which only 129,530 (28.8%) came from pharmaceutical companies. It is true that most, but not all, of the companies who contributed were manufacturers of SSRI antidepressants and I have no doubt that one of their major motives was the hope that an increased recognition of depressive illnesses both by the general public and by general practitioners would result in increased sales for them. For our part, we recognised this from the outset, which was one of the reasons we were particularly careful never, to suggest that SSRI's were superior to other antidepressants, or that any individual drug was better than its competitors. As to the College's policy with regard to accepting financial assistance from commercial organisations, we are prepared to accept funds from any reputable organisation provided there are no strings attached i.e. provided that there is no attempt by the company concerned to influence College policy or College statements and no implication that we are recommending their products. We are also perfectly prepared to defend our policies to the Charities Commission if need be.

You ask for a copy of the audit of the campaign. There was no single audit, though a series of studies was mounted to measure changes in public attitudes to depression, based on 3 successive MORI polls of random population samples, and to measure changes in the attitudes and prescribing policies of general practitioners. Preliminary results have already been reported in Psychiatric Bulletin (Vol 21, pp 148-150 1997), in the American Journal of Psychiatry (Vol 154, Festschrift supplement, June 1997) and in the British Medical Journal (Vol 313, pp 858-859, 1996). The final evaluation of the impact of the campaign on general practitioners is currently being prepared for publication. The results of the final MORI poll will take some time to analyse but they do show widespread though modest changes in public attitudes and they will be published in due course. Finally, I enclose for your interest a brief paper by Professor Malcolm Lader written in response to a request from me to comment on the claims to in your letter that SSRI antidepressants have now been shown to produce dependence analogous to that produced by benzodiazepines.

Yours sincerely,

R.E. Kendell
Copy: Professor Pereira Gray
Royal College of General Practitioners


Contents page
Correspondence: matters arising