Social Audit Ltd

P O Box 111 London NW1 8XG

Telephone/Fax 44 (0)171 586 7771

Dear Dr Kendell

Six weeks have passed since I sent you a copy of my paper, The Antidepressant Web, which is shortly to be published on the Internet in the hope of promoting some discussion. I do therefore look forward to hearing from you, to learn whether the College has to any extent revised its views in the light of the evidence I sent. I have received no acknowledgement yet.

I would of course also welcome a response to the points raised in my interim response to your letter of 28^th November and the enclosed Comments from Malcolm Lader. Now that I've had a chance to think about his paper, some further questions arise:

1. Lader argues (section 6) that BDZs, not SSRIs, produce a "central withdrawal reaction" and says this makes dependence on BDZs more of a reality and more of a problem. He says the distinction is marked by the appearance, with centrally acting drugs, of symptoms on withdrawal "which were not present when the medication was initially given". This reasoning seems extremely perplexing, and conflicts with the views of the College and the Committee on Safety of Medicines, as well as some of Lader's own:

[a] It was precisely the lack of evidence of "new" symptoms on withdrawal that persuaded a whole generation of psychiatrists and GPs that dependence on benzodiazepines wasn't a problem. Malcolm Lader was one of the very first to get the point, but even he didn't put his finger on it for over 20 years. Indeed, the lack of "new" symptoms emerging with BDZ withdrawal is emphasised in the recent College paper, Benzodiazepines: Risks, Benefits or Dependence? (January 1997)

"withdrawal symptoms … may be hard to distinguish from other anxiety-related disorders … There is controversy about whether symptoms persisting for weeks and up to a year are really withdrawal reactions or the manifestations of a chronic underlying neurosis or an exacerbation of the underlying condition triggered by tranquilliser withdrawal."

Malcolm Lader claims that these 'new' symptoms, "are clearly seen" when benzodiazepines are withdrawn, whereas the CSM attributes the huge under-reporting of BDZ withdrawal problems to the fact that this evidence was "unclear":

"As the symptoms of withdrawal were often the same as those that were the reasons for treatment, it may have been unclear to doctors that whether this was a withdrawal reaction or a recurrence of the underlying illness, the latter potentially leading the further benzodiazepine therapy. It is for this reason that we suspect that there have been very few suspected adverse reactions to the CSM regarding withdrawal reactions, as the issue was not identified by prescribers."

Moreover, how would one explain the great preponderance of withdrawal reports for SSRIs? Would this really have happened, if it were so much harder to distinguish withdrawal from symptoms of relapse or unwanted effects of these drugs?

[b] The following are examples of withdrawal symptoms reported for SSRIs which might be counted as 'new' which are not conspicuously associated with the symptoms of "complex depressive illness" to which Lader refers. The list is not complete: sensory disturbances, hallucinations, delirium, fever and flu-like symptoms, craving, disequilibrium, paraesthesias, tingling extremities, gait instability, rigors, tinnitus, twitching, tremor and other movement disorders and occasionally extra-pyramidal effects, dizziness, vertigo, poor co-ordination, visual disturbances, gastrointestinal symptoms (nausea, vomiting, diarrhoea, abdominal cramps, distension), and tachypnoea, fever, cyanosis and possible fitting in a neonate. Numerous other psychiatric and neurological symptoms have of course been reported. Nevertheless, Ashton & Young (in press) have concluded that SSRI withdrawal symptoms "differ qualitatively from the usual side effects profile of SSRIs and from the illness for which they were prescribed"

[c] Not least because "depression" has become a term which now embraces countless signs of psychic distress, SSRIs are frequently prescribed for conditions other than "depression". The US paper by Gregor et al. (1994), cited in my last letter, reported that "depression" was listed in the prescriber's notes for only 20% of women outpatients, 12% for men.

I mention this not because I would want you to think I need reminding again how serious depression can be, but to emphasise this would limit the numbers of 'new' symptoms that might be counted when SSRIs are withdrawn.

In short, I think the distinction between BDZs and SSRIs, in respect of appearance of 'new' symptoms on withdrawal is spurious and confusing. I think many people may be as baffled as I by the statement (section 8) that "benzodiazepines produce problems related to central dependence (what is this?) whereas the antidepressants probably do not induce these changes (what are these changes?). Rather there is a rebound of the changes in central biochemistry which the antidepressants produce". Apart from the points made above, Lader has argued elsewhere that BDZs produce rebound changes in central biochemistry and that rebound is due to the same mechanisms as withdrawal symptoms.

2. Malcolm Lader states (section 7) that "Unlike the benzodiazepines there is no evidence of escalation of dose and no evidence whatsoever that antidepressants can be abused on the street". However, he will appreciate more than most how understandings of these matters change with time - and I'm sure would agree with me that concluding that there is no risk, merely because "there is no evidence …" of it, would be like a politician saying "never". Look how quickly understandings have changed in this field: after the Defeat Depression Campaign was launched, the RCGP said it could find no evidence that antidepressant rebound and withdrawal ever happened (See my paper, p.97). The fact that this evidence had been there for years only underlines my point.

I hope Lader will change his mind on this, also taking into account the contradiction here between what he and the College have said. He contrast the BDZs with SSRIs in this respect: "However, with the benzodiazepines there are other problems such as dosage escalation …" - but the College has concluded that "Dosage escalation in uncomplicated cases is rare"

Is the Royal College really not concerned about the evidence relating to SSRI dosage escalation and diminution of drug effect? I am getting increasing numbers of letters from people who, like me, would very much want to know.

There is in fact some evidence of street abuse of SSRIs and I understand this topic will be also covered in the Ashton & Young paper. Even so, Malcolm Lader would confirm through his involvement with the CSM, that

the MCA/CSM never issued warnings on BDZs relating to any feature of dependence other than withdrawal symptoms, and that no such warnings ever appeared in data sheets. Since 1963, the CSM has received negligible numbers of Yellow Card reports relating to any manifestation of dependence other than withdrawal symptoms - conspicuous as these would be. Three months ago, I invited the MCA "to cite any published report giving convincing evidence that craving, drug-seeking behaviour or dose escalation are remotely characteristic of benzodiazepine dependence in therapeutic settings." They have failed to do so, but if the College knows otherwise, I should be pleased to learn from you.

3. Malcolm Lader's view (section 8) that, "the question as to whether there is a risk of dependence is … a matter of semantics", seems to me to underline the point that the College has behaved quite improperly in telling people there is no risk of dependence with SSRIs. It certainly seems ironical that the College's ability to disabuse people of the facts should have been used as one of the main measures of the effectiveness of the Defeat Depression Campaign. I find it quite irksome that some £65,000 of public money was spent on measuring the extent to which the public believed this propaganda; this hardly amounts to an "audit". It was, however, gratifying to learn from your letter that the College procured only "modest changes" in the public mind. Shades here of Samuel Butler: "The public do not know enough to be experts, yet know enough to chose between them".

4. Lader claims that "the data would indicate that relatively few patients (on SSRIs) compared with the benzodiazepines are disadvantaged clinically to any material extent". I do not think this assertion can be supported, but what evidence does he have for it? None is cited, but I would be pleased to be informed of any. In the meantime, I refer you to the MCA/CSM study, poor as it is, which suggested that, in about one in five of the withdrawal cases surveyed, the patient had to go an another drug - about the same proportion as restarted paroxetine and were unable to withdraw from it within three months (Antidepressant Web, p. 105). That seems highly material to me.

5. I expect Malcolm Lader will revise his view about claims for long term effectiveness of SSRIs (section 9), when he appreciates that the studies involved in drug registration utterly failed to distinguish between relapse and withdrawal. See my paper, pp 115-6. This is a certain recipe for getting it wrong, and it alarms me that the regulators let it go by.

I am sending a copy of this letter both to Malcolm Lader and to Professor Pereria Gray at the RCGP and look forward to hearing about the position of the Royal College of Psychiatrists on these matters.

Yours sincerely,

Charles Medawar

CLICK HERE TO READ ON

	Contents page
	Correspondence: matters arising