|Coleg Meddygaeth Prifysgol Cymru
University of Wales College of Medicine
|Adran Meddygaeth Seicolegol / Department of Psychological Medicine|
|Adran Cymru y Gogledd / North Wales Department|
|Pennaeth Adran / Head of Department|
|Yr Athro /Professor Michael J. Owen|
|Dr. D. Healy (Cyfarwyddwr)|
|Sarah Wark, Senior Scientific Assessor|
|Medicines Control Agency|
|Market Towers, 1 Nine Elms Lane|
|LONDON SW8 5NQ||17th December 2001|
Dear Miss Wark,
Many thanks for your letter of the 7th of December. I am surprised, however, that it is taking so long to address the question of the evidence base for warnings about the increased risk of suicide with benzodiazepines. I await with interest any details on how you propose to resolve the quandary this delay suggests you are in.
On the issue of agitation and SSRIs, I am pleased to hear that agitation will be listed as one of the psychiatric reactions listed as an undesirable effect of SSRIs in the forthcoming SPC. But if this is the case, I fail to understand how you do not feel obliged to put a further warning about the likely consequences of agitation in patients who are ill to begin with. It baffles me that, at least in principle, you consider that agitation might not express itself in terms of suicidal ideation.
You raise the difficulty of distinguishing psychiatric effects of SSRIs from the underlying disease, but I cannot accept your point that this is difficult, given the background of healthy volunteer studies you have where agitation is clearly a consequence of SSRI intake. In fact you have had a study available since 1986 by Saletu et al, published in two different journals, showing a dose dependent induction of agitation in healthy volunteers with sertraline.
When it comes to reviewing the other healthy volunteer studies you have for evidence of agitation and related problems, I would hope your reviewers do not take the excruciatingly pedantic approach companies appear to take at a time like this when they look simply for the term agitation as it is used to describe experiences of healthy volunteers. They do this even though different companies, and each company at different points in time, have used various mapping systems for a range of terms such as tension, irritability, akathisia and anxiety as well as agitation and have mapped these onto agitation. If your search of this evidence, in contrast to company searches, is disinterested, I believe that it will become clear that effects, which may be generically described as agitation, occur following SSRI intake in healthy volunteers at a much higher rate than they occur on placebo or spontaneously. There will be a problem if this is not the result you get, as there are a lot of internal company documents pointing to such a link.
As regards the CSMs difficulties in distinguishing increased suicidality on SSRIs stemming from the drugs rather than from the disease source, the evidence is now rather conclusive. I would like at this stage to supplement the report I sent you earlier with a later draft of an article detailing this evidence. From this draft article, in Tables 1A & 1B you will see that the evidence for an increased number of suicides on SSRIs versus placebo is statistically significantly greater as is the evidence for increased suicidal acts of any sort on SSRIs versus placebo. The overall relative risk of suicidal acts on SSRIs versus placebo is greater than 2.0. The excess of suicidal acts on SSRIs holds whether the data analysed in terms of patient years or absolute numbers of patients. A similar finding is reported by Dr Laughren of the FDA in an article in this months European Psychiatry.
Investigational Drug, Patient No Suicide No
To make things crystal clear I have enclosed the figures above in this letter. In addition to the increase in suicidal acts, one of the interesting points about these figures that I have drawn to the attention of a number of scientific audiences and media sources is that companies have been categorising as placebo suicides and suicidal acts, suicides and suicidal acts that occurred during the washout period of trials or following the termination of the trial. I think there will be questions asked of these companies as to why they engaged in such practices. It is difficult to see how these questions will not rebound on you in the MCA. Were you aware this was happening? Did you condone it? What do you make of the correct figures?
Whatever way you analyse the data, either by adjusting the denominator to include washouts, or by excluding washouts, whether by analysing in terms of absolute numbers or by patient exposure years, there is an excess of suicidal acts on SSRIs and this is statistically significant. And, as you will find from the draft article, these findings are consistent with epidemiological data drawn from primary care in the United Kingdom. Indeed the degree of consistency is quite striking with the figures from RCTs closely overlapping with the figures thrown up by the Drug Safety Research Unit studies, whether these figures are expressed in terms of absolute numbers or patient exposure years.
As regards the issue of dependence on the SSRIs, a great deal hinges on your definition of dependence. I would accept that SSRIs are not addictive. By the statement that SSRIs are not addictive, however, I mean just like you that they do not cause craving and do not cause tolerance. If we all accept this definition, then the term dependence syndrome must mean something else. I dont believe that the MCA can have their cake and eat it on this one by then suggesting that a dependence syndrome is characterised by the production of cravings and tolerance.
If the presence of tolerance and craving characterises a dependence syndrome for you then I would respectfully suggest to you that based on comparator studies of the animal and human literature there is no more evidence that benzodiazepines cause a dependence syndrome than SSRIs do. If this is your position, then just as I have asked you to consider the apparent inconsistency between your views on benzodiazepines and SSRIs as regards suicidality etc, I would also ask you to consider the inconsistency in your views as regards dependence syndromes with benzodiazepines and SSRIs and remedy one or other of your positions.
You have talked about withdrawal reactions with SSRIs being relatively mild and lasting for the most part only a few days. This statement is at odds with both clinical experience and popular perceptions. I suspect the greater part of your data on SSRIs stems for the 60% of patients we know from research discontinue treatment with SSRIs within a month and I can accept that the characterisation of withdrawal reactions from SSRIs you outline might well apply to this group.
I do not believe, however, that you have data on patients who have been taking SSRIs for months or years and cannot then stop. It follows that I do not believe that you are in a position to properly characterise the problems faced by such patients. But even in the case of the milder withdrawal reactions that you concede occur, this would be evidence for most people in the street of a dependence syndromes.
As we correspond with each other Glaxo SmithKline and other companies are actively stating that their drugs do not cause dependence syndromes, or stating that they are not habit forming, or stating a range of similar things that will mislead the public into believing that these drugs do not cause withdrawal reactions of a severity that means that the patient may be effectively unable to halt treatment. Do you condone company statements of this sort or do you think companies are just not saying these things?
As regards the question of any lack of transparency by pharmaceutical companies, I think you need to revisit the experts who have given you an input on psychiatric matters, since the SSRI controversies first blew up, and ask them to outline for you exactly what potentially conflicting interests they have had, and exactly when they acquired these other interests. You should also ask them for any reports they may have written for any of the companies involved on issues to do with suicidality or dependence, that have you might not have seen. You might also ask for any unpublished data they have pertinent to these issues.
Finally, as regards your review of the current evidence on SSRIs and suicidality, can I make the following suggestion? There is a risk of an ongoing correspondence, unless your experts have the opportunity to persuade me that Im making an error of some sort in the points that I have been making on both scientific and public platforms. Effectively I would like to offer the MCA the chance to silence me (it would be a great relief to be able to withdraw from this issue). I suggest therefore a meeting and would be happy to travel to London at short notice. In the light of the serious nature of the problem, and the fact that this correspondence has now been running for the better part of two years, I suggest this meeting should be scheduled for sometime in the near future
David Healy MD FRCPsych Director, North Wales Department of Psychological Medicine
|Uned Hergest, Ysbyty Gwynedd, Bangor, Gwynedd LL27 2PW|
|Ffôn: (01248) 384452 Ffacs: (01248) 371397|
|Hergest Unit, Gwynedd Hospital, Bangor, Gwynedd LL27 2PW|
|Tel: (01248) 384452 Fax: (01248) 311397|
CLICK HERE TO READ ON
or HERE for full index of this correspondence