The following attempts to summarise the main points arising in two further letters exchanged on this subject - SA to MGH on 5 January 1999, and the MGH reply a week later.
1. Re "the initial publication of your work in the name of two Lilly employees, one year in advance of your definitive report". |
|
I agree that such a presentation scarcely counts as a publication - but then it is all the more perplexing to see Dr Rosenbaum (as co-author) actually citing his own findings to Blomgren et al., 1997. (See Ref. 9, in Zajecka et al., 1988). |
|
2. Re: authors' names omitted from the report published as Rosenbaum, Fava et al., 1998. |
|
|
|
3. Re: qualification of Lilly employees as authors |
|
In that case, Wilson (and presumably also Blomgren) do not meet the criteria for authorship laid down by NEJM. "Each author should have participated sufficiently in the work to take public responsibility for the content" and "authorship credit should be based only on substantial contributions to (inter alia) final approval of the version to be published". I cannot reconcile your claim to be fully in control of publication if your findings are cited to a non-publication whose principal author, and one other author, appear not to have been qualified as authors. |
|
4. Re: input of Lilly employees into writing of the manuscript. |
|
|
|
5. Re: the influence of trial design on the conclusion that there was little or no risk of withdrawal symptoms with fluoxetine. |
|
My point was that research carried out to this design could be expected to quantify the obvious - to point to the relative lack of withdrawal symptoms, within one week of drug discontinuation, with a drug (plus active metabolites) having a half-life of several weeks. |
|
6. Re: similarities between withdrawal problems with SSRIs and benzodiazepines (BDZs) |
|
I accept what you say, though it greatly surprised me. In the early 1990s, you were among the first to recognise that withdrawal symptoms with SSRIs could be a problem, and you say you were concerned that others were not aware of this. Presumably you suspected that, with SSRIs, "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". This was how the APA (1990) defined dependence on benzodiazepines - a problem that had been also overlooked for years, which had only just come to light. I cannot understand why you concluded there was no possibility of some link between the two. |
As articulated in prior communications, that symptoms emerge following discontinuation of medication to which there has been physiological adaptation is a commonplace in medicine. So perhaps the hypertensive is dependent on medicines in that sense. The differences between SSRIs and sedative-hypnotics are enormous, however, in terms of drug-seeking, dose escalation, rapid as opposed to delayed (weeks) therapeutic effects, reward or pleasurable sensations from acute dosing, toxicity etc |
6. Re: BDZs and SSRIs - continued |
|
|
|
6. Re: BDZs and SSRIs - continued |
|
|
|
6. Re: BDZs and SSRIs - continued |
|
|
|
6. Re: BDZs and SSRIs - continued |
|
|
7. Re: methodology used to evaluate withdrawal reactions to SSRIs having very different half-lives. |
|
My point was that you used sharp instruments and a limited time-frame to investigate withdrawal reactions with paroxetine and sertraline, but not fluoxetine. |
|
8. Re: depression as a symptom of withdrawal |
|
You previously participated in trials in which 30% of patients became measurably depressed, following withdrawal from fluoxetine - and yet you developed a 43-item list of Discontinuation-Emergent Signs and Symptoms (DESS) which doesn't mention the word depression. Not much prospect of agreement here. |
|
8b. Re: what if the study reported in Biological Psychiatry had been extended for several weeks? |
|
What you would see would seem to depend very much on what you were looking for, and how you went about it. You seem to be reasoning that dizziness would be the hallmark withdrawal symptom to look for, simply because it was the most often reported DESS both with paroxetine alone and for paroxetine and sertraline combined. But you already knew that 30% of patients became measurably depressed, following withdrawal from fluoxetine - whereas dizziness hardly featured as a withdrawal symptom at all. Therefore you might have expected to see a higher incidence of reports of psychic distress of the kind reported in your study in Table 2. See how low in the order dizziness comes, by contrast with symptoms linked with depression: |
Our answer to your question about an extended interruption of fluoxetine was based on data from an experiment that had been done several years earlier as you know |
9. Re: predominance of depression as a withdrawal symptom |
|
On the contrary, I am challenging your assertion that depression should never be regarded as a withdrawal symptom with fluoxetine. You clearly demonstrated the predominance of depression as a withdrawal symptom with sertraline and paroxetine - though you label it 'relapse' - and the same symptoms (notably sudden worsening of mood, irritability, agitation) predominate with fluoxetine too. |
|
10. Re: Your emphasis on withdrawal symptoms other than those linked to depression. |
|
My main point was that your emphasis on withdrawal symptoms that would typically be attributed to the 'flu' or to some other medical illness, seemed highly inappropriate given the prevalence of depression resulting from fluoxetine withdrawal. |
|
11. Re: the timing of onset of depression, in relation to drug discontinuation, as indicators of either a withdrawal problem or relapse. |
|
I was drawing your attention to well-established advice to the effect that "withdrawal syndromes developing within a few days of withdrawal cannot be attributed to a relapse of the disorder for which the antidepressant was first prescribed, because this would take several weeks to appear " You did not comment on this, and appear to have disregarded it in your study - and essentially on this basis you concluded there is no significant risk of withdrawal with fluoxetine, and that fluoxetine is effective in the long term. These conclusions seem quite unwarranted to me. |
|
12. Re: defining as a "depressive relapse" more severe cases of depression occurring during the period of drug withdrawal. |
|
I'm afraid you have completely lost me. My point was that this was not depressive relapse but that depression should be counted a withdrawal symptom. If depression 'melts' with restoration of medication, it adds to the evidence that depression is a withdrawal symptom, and nothing to do with 'relapse'. |
|
13. Re: risks resulting from the failure to recognise depression as a major symptom of withdrawal. |
|
The risks are of widespread drug-induced depression, compounded by the mistaken belief that antidepressants are effective in the long-term. The risk with fluoxetine is so much the greater because depression (rather than 'flu-like symptoms) is the most distinctive withdrawal symptom, and readily mistaken for 'relapse'. The data in the four Lilly-sponsored trials clearly points to this risk, even if the interpretation and presentation of data point studiedly elsewhere. |
|
--- END ---
From: | Charles Medawar |
Date: | 13 January1999 15:38 |
To: | Rosenbaum Jerrold F., Md |
Re: | Your comments of 12 January |
Dear Drs Rosenbaum & Fava,
Thank you for your email. It was helpful to get clarification on points 1 through 4, though it seems there are profound disagreements between us on the substantive issues. I would also think there was little chance of resolving them through further correspondence but perhaps, if you had a chance to read the documents I sent, you might get a better idea of my perspective.
If you would like me to respond to your further points, please let me know. I'm quite prepared to spell out in more detail why I remain very concerned, but would be reluctant to do so unprompted because I fear this dialogue must already seem pretty indigestible to most website visitors. Incidentally, they are hardly a "following", and few are patients. Though patients have commented and contributed much more, most visitors seem to come from pharmaceutical companies and from professional backgrounds - mainly in medicine, academe and government. It is of course open to any of them to raise any queries they may have with any or all of us.
Yours sincerely, |
Charles Medawar |
From: | Rosenbaum Jerrold F., Md |
Date: | 13 January1999 17:45 |
To: | Charles Medawar |
Re: | Your comments of 12 January |
> Dear Drs Rosenbaum & Fava,
Thank you for your email. It was helpful to get clarification on points 1 through 4, though it seems there are profound disagreements between us on the substantive issues. I would also think there was little chance of resolving them through further correspondence but perhaps, if you had a chance to read the documents I sent, you might get a better idea of my perspective.
[Rosenbaum, Jerrold F.,Md] We'll try to have a look when there is an opportunity to do so. If you would like me to respond to your further points, please let me know. I'm quite prepared to spell out in more detail why I remain very concerned, but would be reluctant to do so unprompted because I fear this dialogue must already seem pretty indigestible to most website visitors. [Rosenbaum, Jerrold F.,Md] [Rosenbaum, Jerrold F.,Md] Perhaps you're right, but if questions come up, we can try to address them; indeed, we may have questions for you. Incidentally, they are hardly a "following", and few are patients. Though patients have commented and contributed much more, most visitors seem to come from pharmaceutical companies and from professional backgrounds - mainly in medicine, academe and government. It is of course open to any of them to raise any queries they may have with any or all of us.Yours sincerely,
Charles Medawar