Thank you Mark. This is a very helpful analysis of where we are.
I respect Awais Aftab very much. However, his recent commentary on antidepressant withdrawal, following the publication of the JAMA Psychiatry paper, left me concerned.
I am not a researcher but I am well versed in much of the scientific literature in relation to antidepressant withdrawal. I remain concerned that lived experience of withdrawal is considered of lesser value by psychiatry as a profession than short-term, largely pharma-sponsored studies.
The Science Media Centre's involvement [approach to] the worldwide distribution of the JAMA Psychiatry is also of concern. The BMJ followed the SMC briefing and uncritically broadcast the views of career-long paid opinion leaders such as Prof Allan Young. The BMJ later had to amend their News article on this. Something is wrong here.
This is a much better and more thorough version of what I was hoping to write regarding the Kalfas paper, so thank you Mark! A few things to add:
Regarding the survey-based studies you have conducted, it’s reasonable to assume a selection bias for patients who had more negative experiences with antidepressant withdrawal. But, there’s no reason to think that a selection bias is occurring *within* the group of respondents, as you referenced.
The fact that depressive symptoms not only resolve, but *within days* of reinstatement strongly suggests that patients are often experiencing withdrawal rather than relapse in these studies. Antidepressants don’t usually induce dramatic mood improvement generally, and certainly not within a few days.
Thanks Daniel. A couple of notes: the survey we conducted in the public health system was a longer survey (we only reported the withdrawal part in this paper) and it asked about people's experience of using and stopping antidepressants. So it is certainly possible that people who answered it were motivated by a worse than average experience of withdrawal but this isn't necessarily the case. But, as you say, the relationship between duration of use and risk of withdrawal is unlikely to have been produced by a selection bias.
And yes strongly agree with the interpretation of the Rosenbaum study - you can imagine in studies (or clinical practice) where withdrawal effects are not measured that researchers (and clinicians) would see large numbers of people relapsing because their depressive scores shoot up. Hence all the confusion out there....
Great critique, Mark!
Thank you Mark. This is a very helpful analysis of where we are.
I respect Awais Aftab very much. However, his recent commentary on antidepressant withdrawal, following the publication of the JAMA Psychiatry paper, left me concerned.
I am not a researcher but I am well versed in much of the scientific literature in relation to antidepressant withdrawal. I remain concerned that lived experience of withdrawal is considered of lesser value by psychiatry as a profession than short-term, largely pharma-sponsored studies.
The Science Media Centre's involvement [approach to] the worldwide distribution of the JAMA Psychiatry is also of concern. The BMJ followed the SMC briefing and uncritically broadcast the views of career-long paid opinion leaders such as Prof Allan Young. The BMJ later had to amend their News article on this. Something is wrong here.
This is a much better and more thorough version of what I was hoping to write regarding the Kalfas paper, so thank you Mark! A few things to add:
Regarding the survey-based studies you have conducted, it’s reasonable to assume a selection bias for patients who had more negative experiences with antidepressant withdrawal. But, there’s no reason to think that a selection bias is occurring *within* the group of respondents, as you referenced.
The fact that depressive symptoms not only resolve, but *within days* of reinstatement strongly suggests that patients are often experiencing withdrawal rather than relapse in these studies. Antidepressants don’t usually induce dramatic mood improvement generally, and certainly not within a few days.
Thanks Daniel. A couple of notes: the survey we conducted in the public health system was a longer survey (we only reported the withdrawal part in this paper) and it asked about people's experience of using and stopping antidepressants. So it is certainly possible that people who answered it were motivated by a worse than average experience of withdrawal but this isn't necessarily the case. But, as you say, the relationship between duration of use and risk of withdrawal is unlikely to have been produced by a selection bias.
And yes strongly agree with the interpretation of the Rosenbaum study - you can imagine in studies (or clinical practice) where withdrawal effects are not measured that researchers (and clinicians) would see large numbers of people relapsing because their depressive scores shoot up. Hence all the confusion out there....