This BMJ Talk Medicine podcast can be listened to here. It was published on 28 March 2023. In this podcast Dr Mark Horowitz gives clear and practical advice. Given the importance of this subject and the limited availability of support services, I have transcribed the podcast in full [as below]. Please note: do not stop your antidepressant without discussion with your doctor.
Antidepressant withdrawal with Tony Kendrick and Mark Horowitz
BMJ talk medicine
28 March 2023
The theory that depression is caused by serotonin deficiency has become embedded in our practice over the years, and can be a persuasive explanation of the condition for patients considering starting antidepressants. We talk to Tony Kendrick about the evidence (or lack thereof) to support this hypothesis, and what that means for the efficacy of antidepressants, particularly SSRIs. Later on, we speak to Mark Horowitz to discuss how to come off antidepressants safely, and manage withdrawal symptoms, and how to advise and support our patients as they gradually reduce, then stop, their medication.
Tony Kendrick is a retired GP, and NICE committee member for the 2022 guideline update on depression in adults. He is also a professor of Primary Medical Care at the University of Southampton.
Mark Horowitz is a training psychiatrist, working as a clinical research fellow at North East NHS Trust, and as an honorary clinical research fellow at UCL, he is also the co-founder of outro.com.
‘Antidepressants and the serotonin hypothesis of depression’. BMJ 2022;378:o1993
‘Stopping antidepressants’. Royal College of Psychiatrists. 2020
Tom Nolan: Welcome to Deep Breath In the podcast from the BMJ sponsored by Medical Protection where we tackle the everyday challenges of being a GP. The idea that depression is a lack of serotonin in the synapses of the nerves has become embedded in our practice over the years and can be a persuasive explanation for patients considering antidepressant treatment – to treat a deficiency of serotonin. But is this theory correct? Or could it do our patients harm? When it comes to stopping antidepressants, overlooked for many years, what’s the best way to advise patients on this? Or are we the best people to do this at all?
In today’s episode we will speak to GP and NICE committee member Tony Kendrick about the serotonin theory and researcher Mark Horowitz for an update on antidepressant withdrawal.
I’m Tom Nolan, a GP and clinical editor of the BMJ and I am joined by our usual co-hosts Navjoyt and Jenny. Hi Navjoyt, how are you?
Hi Tom, I’m Navjoyt Ladher, I’m a clinical editor at the BMJ and locum GP in London.
And Jenny, Hi.
Hi Tom, I’m Jenny Rasanathan, a family medicine doctor and clinical editor for the BMJ.
So welcome, we are going to talk about depression today, which probably we should talk about more, as it is literally every single day as a GP that you can be pretty much guaranteed to see someone with depression, em, and so it forms a big part of our work, doesn’t it. Eh, we have talked about explanations on the podcast before and how important they are. I’d like to start with the explanation about depression: and you know, there is a serotonin explanation, there’s others. Is this something you feel good at? Comfortable at? Jenny, you are looking kind of perplexed.
Jenny Rasanathan: I would not say that I feel overly comfortable with it. I very much agree that this is something we see all the time and I think, you know, especially during the pandemic you would see kind of an anxious flavour of depression as well. Just like so often anxiety feeding into depressed feelings. I think there was, perhaps was a less need to try and explain what depression is, or, em, even what anxiety kind of was on a biochemical level, because it was so prevalent. Em, everybody was feeling so bad and I think that was the most useful way in, as it’s not surprising given, you know, the pandemic and all the other negative things that have happened in various places that you might be feeling this way. And, em, that seemed to resonate with people.
Tom Nolan: So it didn’t really need an explanation because it was mostly pretty clear and you through in a few other life events, and . . .
Jenny Rasanathan: Yeah, I think most of, most of the consultations, you know, often people would be tearful and I think they already kind of knew how they were feeling and that they probably already had an idea what they were hoping to get out of the consultation: whether it was a referral to a therapist or, or starting an antidepressant. Em, so a lot of the consultations focussed on that kind of discussion around when is it appropriate to start medicine and what that medication does.
Tom Nolan: Yes, I’m not sure, or if it was my inexperience, or something changed, but I feel that I talked less about naming the diagnosis or trying to explain what’s going on than I used to, but . . . do you feel that way Navjoyt?
Navjoyt Ladher: Yeah, I was just thinking that as you were talking there, that’s exactly my experience as well, but I feel that for depression I don’t explain what the cause is, because I think we don’t know what the cause is often. I think often it’s multifactorial, there are so many things that can feed into that, and that might be something that you know I, we might discuss with the patient. But also, I think in the course of putting this episode together I think, you know, for a condition we do see all the time in general practice and we treat often, there is a lot that is generally not very well understood about depression. Em, you know, we don’t really know how the medications work, we don’t really know why it happens. Em, you know, we know what potentially some triggers and precipitants can be for an episode but, you know, the kind of you know, it is so much more than a serotonin deficiency that you can top up by taking these medications. I think that we have always known that, em, but, em . . .
Tom Nolan: But have you ever . . . do you think you have ever used that serotonin hypothesis explanation for somebody who is . . .
Navjoyt Ladher: No, I don’t think I have. I think, I think I know and can remember drawing diagrams of like, if you were to take these medications, as in SSRIs, we think that it will bring your serotonin levels up, but that is not predictable, and em, we don’t . . . and hopefully that will help you.
Tom Nolan: Yeah, what was the diagram of? Was it like a synapse?
Navjoyt Ladher: It was a graph . . .
Tom Nolan: Alright, okay . . .
Navjoyt Ladher: Well actually what I was trying to convey in that graph, which will probably lead on to our next, the other interview that we have on this episode. Was that I was trying to convey that when you stop, when you stop taking SSRIs the levels don’t come down immediately but you have to have been taking them for a little while. I think that I learned that from my GP trainer at the time. I have been thinking about this as we were putting this episode together and I was thinking I’ve no idea if that is right or wrong? And possibly it is wrong, because we are learning so much more about the kind of, you know, we are learning that there seems to be more looking at these medications and questioning some of our previous assumptions. So, em, I just find all that really fascinating that for a condition that is so common and so part of our everyday management . . . there is really a lot we are familiar with in our management and can manage well, but a lot of the underlying mechanisms are a bit kind of less well understood.
Jenny Rasanathan: And that can be really frustrating for people, right, because, it is part of the popularity perhaps of the serotonin hypothesis as it is so convenient. Like, how easy . . . if it did work to be able to say: oh well, you know you have to have certain levels to maintain regular mood and when they go down we can replace them. I understand the appeal of the model and in some ways the simplicity of being able to explain the, em, mechanism and use of a medication.
Tom Nolan: Yeah, yeah, it’s a tricky little . . . it feels like a great short cut, doesn’t it. Okay, so serotonin . . . and people say, yes, yeah, I get it. And so our first interview is about that and so there has been some, em discussion . . . there’s been some in the BMJ and elsewhere about the serotonin theory and so we spoke to one expert about that, em, which we will get on to it. Now that I’ve mentioned that I feel that we have to go straight to that interview. Shall we do that? So I spoke to Tony Kendrick, he’s a now retired GP and also NICE committee member for the recent depression update, and he wrote an editorial for the BMJ about the serotonin hypothesis.
Tony Kendrick: So I am Tony Kendrick, I am a GP by background although I recently retired for the second time. Eh, I am professor of Primary Medical Care at the University of Southampton and I do research on depression, its assessment and treatment.
Tom Nolan: And recently in an Editorial for the BMJ you discuss the serotonin theory of depression and antidepressants. Could you, em, maybe start with an overview of that? What made you want to write about that?
Tony Kendrick: Well I was asked to write about it. The BMJ commissioned the Editorial. Em, I think it was a very important issue, which, eh, really speaks to what do we say to patients about antidepressants when we are, eh, discussing whether or not they should start them. It was about Joanna Moncrieff and colleagues review in Molecular Psychiatry, where they did an umbrella review, that is to say, a review of reviews, em, of evidence for the serotonin hypothesis. The idea that somehow a deficiency of serotonin, a lack of serotonin perhaps in the synapses between the nerves causes depression and that antidepressants and most specifically the SSRIs work by correcting that kind of deficiency, em, or chemical imbalance, em there is a popular term of serotonin. The review was pretty convincing that there is very little evidence for this hypothesis and, eh, therefore, this kind of casts doubt on the rationale for taking antidepressants. Em, where the review went too far I think was in suggesting, not in its conclusions, but in its press release and in appearances by Joanna and others afterwards, that this meant that antidepressants don’t work. Em, and so, my first take on this was, along with a number of other psychiatrists, em, neuropsychopharmacologists, em , is the fact that this particular hypothesis doesn’t em, explain how SSRIs work, it doesn’t meant that they don’t work. Em, we judge that on the basis of clinical trial evidence and, em, taken in the round, there is evidence that antidepressants work, although em, they are not perfect, they don’t work for everybody. Em, sometimes we have to try them to see if people respond, but there is a greater response than there is to placebo even when you take into account the fact that people do get unblinded in placebo-controlled trials due to the side-effects. So when you compare antidepressants with placebos that have anticholinergic effects, you still see a benefit from antidepressants. So I think we can say that antidepressants work for many people, em, in the short term.
The other issue though, this is where I think em, Professor Moncrieff and colleagues have a point, is that we should not be telling patients that they have a chemical imbalance. We should not be saying that, em they have a deficiency in any way of serotonin. We have shown previously, em, in a meta-analysis, a meta-synthesis of qualitative research, em, that of studies showing what effects people taking antidepressants long term, that a significant factor in that is their belief about whether antidepressants are needed long term and that often hinges on what they have been told when they start taking them, and I, as a GP and many of my colleagues I have spoken to have been used to in the past to say; ‘take this antidepressant it will build up your levels of serotonin over time and you will hopefully feel better, it may not work, but em, it is one of the treatments that we have’. So I’m guilty of saying, in a way that you have a deficiency of serotonin. The problem with that is, that rather like the deficiency of thyroxine or a deficiency of insulin, it kind of suggests that you should take it for good. That this is maybe something you need for life and it adds to the problem that people have coming off antidepressants because we know that there are withdrawal symptoms, we know there is risk of relapse when people come off. But we do know people can come off successfully and that it is probably better for them to come off after a couple of years or longer if need be, em, rather than em, to take them for good. Particularly after the age of 65, and as a 66 year old I start to take this more seriously, you do get more and more side effects of antidepressants and some of them can be quite severe.
Tom Nolan: So what do you say or what do you advise listeners to say to their patients then as an alternative?
Tony Kendrick: If you were to come and talk to me in my surgery and tell me that you were feeling depressed, once we had gone over the symptoms of that and what was going on in your life, in your relationships, in your work, in your family life, I might em, consider offering you an antidepressant but it would necessarily be a discussion between us and you would need to think about whether you wanted to take one. You might ask why should a chemical help with depression if it is caused by problems in my life, em, or in problems in the way that I am feeling in the way I am thinking, in other words social problems or psychological issues, em, well we do know that there are biological issues in depression specifically things changing in the brain and although it is not a simple deficiency of serotonin that we can correct with an antidepressant, nevertheless the chemical effects of antidepressants we know can help the brain adapt to the stress of environmental problems, em, the stress of what’s going on in your social life and indeed the way you think, and em, therefore can actually be of some use in depression. But they have side-effects and it’s a discussion we should have giving you the full information.
Tom Nolan: So I thought that was pretty clear that don’t tell people they have a serotonin deficiency, which em, which is helpful isn’t it. It is rarely we can be so definitive about something particularly when it comes to mental health. And em, and that point when at which you are prescribing an antidepressant is really important 6 months, a year later down the line. Eh, em, I also, people also seem to appreciate that thinking ahead to their recovery as well. When you are better, which can be a nice, quite important part of that conversation as well when people are often feeling quite helpless and hopeless and not seeing that that would happen. That this is something that they can stop and I suppose carry on with their lives.
Navjoyt Ladher: Yeah, I completely agree. I think you are right, that is a kind of encouraging thing to be thinking about when you are probably really wanting help. At a stage when you are about to start a medication. I do wonder, I don’t know how many GPs would describe depression as a serotonin deficiency? My sense is most people get that’s not what’s going on. But, perhaps people like me, are explaining the mechanisms of how SSRIs work in a way that suggests that it is a serotonin deficiency. But, I think, my sense is that, yeah, GPs and probably patients know it is more complex than that because of all the other things that we do to treat depression as well, talking therapies and that kind of thing. The kind of theory falls down which you think of all the other interventions which are also effective for treating depression. But so, I think it is a useful reminder of not only how we explain things but then also, as you were saying, how we can . . . what things might be important to cover when we are kind of counselling patients at the start, when we are initiating these medications.
Tom Nolan: When I think of this, because I feel like I agree, that I think we have also moved on from that, but em, it almost seems the same thing when it comes to CBT, em, talking therapy solutions, like maybe sort of over-promising: like, ‘it’s ok because when you have CBT you will feel better’ and well I guess the new NICE guideline has been updated specifically to say actually that it’s not just everyone that gets CBT, but people should be offered a range of talking therapies according to their preferences and what they have had experience before and I suppose what will suit them most. I am yet to see that change really in terms of what talking therapies are offering. But I get the same kind of response often from patients these days when you talk about CBT, they like kind of ‘whooo, I tried that it didn’t work’, em, in the same way, well that antidepressants didn’t help, and so I feel like sometimes we are running out of options. The last thing you want to do is to leave a patient feeling helpless or that you have run out of ideas.
Navjoyt Ladher: Yeah, it is always worth exploring those kind of, em, you know when a patient says that though isn’t it, because you know the antidepressants didn’t work, it may be that one was tried and you know maybe not even at, might have been a sub-therapeutic dose or something, and so, em, yeah. But I hear what you are saying. You know, I think in terms of offering a range of talking therapies I kind of offer a referral to IAPT and then kind of it’ s in their hands, so you are right, that kind of change hasn’t filtered through to me, for sure, yet.
Jenny Rasanathan: Just picking up on a slightly different point, I was going to say that what I have seen, kind of increasingly, em, in conversations with people who decide to start an antidepressant is that they kind of ask more, or now ask, whereas they didn’t before about ‘going off it’. Right, like it was, it has always been a common concern, understandably, you know, does this mean ‘I have to be on this medicine for the rest of my life?’ But people specifically saying is it easy to go off these, how would I want to taper down, I think that there is this, em, more common understanding of these medicines as something you do use kind of as a temporary assist, em, as opposed to you know necessarily being on it for years. Em, so frequently people saying, ‘and when I want to go off this how do I do it?’ I say, you haven’t even tried it yet.
Navjoyt Ladher: Well Jenny, it is so interesting that you brought that up because our next interview is kind of very much related to this point and I kind of think is probably something that will affect the way we talk about antidepressants when we are initiating them, because, we, well I had a chat with Mark Horowitz who is a psychiatrist, academic clinical fellow, who has done quite a lot of work on tapering antidepressants and, em, suggests that actually the way it is done mostly at the moment is too quick and too much for patients and that withdrawal from antidepressants is a kind of under-recognised phenomenon. So, he has some tips on how we could maybe approach it better.
Dr Mark Horowitz: My name is Mark Horowitz, I am originally from Australia where I finished medical degree and I have worked in London. I am currently working as a clinical research fellow in psychiatry at North East London, NHS Trust, and I am an honorary clinical research fellow at University College London and a lot of my work is around rational psychopharmacology and safely stopping psychiatric drugs.
Navjoyt Ladher: Okay, yeah well on that note that’s what we are hoping to talk to you today about: about antidepressant withdrawal syndrome and its management, which is a topic that, em, has got a bit more attention recently or at least that’s how I came to notice it with a new NICE draft quality standard about stopping antidepressants that was published, em, I think it was in early January around that kind of time. My initial impression was like, well I think as a GP I am already kind of tapering antidepressants when I stop them but actually from the work you’ve been involved with, em and the work that you’ve done, em it would suggest that antidepressant withdrawal is more of an issue than we perhaps have realised and current practices are falling short for many patients, So, I wonder if that is where we could start really with the basics, like what actually is the problem here?
Dr Mark Horowitz: Sure. I think that you are exactly right, there has been more attention, em to this issue of antidepressant withdrawal over the last few years. In part it has come up because of patient groups complaining that their withdrawal has not been overseen well, that they have suffered withdrawal effects, it’s been misdiagnosed as relapse, and they feel that doctors are tapering them too quickly off their medication causing them trouble. This has led to a lot of patients seeking help from outside of the medical system. They end up going to peer support websites or Facebook groups to get advice from other patients on how to come off their drugs and reported often that it is more successful than the advice they get from doctors. Which is quite surprising, eh, my take on this is, eh, I myself, learned how to come off my antidepressant from peer support websites despite the fact that I have done a PhD at the Institute of Psychiatry on the way that antidepressants work, and I work with leaders in the field. So it was quite an inversion of the usual flow of information. For many years the NICE guidelines had said that antidepressant ‘discontinuation’ – a euphemism that actually came from drug companies – is mild and self-limiting for about a week or two, occasionally it can be worse. And that is what has informed GPs and psychiatrists for many years. It turns out that from studies that have been done and peoples reports that antidepressant withdrawal is more common, more severe, and more long-lasting than we had first thought. So, a systematic review done a few years ago found – you could just look at double-blind, randomised control trials, about half of patients who stop antidepressants will experience withdrawal effects. There is less certain evidence on what proportion will experience severe effects. From surveys of patients, up to half of those who had experienced withdrawal effects about 1 in 4 will experience severe withdrawal effects – that might be a slightly skewed population, people with worse than average withdrawal effects but it is quite a large proportion of people and there are case series of people who have experienced months and sometimes more than a year or years, of effects from withdrawal symptoms. So, first of all people find that quite surprising: how can withdrawal last for months and years when most of the drugs only take days, or in some of the longer acting drugs like Fluoxetine, weeks to come out of the blood. So why do the withdrawal symptoms last so long? The explanation for that is that it is not about the drug leaving the body it is about the residue of the drug’s effect on the brain. So the brain adapts to the presence of the drug, like lots of drugs and medications that we take. It’s those adaptations to the drug that can take months or years to resolve. It is the time it takes for the brain to get used to less drug around that leads to withdrawal symptoms. Withdrawal symptoms are really the difference between what the brain expects and what it has become accustomed to for a drug and what is actually supplied by the drug. We know from neuroimaging that antidepressants can leave changes in the brain that last for months or sometimes years.
Navjoyt Ladher: Right, that’s fascinating. I just wanted to ask you when we refer to antidepressants are we thinking specifically of SSRIs or is this other classes as well?
Dr Mark Horowitz: So, all classes of antidepressants have been found to cause withdrawal effects, the tricyclics as well as the newer drugs. The drugs that I am primarily focussed on are the drugs that are commonly used in clinical practice and they are the SSSRIs, the SNRIs, and drugs like Mirtazapine. These antidepressants effect serotonin, noradrenalaine and a variety of other neurotransmitters throughout the body and brain. They have a myriad of effects: they effect the hormonal system, they effect the haematological system, they don’t just target the brain. All of those systems will adapt to the presence and when you take away the drug there will be withdrawal effects that effect all of those organ systems. So you get two major groups of withdrawal symptoms: you get physical symptoms, things like dizziness, shakiness of your feet, headache, nausea. There are some quite typical sensory symptoms, the famous brain-zaps. Then there are the emotional withdrawal effects and some of the emotional withdrawal effects from antidepressants include, low mood, increased anxiety, trouble sleeping, panic attacks, tearfulness, and sometimes people will have suicidal impulses or thoughts. And the first thing to say is you can see how this could lead to confusion because it overlaps a lot with the symptoms of depression and anxiety, the reasons that people were often put on these medications.
Navjoyt Ladher: So, I guess the first step in distinguishing withdrawal symptoms from relapse symptoms is just an awareness that you can have these emotional symptoms, but then in practice what else might you be looking for as you described in that example. Is there new symptoms or is there anything else that clinicians could be attentive to or patients in that?
Dr Mark Horowitz: That’s right, this is the key issue. How can you distinguish because it can be confusing. There are three main ways to help distinguish withdrawal from relapse. The first one is the timing. If someone has stopped their medication a day or two, or a week or so ago, there is a high likelihood that it’s withdrawal. Relapse shouldn’t happen for weeks or months, it will depend on if someone has a relapsing-remitting condition, it should depend on their normal rhythm, which might be months, it might be years. The second one is the nature of symptoms and the accompanying symptoms: so if depressed mood and anxiety makes it difficult to distinguish, so if someone is feeling terrible the next questions to ask: are you also dizzy, have you had a headache, have you had nausea, have you had any electric zaps – there is quite a good list in the NICE guidelines what to ask people. And I would add to that, the point I made before, it is good to distinguish what were the symptoms of the underlying condition to more clearly distinguish from what is going on now? And the third one, which is more helpful in retrospect, is, if you reinstate the drug then the symptoms generally dissipate within a few days if its withdrawal, and generally takes a bit longer if it is a relapse. That relationship becomes a bit little less clear the longer you wait before re-instating. So if you re-instate within a few days or a couple of weeks afterwards almost everybody responds quite quickly as you get later out it is not quite as clear the response to withdrawal symptoms.
Navjoyt Ladher: So say, em, you are seeing your patient who is coming off their antidepressant and you make the judgment that the symptoms they are experiencing are due to withdrawal, I can imagine that can be difficult if you are not experienced in managing this in practice where you know . . . my instinct would be that patient needs to go back on their antidepressant. How do you tend to manage that in your clinic? How frequently are you seeing patients for example to follow up?
Dr Mark Horowitz: Right, so I mean the first thing to say is that if someone is experiencing withdrawal effects, it is not a sign that they need the drug, it is a sign that they are coming off too quickly. Because you can come to a similar conclusion for other drugs. If someone is experiencing withdrawal from Valium you wouldn’t conclude that they must need the Valium. The same for smoking. If someone had withdrawal symptoms from coming off cigarettes you wouldn’t conclude they must keep smoking. So, I don’t think withdrawal symptoms are a good reason to put someone back on the drug and keep them there. Relapse might be but withdrawal symptoms are a different story. So I spend a lot tome talking with patients about what they might expect when they come off the medication and putting in place contingency plans. The contingency plan is generally the same: which is if you have withdrawal symptoms, let’s either pause to wait for them to go away, as they are often temporary if they are mild; if they are quite unpleasant to go back a step or two till things have resolved and then go down slowly. So we have learned a lot over the last few years about how to safely take people off antidepressants to avoid withdrawal effects. There are two overarching ideas. One is to go at a rate that the person can tolerate, which is often much more gradual than people have been told in the past. The guidelines now from NICE, which links to a very useful educational pack from the Royal College of Psychiatrists, recommends that people should stop over weeks or months, and in fact some of the guidance suggests that people should stop over a year, especially for high risk antidepressants like paroxetine or venlafaxine and particularly for people who have been on them long term or have had trouble stopping in the past. Antidepressants don’t act on the brain in a linear way. That means that doubling dose doesn’t double the effect. They actually have a pattern of effect on the brain that is a hyperbola, which basically means that very small doses have quite large effects on the brain and the effect bottoms out at high doses. For example, taking a very common drug like Citalopram, 1mg, a tiny dose, actually has almost half the effect of 20mg. That means, as you reduce the drug, so going through what a clinician would normally do – based on historical practice – going from 20mg to 15mg only causes a very small decrease of effect on the brain, going from 15mg to 10mg has a slightly larger effect, 10mg to 5mg larger again, the reduction from 5mg Citalopram to 0mg actually has about a fifteen-fold effect on the brain as compared to 20mg to 15mg. And so whilst it makes intuitive sense to go down by these nice even numbers but in terms of what the effect on the brain is it gets stepper and steeper, it’s like walking down a very shallow path that suddenly turns into a cliff. I think a lot of people get into trouble in those last few milligrams, that’s what people tell me: the first few reductions was relatively easy but the last few milligrams put me in a screaming mess. Because people have been unaware of this they can’t see this very steep last step and they keep falling over it in the dark. So what makes more sense, rather than reducing by even amounts of dose, say 5mg for each step with Citalopram, it makes more sense to reduce by even amounts for effect on the brain, which means as you go down a path that is steeper and steeper you have got to slow down and in fact you may need to go down to doses, less than a milligram for lots of antidepressants that are used. So, for example, rational, pharmacologically rational dose reductions would be something approximate to: 20mg, 10mg, 5mg, 2.5mg, 1.25mg, 0.6mg. Those are 50% dose reductions at each step, getting smaller and smaller to follow that curve of effect on the brain. That is the example regimen given by the Royal College of Psychiatrists advice on stopping Citalopram and in practice I have seen that work for a lot of patients who couldn’t get of their medication doing the usual 10mg, 5mg, 0mg reduction regime. Now, the question that everyone puts here is: how could you possibly do that in practice when antidepressants, for example Citalopram, comes only as a 10mg tablet – that’s the smallest possible.
Navjoyt Ladher: You read my mind that was going to be my next question.
Dr Mark Horowitz: So I give lectures to GPs and they sort of understand what I am saying then they groan because I am giving them a problem and I am sorry I am trying to solve a much bigger problem by giving this smaller problem. I am very apologetic. It is not a good idea to dose every other day. The half-life of Citalopram and most antidepressants, except Prozac [Fluoxetine] is about 24 hours. If you dose every other day that is two half-lives, the drug goes down by half and then to a quarter. That is quite a large change in plasma levels and we know that can precipitate withdrawal effects in susceptible people. So the first thing you can do is split tablets. Round tablets can be halved and quartered, so you can actually get down to 2.5mg using a tablet cutter that you can buy at local pharmacies or online. Doses beneath that, basically you need to use a liquid preparation of medication. So once you are using a liquid version of a drug you can make small adjustments. If someone is saying, the rate of reduction I was outlining before which was 50% reductions about every fortnight or month, if that is causing too severe withdrawal effects, you can halve the rate and go to 25% per fortnight or month or, I should point out, there are some patients out there that are only able to tolerate reductions of around 10% of their most recent dose made every month or so. That means, because it is of the most recent dose, the reduction becomes smaller and smaller as you get down to lower doses and that’s how it takes some people a year, or a couple of years to come off their antidepressants. I have certainly come across patients who have told me they have taken three, four or five years of slowly titrating down liquid drugs, particularly of drugs like Venlafaxine or Paroxetine, in order to come off those medications.
Tom Nolan: Well
Jenny Rasanathan: That was so great.
Tom Nolan: So much new and incredibly useful practical information. Thank you Navjoyt, thank you Mark. That was really helpful. Maybe one of our most useful interviews ever.
Navjoyt Ladher: Wow, yeah, I mean I think Mark Horowitz has done a lot of, I guess teaching and lecturing and writing on this topic and I think it all stemmed from his own experience as well of trying to come off his antidepressant. It is really interesting isn’t it, as I was talking to him I was reflecting back on situations where I thought someone was having a relapse on the basis of the symptoms they were having, because our convention is that you can come off an antidepressant, you know 4 weeks, start tapering, halve the dose, do every other day, if that’s a bit fast . . . you know, might need to be an extra week or something, but. So this was eye-opening. I think in more recent years obviously there has been, more, these connections between patients, em, your patients peer support groups and that kind of thing that has really raised attention on this kind of problem and that is a good example of how listening to patients can be so important and trying to connect those dots.
Jenny Rasanathan: Yeah, I totally agree. That was so clear. So practical you can tell that he has been a GP educator. Really anticipating questions and thank you Navjoyt for asking about how to divide pills, even to cut doses, and the piece about not taking a pill every other day was something that I did not know. I learned a lot from that.
Navjoyt Ladher: Yeah, I think it links a bit to what we were talking about before we went into that interview about conversations that you have when you are initiating antidepressants. One of the things I spoke to Mark about, which we didn’t have time to air in that interview was about a question that I suspect we all get fairly commonly when we are discussing antidepressants with patients: are they addictive? And the answer to that is no, not in the pure psychiatric sense of the word ‘addictive’ but I have always felt very confident about being ‘No, No. They are not addictive you can stop when you need to’. But actually I think there is a conversation to be had that for some patients coming off, can take months if not longer and that is because the body does adapt to these medications. So that will probably change the way that I speak to patients about antidepressants when I am starting them.
Jenny Rasanathan: I very much agree. Even thinking about the intro to this episode and the serotonin hypothesis, not that any of that proves that hypothesis or anything, but if we are again thinking about the biochemical basis of this, you know, when you are putting something in your body that changes the levels of those neurotransmitters – not that it is replacing them or whatever – but when there is an adjustment in those levels it does make intuitive sense that you can’t just immediately take them away, because your body has acclimated and adjusted to that. And again, it not a one to one replacement, em, but it does make sense that there is perhaps more of a conversation to be had about the fact that people may really struggle and it is not as fast as go down by half every two weeks and you will be fine.
Navjoyt Ladher: Yeah, yeah, I think this idea that it is just a bit slower and a bit more drawn out than perhaps we were used to doing is, em, is something I will definitely talk to patients about.
Tom Nolan: So I have a question or something we could perhaps work through together. So when that patient comes in for the appointment that which is ‘I’ve been taking this for this long and I want to discuss stopping this medication’. Em, I suppose the best case scenario is that for many people they can still stop their medication within that one month or two months – obviously there is factors to discuss. I didn’t quite catch in that interview, whether for most people that is achievable?
Navjoyt Ladher: Yeah, I think from the evidence that mark cited, you know from RCTs, he was saying about 50% of patients get withdrawal symptoms when they stop, and obviously, I didn’t ask and I don’t know, like what the rate of taper was in those studies. But yeah, so I think it is a proportion of people it is not everybody, from what he was saying.
Tom Nolan: So it is still reasonable to, em, discuss in the best case scenario that this is your original plan and if that isn’t working for you, if you are getting withdrawal effects, then here’s what . . . I guess that’s more appointments. I am always thinking about how many appointments is this going to take? It is really bad. But if I can cover it now and send the patient armed with, you know, the information and the understanding they need to do it themselves if they want to take that on. I think that is a better thing.
Jenny Rasanathan: But it is so funny you saying that as I was thinking the same thing in my head. You know, I don’t want to set someone up for failure, by saying ‘oh some people take a really long time, you might come to struggle to come off.’ I don’t want of like to give them the expectation that this is a hard thing to do, because that might not be their experience. Not because it couldn’t happen but because it might not. 50% of cases, right. So in my head, what is the approach? I worry that I veer too much towards giving too much information, like ‘oh this bad thing might happen’ . . ‘so make sure you reach out if it does’. Versus do you just set up a follow up appointment in a month’s time or do you put a reminder in your calendar to check in with this person, you know, and all of that takes resources.
Tom Nolan: And in a system where you can get an appointment in the next two weeks I guess you don’t feel so bad, but em, having an appointment where it is actually going ‘fine thanks’, is, em, okay . . .
Navjoyt Ladher: The Royal College of Psychiatrists guidance that mark mentioned in that interview is actually directed at patients. It is about how to come off your antidepressant medication. So I think that could be a really useful resource that we can share with patients, and then, you know depending on the patient, you can get in touch and tell us how this is going and we can see you again.
Tom Nolan: So that was my question right back in the intro is, you know are we even the right people to guide people, obviously we are we are GPs, we are good at this, but you know if some people have a better experience on a Facebook group, within certain caveats for that, is that a reasonable alternative for people?
Navjoyt Ladher: It is a good question isn’t it. I guess people have turned to these resources because they are not getting what they need from their GP surgery or whoever their healthcare provider is, so I don’t know . . .
Tom Nolan: It is self-help, do we . . .
Jenny Rasanathan: But you know the time required to get an appointment with the doctor. If you realise you are struggling when you are trying to taper off these medicines and even if you weren’t to get an appointment for the next morning you are still suffering. Hey I get it, I get why you would go online and try and seek online support resources, trying to figure out what is going on, even if that was to prepare you for your appointment or what have you.
Tom Nolan: Yeah, so rightly or wrongly I think increasingly I am trying to give people permission to, you know, or at least to have a conversation about how they might look to do things like this without me. Shall I stop doing that? You are both looking at me like . . .
Navjoyt Ladher: Well the thing is, actually how much control do we have anyway. Like patients could do that if they wanted to anyway. So, em, I don’t think that you are wrong Tom.
Jenny Rasanathan: I don’t think you are either, I just em, I think you know when I was most recently in practice in New Zealand, I am really bad at keeping boundaries, I’d be like ‘oh, just feel free to check in with me’ and I would be like overworking, by a lot, you know. I don’t think you are wrong.
Tom Nolan: Do you give out your number? Oh no, that is another conversation.
Jenny Rasanathan: Another podcast episode. I used to actually when I was in residence. I don’t any more.
Navjoyt Ladher: There we go, everyone go and see Jenny. [laughter]
Jenny Rasanathan: I am not advocating that approach.
Tom Nolan: So Navjoyt, I want to go back to those diagrams, I feel like you need to . . .
Navjoyt Ladher: No I stopped doing those a long time ago
Tom Nolan: But was it a hyperbolic graph, was it? What was the shape of your curve . . .
Navjoyt Ladher: It was an exponential curve, no it was just a linear line. Linear line, that doesn’t make any sense [laughter]
Tom Nolan: You work for the BMJ do you? Research editor [more laughter]
Navjoyt Ladher: It is just a positive correlation. But I have stopped doing those graphs and if I haven’t I definitely stop. But yeah, I have definitely learned a lot and like I said when we first started this conversation, once again I am reminded just how much there is to know and actually how much we can learn from our patients.
Jenny Rasanathan: I have a question for you both: what else do you tell people when they come in and say they want to get off an antidepressant? Cause I got into the habit of kind of trying to have a little chat about what is going on in the rest of their life? Do you think it matters? Like people say there is never a good time to do X, Y or Z. Is it that way with going off an antidepressant?
Tom Nolan: Yeah, I think I would try to do that. Yeah, just the story of how, what was the helpful things in their recovery, assuming that is the context. I always learned this thing that you do your CBT to change, that is a thing that will change your thinking patterns and behaviour patterns longer term and once those are established that is the best time to start thinking about stopping the medication. Em, because you have still got those loops or pathways, if you want to keep a neurological way of thinking about it, established. I don’t know if that is right or not?
Navjoyt Ladher: That makes sense. I do something similar, I do explore kind of where patients are with them and sometimes, you know it kind of depends on the kind of nature of what the patient is saying. Like you know ‘I’m thinking about it, do you think the time is right?’ Or for some people it is like ‘I want come off this, how should I do it safely?’ Those to me are two different conversations.
Jenny Rasanathan: Yeah it’s a good point. I sometimes wonder, you know if it is worthwhile like planning it out a little bit? And then of course there are other circumstances if people are trying to conceive etc etc which is for another podcast as well.
Tom Nolan: So I think that is a good place to end. I have learned a lot from today’s episode. Thank you to our guests, Tony Kendrick and Mark Horowitz and thank you Navjoyt for that great interview and thank you Jenny for your insights as well.
At the foot of this podcast, there is an opportunity to comment. The character limit is 1000. This is a full version of my comment:
Re: Antidepressant withdrawal with Tony Kendrick and Mark Horowitz
Dear Tom, Jenny and Navjoyt,
Thank you for this explorative and helpful podcast on antidepressants.
It is always welcome to hear a primary care perspective, particularly in the open way this podcast considers the knowns and unknowns.
Both your speakers helped guide your open discussion. It was disappointing that Professor Tony Kendrick, a career-long key opinion leader, who has helped shape guidelines and national policies on antidepressant prescribing, failed to address, in any practical way, antidepressant dependence and withdrawal. Instead, Professor Kendrick put forward an argument that a key issue in widespread longer-term prescribing [prescribing out with evidence-based recommendations] may relate to the “beliefs” of patients. Up until a period of years before his retirement, Professor Kendrick was a paid partner of the pharmaceutical industry. Professor Kendrick should not be singled out here, rather, we should note that he has long been part of a most influential group of ‘educators’ on prescribing.
The interview with Dr Mark Horowitz, in contrast, felt like a new horizon where the current generation of doctors are thinking a new.
In this podcast it was stated that Dr Horowitz is “a good example of how listening to patients can help” Tom you suggested that Dr Horowitz may have given “one of our most useful interviews ever”. I reckon that many patients taking antidepressants might agree.
Thank you again for this important podcast.
Dr Peter Gordon
Retired NHS Psychiatrist
Conflicts of interest: I have been a campaigner for the UK to introduce Sunshine Legislation. My wife is a GP. I am dependent on a SSRI antidepressant: a prescribed drug that I have been unable to successfully withdraw from.
Some relevant background material:
The following petition is now closed. It is worth reading the written submissions in response to this petition. It is my understanding that this petition received more responses than any other petition lodged with the Scottish Parliament:
Lay peoples attitudes to the Defeat Depression Campaign, published in the BMJ, 1996
Profile: 36 years at the Royal College of psychiatrists, Psychiatric Bulletin, December 2016:
Some of the past declarations of competing interest as given by Professor Tony Kendrick in his publications:
Declarations by other paid opinion leaders in British Psychiatry can be found here.