How to choose an antidepressant

Welcome to the Australian Prescriber podcast. Australian Prescriber. Independent, peer-reviewed and free.

I'm Dhineli Perera, your host for this episode, and it's a pleasure to be speaking to Professor Philip Boyce. Philip is a Professor of Psychiatry at the University of Sydney and Head of Perinatal Psychiatry Clinical Research at Westmead Hospital. Philip writes about the combination of evidence-based treatments and lifestyle approaches, together with shared decision making, for the management of depression.

Philip, welcome to the program.

Hi.

Philip, your article introduces the concept of biopsychosocial and lifestyle models when managing depression. Would you mind explaining what each of these terms are? What's the terminology there?

Okay. Well, the biopsychosocial and lifestyle approach really involves taking a very comprehensive assessment of your patient. That means understanding what are the things that are contributing to both the onset and maintenance of the patient's disorder. Oftentimes, we do find these are lifestyle factors such as poor diet or poor sleep patterns or poor sleep hygiene, use of illicit substances, too much alcohol. All of these may be contributing to the depression, and if they're present, they will maintain the depression.

We also need to take into account social factors that may be contributing. I work a lot in perinatal psychiatry, and so often the contributors to women becoming depressed are they don't have social support networks to support them or they may be in difficult social relationships, particularly interpersonal relationships, where there may be abuse or control. All of those things are contributing to the depression.

So, doing a biopsychosocial assessment really means trying to identify all the factors that are contributing to the person becoming depressed and maintaining their depression.

Mm-hmm (affirmative), and, I guess, acknowledging that without addressing those factors, that medication is only going to add a certain amount of relief to that. Removing the trigger is what you're saying, or at least addressing the trigger is what's really important.

Absolutely. You can diagnose depression using checklists and following diagnostic criteria, if they've got five out of nine symptoms, it means they have a major depression, but it really tells you nothing about the person that has that depression. We need to understand the person that's got the depression so we can effectively treat them.

What is psychoeducation and psychosocial factors? Are these things different, or is it a combination of things to address what you've already mentioned?

Yeah. The psychosocial factors are the psychological factors, chronic low self-esteem or being very anxious, that may be contributing to the depression. That's the psychological factors. The social factors are the things I've talked about. Social support networks. Have they got a job, for instance? Those are all very important factors.

Psychoeducation is what we explain to the patient about their depression. We need to explain what depression is. It's an illness that can be treated, they will get better from it, and, "These are the things that are causing you to become depressed. Maybe you're being bullied at work and that's a major contributor for depression, and we need to be able to deal with that to be able to help you come out of your depression."

We need to make sure they take the appropriate steps to help themselves in their depression, and this is where we need to educate them about lifestyle factors, having a healthy diet, exercising, having good sleep hygiene, and so on.

I found it interesting that you state that cognitive behavioural therapy, so I'm going to call that CBT from here on, is equivalent to drug therapy in mild-to-moderate disease. What do you think are the factors that stop CBT being used more frequently?

Access. It's fairly straightforward. We have created more access for people to have CBT with focused psychological treatments and the better-outcomes approach, so GPs can really refer patients to psychologists to get CBT, but in some places there are not many psychologists. That is just a fact of life that there may not be people that can provide CBT for patients, particularly in rural and remote areas. Access is very, very difficult. Fortunately -

What about, though, with telehealth? Especially last year, we saw the emergence of that becoming almost mainstream. Does that change things for access to CBT?

That's certainly improved it. I really do think it has. During the last year, we've all been doing a lot of telehealth and providing CBT on Zoom. But really, we also need to recognise that there are not enough providers of CBT, and that's where the problem may be, with long waiting lists. I know people I see that if they want to be able to get to see a psychologist, they may have to wait three or four months. So there is access; there's also a lack of providers for those services. Although, it's getting better, I think, with telehealth. But another way it can be provided, and that's online therapies. There are some very good cognitive behavioural programs which people can do on their own computer or their tablet or their mobile phone.

Is there an easy point of reference or a one-stop shop for GPs to refer people on?

I don't really think there is a sort of central reference point. I think it really is the GPs knowing their own network and what resources they have in their own network. I actually haven't got the website where you can go and find out about digital therapies, but there is one [E-Mental Health in Practice resource library]. These are online CBT courses, generally six sessions, that the person does on their computer or their phone.

Okay, great. We can always put a couple of website addresses if you do come across them in your time.

Sure.

We can attach it there if people are interested.

Okay.

What are the main types of depression that antidepressants are indicated for, and what type of concurrent psychological therapies are recommended? We've talked about CBT.

Yeah. Well, interpersonal therapy is another psychological therapy that's shown to be quite effective. Mindfulness-based CBT is another one that's got good evidence base. But when we're thinking about medication, I think there are two things we need to take into account here. Firstly, we need to take in the type of depression the person's got, and also we need to pay a lot of attention to what their preference is. Some people want to be on medication for their depression, and if they're offered psychological therapies, they tend not to do so well, and vice versa. Working with women in the perinatal period, many of them don't want to be on medication. They prefer to keep well away from it.

Mm-hmm (affirmative). Yeah, okay. So, the types of depression, then, would you say, are they the more severe depressions that you would…

Yeah, absolutely. I think some people who've got really a severe major depression, I'll be thinking of going in early with medication but also providing the psychological therapy as well. If they have melancholic-type depression characterised by psychomotor symptoms, that's sort of slowed down or got agitation, they absolutely need to be on an antidepressant. If they've got a psychotic depression where there are delusions and/or hallucinations present as well, they should be on medication.

Mm-hmm (affirmative). Your article also neatly lists the four major considerations for choosing an antidepressant, and I guess this is the crux of your paper.

Yeah.

Could you please describe these for us?

Yeah, look, I think the things are you need to find the balance between tolerability and efficacy. We need to find an antidepressant that's effective for the person and is well tolerated. Otherwise, if they don't take the antidepressant, they're not going to get better, and if they've got significant side effects, they're not going to be taking the medication; they won't persist with it.

Unfortunately, we have a conundrum in a sense in that the more efficacious the antidepressant is, the more grunt it's got, if you like, the less tolerable it is. They've got the more side effects. So we've got to balance that out, and I'll just do a promotion for the College of Psychiatrists. They've just published an update on their mood disorders guidelines, and in there we talk about what antidepressants would you choose in the first instance for this particular patient. Here, we've really balanced out the tolerability with efficacy. So we find reasonably efficacious... well, all the antidepressants, they're all relatively equivalent in their efficacy, but efficacious ones that are actually also tolerable, as our first-choice antidepressants.

So, that's the first consideration we have to make. Secondly, we try and match the antidepressant to the symptom pattern the patient's got. If they've got significant anxiety, they may do well with an SSRI. If they're having trouble sleeping, they will do better with a sedating antidepressant, something like mirtazapine. We also need to think about safety, and that's absolutely critical. That means are there going to be any significant interactions with any other medication the person might be on? Are they going to have side effects that might affect the person because of their age? Again, when I'm working with women and they're pregnant, I need to make sure the medication is going to be safe for pregnancy. So we really need to take into account safety considerations as well.

The most important thing, again, it comes to patient preference. You'll say, "Look, I think you need to benefit from an antidepressant. What are your thoughts about that?" and they say, "Well, my mum was on antidepressant X and she had terrible side effects, so I don't want to take that one," or, "My friend recommended being on a different antidepressant, and she did really well on that. Can I take that antidepressant?" So really listening to what the patient tells you. Or they may have been on an antidepressant beforehand, and they say, "Look, I was put on to this particular antidepressant and all I got were terrible side effects and no benefit from it at all."

So, we're talking about efficacy, tolerability, safety and patient preference.

Absolutely. Yes.

I think that really neatly segues into what I was thinking as I was reading your paper, that this is really an interesting shift in results because a lot of people, when they were at university and learning about antidepressants, the general rule of thumb was that some of the what we call dirty drugs, that's not really a term we'd use with a patient, but were the ones that had impacted so many receptors, so their breadth of side-effect profile was large and their tolerability low. But from the table in your paper, it looked like things like, for example, TCAs were now considered to be more favourable. Has there been a shift in opinion of things like TCAs?

Yeah. Look, that's a very interesting question. Obviously, there's been a huge change in the way people think about them. There was a time that's all we had, that and the MAOIs, and then, when the SSRIs came onto the market, it really changed the game.

Yes. They were considered the clean ones, right, because they were so selective. Yep.

That's what they said. They're not that clean, I don't think, at all.

No. No, they're not.

They were considered very easy to prescribe. You only had basically one dose: one tablet, and that was all you needed. With the old tricyclics, you had to get five or six tablets at nighttime, and it was complicated dosing and so on.

I think people then became unfamiliar with using things like the tricyclics or the MAOIs and they did fall out of favour, but I still find them very, very effective for the right patient; right clinical profile for that patient. If they've got a severe depression or severe melancholia, I generally go to a tricyclic pretty early on because I know they're effective; I know how I can manage their side effects.

Again, one of the other things we used to say when the SSRIs came on the market, it didn't have any side effects. Well, we now know that's really not true and patients often don't talk about the quite debilitating side effects they get with SSRIs and SNRIs, particularly things like sexual dysfunction, which patients, unless you ask them about it, won't tell you.

Part of the psychoeducation, going back to what we were talking about before, is when you prescribe a medication is explaining to them how will they take the medication. It sounds crazy to say this, we've got to explain to them carefully you've got to take it every day. I have patients that come back to me and I said, "Well, how do you find the medication?" They say, "Well, some days it works and some days it doesn't." I say, "What do you mean?" and they say "Well, I only take it when I'm feeling sad."

So you need to explain how they take the medication; what side effects they can expect to get. When they start an SSRI, they're going to feel a bit nauseated. They may get a headache for the first few days, they may get some diarrhoea, they may feel a bit agitated, so explain to them what's going on so it doesn't come as a surprise to them.

Okay. Then, along that same vein, Philip, can you tell us a bit more about why the duel-acting antidepressants seem to be more efficacious, based on that table you listed with the odds ratios?

Absolutely. Well, they're more complex drugs. Things like venlafaxine, it's a more complicated drug. It's targeting both the serotonin and noradrenaline receptors, and so you're getting a broader spectrum of activity. But it's targeting two different systems, and so we'd expect there to be more potency from them.

Yeah. I guess the other table that I found interesting was the limitations table, and it's a great summary of adverse effects and their frequency. I wonder, is there a questionnaire or survey available for prescribers to run through with their patients that will help prioritise these limitations for individual patients' needs and their preferences?

I don't know of any particular questionnaire. I think a lot of decision aids will go through this. One thing we do talk about in those set of limitations, and something we probably have not paid enough attention to, is the withdrawal of antidepressants. I think, when we're prescribing an antidepressant to a patient, we forget to tell them, "If you come off this medication, you're going to get some symptoms; withdrawal symptoms. So don't come off it abruptly, come and talk to me when you want to come off your antidepressant." We think withdrawal syndrome is an important limitation in the antidepressants that we use because most patients... only about 50% will respond to the first antidepressant we give them, so we may need to switch antidepressant. So we want one we can switch easily to and from.

Mm-hmm (affirmative), because I really liked the rule, the little rule that you introduced, which was that tolerability was more important than efficacy for mild-to-moderate depression, whilst efficacy trumped tolerability for the severe depression.

That's correct. Yes.

When prescribers are trying to match the antidepressant to other comorbidities like diabetes and cardiovascular disease etc., how do they decide which are the most important comorbidities to consider?

Again, that's a really difficult question, and it's a very important clinical issue. I guess, first of all, you've got to look at what comorbidities the patient has, and patients with depression will often have other comorbidities. For example, if they've got diabetes, you don't want one that's going to make them put on weight or increase their appetite. That's fairly straightforward and simple. If they've got cardiac problems, we don't want an antidepressant that's going to have a detrimental effect on the heart. For example, like the tricyclic antidepressants.

We also need to look at what other medications they're taking. Often they will be on a tranche of other medications, or may be, and so we do need to be looking at any individual interactions that the medications may have with each other.

Mm-hmm (affirmative). Are there common ones that you keep an eye out for?

I don't. I often have to look it up because I can't remember all the drug interactions.

No, of course not. No.

I think, if you try to just go on memory, you make mistakes, so I generally... I need to do a quick check on the list.

No, that's a good answer. That's usually my answer as well when people ask me about drug interactions: I look it up.

Yeah. For example, an important one is if they're on an anticoagulant and you want to prescribe an SSRI. I need to check to see if there's going to be an interaction.

Yes, definitely.

Because that's a really important one because the SSRIs can increase bleeding.

Yes. Okay. Obviously, that's something that if it's an internal bleed, then the patient might not know until very late in the piece, which is-

That's absolutely very true. Yeah.

Yeah. Have you had patients, Philip, who have refused to take every antidepressant you've prescribed, and have you ever retried something that has failed due to them not giving it enough of a chance?

I've had patients, you try them on an antidepressant, they feel a bit nauseated the next day and it didn't make them feel happier.

Yep. They then stop it.

Yeah, it does happen, and you'll try something else, the same thing happens. Then you have to spend a fair bit of time talking with them, you have to rethink how you're going to prescribe it, and the adage, "Start low, go slow," is very, very important. So you maybe get them to take half or a quarter of the dose and slowly build up the dose so you're not giving them intolerable side effects the first instant they take the antidepressant. And stick with them and talk with them, listen to their concerns. Most often, I can get them to take an antidepressant, but it does take a fair bit of work.

Would you almost recommend that if GPs are managing this that they do schedule in appointments to see patients almost weekly for the first month when they start an antidepressant, just to touch base, or at least a telehealth chat?

Yes, absolutely. I think waiting a month before you see them again when the prescription runs out is not a good idea. We know, curiously, that sort of 50% of people don't pick up the second prescription. Plus, I think we need to be interested to see how they're going, not forgetting that sometimes, particularly with the SSRIs, they can be quite activating, increase suicidal ideation, and I think we need to be monitoring that really carefully. So I think a telehealth consultation after a week is probably absolutely necessary.

Excellent. Well, that's unfortunately all we've got time for for this episode. I feel like I could chat to you for ages with this, Philip. But thank you so much for joining us today.

It's been a pleasure.

[Music]

The views of the hosts and guests on the podcast are their own and may not represent Australian Prescriber or NPS. I'm Dhineli Perera, and thanks for joining us on Australian Prescriber Podcast.