Management of Depression and Anxiety in Primary Care

Depression and anxiety are common in primary care and primary care physicians manage them regularly. In the Penn system, a collaborative care model known as Penn Integrated Care (PIC) allows for collaboration between mental health providers and primary care physicians. In this podcast, Kendal Williams, MD, MPH discusses this model with Joseph Teel, MD and Dr. Eleanor Anderson, a psychiatrist. We also review the fundamentals of the management of depression in primary care.

Kendal Williams, MD, MPH, is Director, Center for Evidence-based Practice, and Professor of Clinical Medicine. Dr. Williams is board certified in Internal Medicine. He currently sees patients at Penn Internal Medicine Radnor.

Refer a patient to Dr. Williams:

• Call our 24/7 provider-only line at 877-937-7366
• Submit via our secure online referral form

Transcript

Download a transcript of this podcast

Kendal Williams, MD, MPH: Welcome to the Penn Primary Care Podcast. I'm your host, Dr. Kendal Williams.

On this podcast, we seek to support primary care providers at Penn and in the larger world by exploring various clinical challenges most often with experts from within the Penn community itself.

If you've been listening to our podcasts thus far, you know that we've been tackling some of the most common challenges we face. And we'll be continuing that today as we explore the diagnosis and management of depression and anxiety in primary care.

Depression and anxiety are two of the most common reasons people seek help from a physician and much of the depression and anxiety that is treated in this country is managed through primary care offices.

With me to discuss this are two people who've given a lot of thought to the management of mental health in primary care:

Dr. Joseph Teel is the Vice Chair for Clinical Operations in the Department of Family Medicine and Community Health here at Penn. He is also the Associate Medical Director for Operations for the Primary Care Service Line and works closely in the Penn Integrated Care Program, a collaborative model between primary care and mental health professionals here at Penn, that many of you are already familiar with.

Hi Joe, welcome to the podcast. Thanks for coming.

Dr Joseph Teel: Thank you.

Kendal Williams: Dr. Eleanor Anderson is an Associate Professor at Penn in the Department of Psychiatry. She is the Medical Director for Consultation-Liaison Psychiatry, and the Chief of Integrated Services. She is the Medical Director for the same Penn Integrated Program I mentioned before.

So Eleanor, Dr. Anderson, thanks so much for coming.

Eleanor Anderson, MD: Thanks for having me. I'm happy to be here.

Kendal Williams: So let's first take a few minutes to talk about the Penn Integrated Care Model, what it is and how it came to be. And then, we'll tackle some of the practical aspects of managing depression and anxiety in primary care. So maybe Joe, I'll start with you. Can you tell us more about the program?

Joseph Teel, MD: Of course, the program is certainly not unique to Penn and none of us can really take the responsibility for developing this model. But we are following a fairly robust national standard around what usually gets referred to as a collaborative care model developed years in the past, that has been scaled in many institutions and even small community practices across the country that really looks to integrate and embed psychiatric access and psychiatric care within the primary care setting.

It allows patients of ours in the primary care world to have a navigated and embedded experience, and be able to address the anxiety and depression that you had mentioned without leaving their primary care home.

At Penn, that actually physically looks like some mental health professionals, such as LCSWs and other counseling professionals who are co-located with us in our practices. And then they have support and expertise through guidance by people like Dr. Anderson, who are overseeing their care and helping to guide clinical decisions and management.

Some of those patients also have to seek specialty care with a psychiatrist and our team is able to then connect them to specialty care, either at Penn or out in the community. It's been a terrific program over the past few years after implementation and a really great support to both the patients and myself as a primary care clinician.

Kendal Williams: Yeah. I mean, I think in a practice that does not have this program as a resource, they're often working with mental health professionals outside of their practice. But I would imagine it's not as coordinated. Joe, you worked in that environment for years, where you were trying to manage folks that were not in your practice. How are you finding this to be different?

Dr. Teel: This was probably one of the most transformative changes I would say to my practice. And I think if you ask the same too many of the primary care clinicians across Penn, that really has occurred in the past several years.

But prior to the implementation of our program, I would just tell people if they needed to connect with a counselor or a psychiatrist, I would literally just have them take out their insurance card, I would turn it over and there would be a number generally in the back that would say something like, you know, customer service or many times it would say for behavioral health services, you know, call this number. And I would say, "Call that number," and that was the extent of my ability to connect them with behavioral health services.

And so fast forward to today when I have staff and professionals as part of an integrated team that can connect, provide services or get them to a specialty service outside of Penn, is really just night and day.

Kendal Williams: So Eleanor, from your perspective, from the psychiatric perspective, how do you see the program?

Dr. Anderson: I think it's tremendously impactful. And I think one reason I've become so passionate about integrated care is that, as a psychiatrist, you know, previously I was an outpatient psychiatrist and I would see patients who had mild to moderate anxiety or depression and I helped them. And I think I provided a good holding environment for them. I helped them get better. But at the same time, it wasn't very psychopharmacologically complex for me, and I couldn't help but notice that some patients couldn't come see me because of insurance reasons and also that I wasn't necessarily working at the top of my training either. And I learned more about the integrated care model and learned that most mental health care in the country is actually provided in primary care clinics because of just lack of availability of mental health resources. And I just thought, "Oh wow. You know, my training can go a lot further through this model."

And so I think we've been able to help patients get access much more quickly to the right level of care with this program.

Kendal Williams: I'm just curious about some issues of ratios and so forth. So Eleanor, you supervise the counselors themselves, right? Or not supervise, but also you sort of support them, I suppose, in their work. How many do you support all at once?

Dr. Anderson: So in my current role, it's more administrative. But, as a supervising psychiatrist, when I first joined the program, I would, with one half day of my time, supervise one or two LCSWs, who would each be in a clinic that would have 10,000 patients each. And at any given time, the LCSWs carry a panel of maybe 60 to 80 patients.

Kendal Williams: So it really expands your ability to be effective to a larger audience in a much more streamlined model. So it's terrific. Joe, where are we at with rolling this out in the Penn practices?

Dr. Teel: So we've over the course of the past few years have been able to slowly scale to increasing number of practices.

We had initially started in some of our practices that had the highest sort of degree or burden of behavioral health concerns within the population. And we have then, you know, moved it into additional practices. We're up to, I think right now, 15 practices with new ones coming online actively over the course of the fall and into the winter. We had three practices that just came up in the past month.

And we are hoping to one day in the coming months to probably one to two years to be able to serve every practice within the Penn Primary Care Network.

Kendal Williams: So providers can expect that if they're not knowledgeable or having experience with the PIC program yet, they will soon. So that's terrific.

So let's spend the remainder of the podcast time sort of going through depression management in primary care, from your perspectives, given your sort of clinical expertise in this.

And I just want to start with a patient presenting to a primary care practice at Penn with symptoms of depression, and let's sort of follow that patient through the process. So in some cases, patients are coming to see us because they feel depressed and that's their primary issue. But in other cases, depression symptomatology is part of their overall set of issues.

If our systems are working properly, they will be screened in triage by a PHQ-2 evaluation, a two-question screen that may identify them as experiencing depression. The providers then, I think this works universally, are asked through the Epic system to do a more formal screening, the PHQ-9, to assess for the presence of depression.

Joe, is that how it's working?

Dr. Teel: Yeah. Certainly, I think it's very common through many healthcare systems universally to use the PHQ-2 to 9 screener as the most common way to identify depression in primary care setting. You're right that our staff are tasked with administering the PHQ-2 during a rooming process, or now even some patients will respond to those two questions before they get to their appointment through their EMR portal.

Once they've screened positive and responded in a positive manner to one of those, then you're right, then the remainder of the PHQ-9 is deployed. And sometimes a clinician will ask for that. And I know in some practices a staff member may hand them a paper screen that they can complete themselves and then hand to their provider for more efficiency.

But the flow is generally the same as you described.

Kendal Williams: So once you do a PHQ-9, it'll give you a score that is correlated to levels of depression, mild, moderate, and severe. So I'm going to ask Eleanor, is the patient diagnosed with depression based on that scoring?

Dr. Anderson: No, it's a screening instrument. So it's meant to really pick up the likelihood that someone may have depression. But it does roughly map onto severity of depression. And so we end up using it as a proxy for depressive symptomatology. But it should always just prompt a more thorough evaluation by the clinician.

Kendal Williams: And so let's get into that more thorough evaluation that the two of you would do.

So let's say you have a PHQ-9 score that, let's say for instance, it's in the moderate range or even the severe range.

What are you doing from that point forward? So, Eleanor, let me ask you.

Dr. Anderson: So I would start off the conversation by just mentioning there's high scores and would ask if it would be all right to talk about it. And I do that first because I think, for many patients, mental health is still stigmatized and they may have felt comfortable writing it on a piece of paper, but it may be scary to talk about it out loud. And so I say, "Hey, could we talk about the scores you put down." And almost all the time, they'll say, "Yes, of course." And they really do want to talk about it on some level. And then I'll say, "What's been going on? How has your mood been?" And use an open-ended question like that to really encourage them to talk about it.

And I know it can be difficult in a primary care office where time is tight and you can't necessarily start a therapy session per se. But even just to say, you know, what's been going on and give them a minute or two to just sort of get it off their chest what's been going on, I think that will help at least establish whether they have depression that's distressing to them or impairing their function in some way that would make it appropriate to talk to them about getting more care.

Kendal Williams: And I imagine having gone through this myself, you're often, as you're hearing what's going on, starting to build a plan if you will to help address it, right? So you're pulling out features that you think that can be addressed by you. So I know when I'm doing it, I'm listening for signs of anxiety. I'm interested in how they're sleeping. I'm interested in sort of things that maybe I can target with therapeutic tools.

Is that how you do it as well?

Dr. Anderson: Yeah. And I think I really try to start from a place of what does this mean to the patient and is it coming from something in their life? Did they chalk it up to a certain stressor, like the death of a loved one or loss of a job, and what's making it worse?

So the classic medical precipitating factors, prolonging factors, almost like you're addressing pain, you know, what is exacerbating or alleviating it?

And one big one that I always want to ask about with depression is substance use, especially alcohol use, but any substances, but alcohol use in particular can really make depressive symptoms worse and can greatly increase the risk of suicide. So that's certainly one modifiable risk factor that I would want to address if I'm asking about depression.

Kendal Williams: You know, when I was learning about depression as a medical student, as a resident, I often thought of depression as feeling the symptomatology of depression without a specific cause, not you lost a loved one or some other specific grief event. But I understand that distinction's been put to the side.

Dr. Anderson: Depression can really be thought of as a physiological state, almost like a fever. It's a very complex interplay of neurotransmitters in the brain.

There can even be structural differences after years of prolonged depression between patients with severe depression and patients without. And it can help explain why a depressive episode started right then, but it doesn't mean that it isn't clinical and deserves treatment.

And I liken it to someone who breaks their leg in a skiing accident. The doctor doesn't say, "Oh, I see why you broke your leg. It's understandable. You were in a skiing accident. Go on your way." They treat the leg. And so even if we understand why somebody may have started to have a depressive episode, if it meets criteria otherwise, that it's really gone on for more than two weeks, they're feeling depressed and sad most of the time or they're having anhedonia most of the time, then they still need treatment.

Kendal Williams: Joe, are there other things in the history that you're thinking as a primary care physician that you want to pull out?

Dr. Teel: Yeah, definitely within the screening survey we mentioned before, one thing I think many of us will always ensure we cover is suicidality or homicidality or any significant risk concerns that may be coming up. And so while, as you both mentioned, you know, making sure I can touch on the different symptomatologies and already thinking about how I may want to handle that, I certainly want to make sure that I'm addressing any risk to life for the patient or others around them and creating, if I need to, a safety plan and this is also really where, as I've mentioned, having an integrated model can be of utmost value.

Because if I were to identify the concerns such as that in the past, I would really have very little resources to none to be able to manage that myself or it would be just beyond me to figure out what to do and that usually would involve some sort of possible engagement with emergency services.

And now, I have the ability to pull in a behavioral health professional, either via phone if necessary for some practices or in person to really be able to address a crisis situation that may be sitting in front of me at that time.

Kendal Williams: Yeah. Having those resources available is absolutely critical. I agree with what you said, Joe. I practiced without Penn Integrated Care and now coming back into primary care, I entered a practice that had the PIC program and it's just been one of the best things that I've seen in my new experience as a primary care doc. It's terrific.

I want to pull in anxiety here as well, because oftentimes we are dealing with anxiety in primary care as an element of depressive symptomatology, but also sometimes it's separate. Eleanor, how do you make that distinction?

Dr. Anderson: It's interesting. Many people think of anxiety and depression as along a spectrum, that they aren't necessarily distinct clinical entities. And so there's been talk of having an anxious subtype for a major depressive episode. I think they can be similar.

You know, Conan O'Brien recently talked about how he never thought he was depressed, he only thought he was anxious. And then someone pointed out, you know, that can be a form of depression and he started to see other things that were like, "Oh, I actually wasn't admitting to myself that I also have these other features of depression." So it could also be even how the patient thinks of it. It maybe for whatever reason more acceptable to have anxiety versus depression. And so that's what they're focusing on and that's what they're reporting.

So there's multiple layers to this phenomenon. But there are different sort of subtypes of depression. And I find that anxiety makes for a very severe uncomfortable and really tormenting form of depression and really requires swift treatment.

And I have to be honest that I'm always a little bit more worried about suicidality in my anxious patients with depression, because anxiety is acute agitation or sort of inability to sit still, that sort of sense that a patient can't get peace, that's a risk factor for suicidality.

Kendal Williams: So let's pull this back into the sort of the therapeutic approach and talk about both the non-pharmacologic and then the pharmacologic approaches to managing this.

Generally, the treatment for psychiatric disorders in primary care or mental health disorders in primary care is divided into the pharmacologic and non-pharmacologic approaches. Psychotherapy is the most prominent non-pharmacologic approach. But, you know, to be honest, as a primary care physician, I don't know that much about it.

So, Eleanor, maybe you can give us a little insight into what psychotherapy is.

Dr. Anderson: That's a great question. And I think psychotherapy is ultimately the ability to heal through talking and listening.

And what great therapists have found is that it doesn't even matter necessarily what specific modality you're using with a patient. It's about what they've called in the literature common factors. There's factors that are common in all therapies. And it mostly requires empathy on the part of the therapist, a quiet listening environment. And I think the ability to encourage the patient to explore their past and how it's affecting their present symptoms.

So it's really learning where the therapist is a teacher and the patient is a learner. Although again, great therapists also manage to learn from their patients and the patients see that they're teaching their therapists. But the patient is there to really learn about themselves and to try on different ways of being in the world.

Kendal Williams: We hear a lot about cognitive behavioral therapy and how effective it is. My understanding is it's a form of psychotherapy. Can you help us understand the difference between that and other forms or what it is specifically?

Dr. Anderson: Sure. So cognitive behavioral therapy was pioneered at Penn by Aaron Beck, Tim Beck. And it really focuses on the fact that there's a triangle relationship between thoughts, feelings, and behaviors, and to really help see how each one affects the other. So you walk through the patient with the patient what they were feeling and what sort of hot thought was associated with that feeling.

If a man is feeling very angry and he is trained through therapy to identify what are all those thoughts going through his mind as he's walking angrily along and can identify what they call the hot thought, which is the thought that really feels like it encapsulates why he's so angry. And then also to notice what behaviors he's engaging in because of his anger. So that could be maybe leaving his house, maybe going to buy alcohol, maybe spending too much time at work and so forth.

And so you can really see the connections between these. And once you start doing it, it becomes even more easy over time to notice this and that works for depression, for anxiety and so forth.

And one difference between CBT and other therapies is the inclusion of homework assignments. So actually other therapies do make this connection between thoughts and feelings and behaviors. It tends not to be as explicit, and it's not as structured as in CBT.

In CBT, you get homework from your therapist and you take it home, you practice exercises, and it really strengthens your understanding of those relationships and you watch for episodes in your own life to draw those conclusions. And that really frees you up to, as I said, try on different ways of being.

Another feature of it is when you get to those thoughts, you evaluate how realistic they are. So you can say, okay, you identified the hot thought, like "Nobody loves me." "Well, really how realistic is that? Well, I guess my parents still love me. That's one thing. They sent me a card. They want to talk to me. It turns out my spouse does want to be with me. She's calling me to say, 'When am I coming home?' So I guess, that's not accurate."

So they can really take a thought that's causing them a lot of distress and can unpack that and assess maybe some of it's not so realistic, maybe they're applying a cognitive distortion to it.

Kendal Williams: So before we talk about antidepressant medications, my understanding is psychotherapy and cognitive behavioral therapy are at least as effective as any medication in managing depression. Is that accurate?

Dr. Anderson: Yeah. It is accurate. It is user-dependent though. You have to have a really good CBT therapist. But it can certainly be as effective as medication in the right patient and it lasts longer with no side effects. So it's a very effective tool we have for depression.

That being said, because it's so measurement-based and so structured, it is easier to study than the other psychotherapies. So other therapies like psychodynamic therapy are also effective. They may be a bit harder to study in traditional models though.

Kendal Williams: I think one of the things that's an aspect of the PIC program is that because I'm more aware of what that process the patient is going through, I feel like as a primary care physician, I'm much less of a pill pusher, if you will. Because before in the other models, the only thing I had to give was one of these medications we're about to talk about. But now, that I'm a little bit more integrated myself into the processes that the patient is experiencing, I feel like I'm part of a more holistic approach.

Let's talk about antidepressant medications and, in any model of care, it is often the primary care physician that's called on the one to prescribe the drugs in the treatment of depression. It's very important for primary care docs to have a good understanding of them. I want to admit upfront some of my own biases when I trained in a time when Prozac was just coming out and the other SSRIs, and there was a lot of popular press. I don't know if you remember this, this was probably about the '90s or so. It felt like a fad. It immediately biased me against these drugs because I couldn't see how it would be beneficial to treat masses of people with the neurochemical when they were dealing with everyday life stresses.

You know, this is the time when antidepressants went from something just psychiatrists would prescribe to really being something that became very common in the public. So there was a lot of skepticism. And then of course, there was this period where there's a controversy caused by a systematic review that suggested that while antidepressants had value in treating depression, it was really no better than placebo.

Now, more recently, there's been more work on this. And I found a large network meta-analysis that's often referred to. Network meta-analyses are very large, looks at data, very comprehensive looks at data. They're essentially meta-analysis of meta-analyses.

And in that view, it was actually fairly impressive that demonstrated that all antidepressants, virtually all of the ones that we use commonly, there were 21 antidepressants highlighted, do better than placebo. All of them do with odd ratios that range from 1.5 to 2. So that's led me to be more comfortable with these medications.

For reference to the study, I'm referring to that that came out in 2018 in the American Journal of Psychiatry, the first author is A. Cirpriani, and that's been referenced quite a bit. Were you familiar with that work, Eleanor?

Dr. Anderson: Yeah. And I'm glad you brought it up because I think your initial doubts and subsequent reactions has been the experience of many physicians. I have my own thoughts about that, that any new medication class, there's going to be some early adopters and so forth.

But I think within five years in the psychiatry community, it was pretty clear that they were safer than the old antidepressants. Because you have to remember before they came along, all we had was tricyclic antidepressants, which of course had lots of side effects and were much more lethal in overdose because of Torsades. And SSRIs were just easier to use.

And then there was then kind of more of a subsequent worry about, well, how beneficial are these. They're getting prescribed all over the place. And so I think this was reassuring that it really does provide benefit.

However, the data also showed that it's really not as beneficial for patients with mild depression. For patients with mild depression, they really should start with non-pharmacological approaches.

So I would say psychotherapy, exercise, sleep hygiene, and so forth. But for moderate to severe depression, antidepressants have a real effect and they can be life-saving.

Kendal Williams: Joe, I'm interested in your thoughts on this generally. I mean, I think you've probably practiced in the same period where a lot of these antidepressants have come out and now are used more frequently. What are your thoughts?

Dr. Teel: Not to make you feel bad in any way, Kendall, probably that I'm a little younger, so I certainly remember the release of Prozac on the market, but was not yet old enough to take it seriously in terms of my career path.

But just more from a general life understanding as a young person. Certainly by the time I was coming into practice through residency and med school, I think there definitely was a little bit of controversy, but the mainstay of treatment really had switched to SSRIs at that point. I think it was basically the answer for every question on your boards. If it had anything to do with depression, the answer was always an SSRI at that point.

And so it really has been the backbone for management through most of my professional career.

Kendal Williams: So let's dig into the SSRIs then. We're all familiar with the drugs, Prozac, Zoloft, Paxil, Celexa, Lexapro and others. Those are probably the more common ones. They're the most commonly prescribed medications for depression. Each have their strengths and weaknesses.

And in this, I'm just going to ask you guys to sort of download, if you will, almost re-associate your feelings about these drugs generally and some of your thoughts on how you use them in specific instances.

Eleanor, is there sort of a go-to for you?

Dr. Anderson: You know, I'm going to admit it's my bias and you can talk to different psychiatrists, they may have different biases. I'm used to treating patients with medical illness. And so I usually go with Lexapro, escitalopram, because it has lower incidence of side effects, fewer drug interactions. And I've seen some good benefit from patients with that.

Dr. Teel: It's interesting that we all develop our sort of initial go-to's, and I think I definitely prescribe escitalopram. But I think it's my bias in some ways.

So I think, it's really influenced by the fact that while I do primary care, I also practice obstetrics and care for prenatal patients. And so I think if you looked at all of our prenatal patients and therefore it sort of influences my choice in general, it definitely shifted towards sertraline, which has a lot of great long-term data in pregnancy.

And so I think it is interesting how we all sort of get grounded in certain medications. I appreciate your input, Eleanor, on how clean Lexapro can be.

Dr. Anderson: No, absolutely. But sertraline is a wonderful medication also. The nice thing about the whole class of SSRIs is that there's not necessarily a wrong option. I think paroxetine is actually my least favorite because it has more side effects. It's got more anticholinergic action, some weight gain in people and the discontinuation can be severe for Paxil. So it's got a short half-life and so patients get very uncomfortable if they miss a dose even.

So that one I tend to avoid, but the other ones I think you could almost always make the case for any one of them. If one doesn't work, you could try another one.

I really like sertraline also, and I think it's important to note that in young women of childbearing age, many pregnancies are unplanned. So whether they're prenatal or not, it might be wise to do sertraline. Not that that's the only safe SSRI for them, but it just has the most data.

Kendal Williams: When I was learning about these, Prozac was the one that was sort of more activating for depression, Paxil was the one for anxiety and sertraline or Zoloft was in between. And so you ended up using a lot of sertraline because it was sort of considered to be the nice happy medium between those as well. I don't know if that's played out in real life.

Dr. Anderson: Well, that's interesting. I will sometimes use fluoxetine for anxiety and one reason is because you can push it to pretty high doses to handle really severe anxiety disorders like obsessive compulsive disorder or just severe generalized anxiety disorder.

I like to be able to say, "Okay, you've gotten some response. I'm going to push it even higher," because the data show that a SSRIs need to be at a higher dose for a longer duration to really have good benefit in anxiety, so you might need to give it longer and push it a little harder for anxiety disorders versus depression.

Kendal Williams: So, let me ask you about that now, just about dosing. Let's say you're starting sertraline for someone. How do you start it?

Dr. Anderson: : Well, I'll typically do 25 milligrams a day for about a week and then go up to 50. So I might prescribe the 50 and say cut it in half for six days, something like that, and then go up to 50 milligrams.

But I do notice a little bit more nausea with sertraline than the other SSRIs. I'm not actually sure whether that's more borne out in the literature, but that's just been my anecdotal experience.

Kendal Williams: So, might you stay at 50 if you're treating depression and someone's feeling better?

Dr. Anderson: Usually, I'll check in every four to six weeks or so. And see how they're doing if they're having a benefit and try to get them, I would say more to 100 milligrams.

I think 50 is more to kind of give them a chance to get used to the medication and some patients do respond to that dose. So I don't want to necessarily push it much higher if they're getting a good response. Also some patients don't want to feel that they're in the presence of a pill pusher and so forth, especially if it's the first medication I'm trying, I just sort of take it slow and I kind of gradually go up every four to six weeks until I'm at 100 milligrams, which could be at the very next visit. You know, go up to 75 for a week and then up to a hundred.

And then if there's really no response at all after that, I might try a different medication at that point. But if there's a partial response that would lead me to keep going and boost the dose up further.

Kendal Williams: And you do a similar approach if you were using Lexapro?

Dr. Anderson: Yes. But as I said, Lexapro doesn't really have as many dosage changes. It's just got a narrower range. So, I would start it at five milligrams and go to 10.

And again, this starting at a lower dose for the first week or so is really just to minimize side effects, headache, nausea, and so forth that might occur in the first week. And then I would just go up to 20, but above 20, there's really very much diminishing returns. And so I know some psychopharmacologists do go up to 30 of Lexapro. But I haven't really found it helpful. So at that point, I just switch if a patient's not getting benefit.

Kendal Williams: So the second class that are used for antidepression are the SNRIs, venlafaxine or Effexor, desvenlafaxine known as Pristiq, and duloxetine, which is Cymbalta. Now, those are the most common. What's your experience with this clause?

Dr. Anderson: They can be really effective. And I think certainly if someone's tried a couple of SSRIs, an SNRI would be very reasonable. I like to use them if there's a chronic pain element, especially duloxetine has, you know, an indication for fibromyalgia and chemically they're similar to the tricyclics, which we also shouldn't forget about as an antidepressant class.

So I think that's why they may have an analgesic effect also.

The thing to watch with venlafaxine though, as with paroxetine, it can have a really severe discontinuation syndrome if someone stops it too abruptly or misses doses. And that's manifested by flu-like symptoms, malaise, fatigue and the dreaded brains zaps, which is that electric shock feeling on the scalp.

So that's the thing to watch out for venlafaxine. Sometimes it can be associated with sweating. Ironically in patients with hot flashes, like maybe they're menopausal or they're taking tamoxifen or Lupron, then a medication like Effexor and Paxil can help those hot flashes. I'm not sure why that is, but that does seem to be born out clinically.

Kendal Williams: You know, it's interesting in that systematic review, that meta-analysis of meta-analyses I've referenced earlier, the one with the highest odds ratio is amitriptyline, which surprised me. It seemed to be the most effective.

So it's interesting that these drugs have some of that same benefits that tryciclics have. I don't recall off the top of my head whether this drug class was in that meta-analysis, but that's an interesting observation.

Dr. Anderson: : Yeah. And the problem with amitriptyline is the side effects, I think. I mean, I usually go with nortriptyline. If I have someone with migraines, for example, and I'm really trying to achieve two goals at once to do the depression and chronic pain.

And the nice thing about nortriptyline is that it has much lower side effects than amitriptyline. There's no like metabolites, active metabolites. So, if I'm not getting a response from nortriptyline, and they're at a pretty good dose and there's really no response, then I could get a level and see if they're an ultra-metabolizer and I can sort of trust that level, and the dose is the lab level.

So if I'm at 50 milligrams of nortriptyline, I should expect it to be about 50 nanograms.

Of the tricyclics, which I don't have too much experience with, because again, we tend to go with these other medications first, but nortriptyline is my favorite to go with. And, I'd be happy to have more of my treatment-resistant depression experts on if you want to dive deep into trycyclics.

Dr. Teel: I think Eleanor, it is a good comment earlier, the one you said about the dual effect. And I think, especially if we go into a little tangent around chronic pain, especially, as we're, you know, in primary care obviously avoiding opioid analgesia when not considered appropriate and for non-cancer pain.

Your comment about duloxetine or Cymbalta as having a role there. And I do find myself, for many years, turning to tricyclics, both amitriptyline and nortriptyline, looking for that either dual effect of emotional impact as well as pain and especially if there's insomnia overlying all of that. I think the TCAs and the SNRIs are a nice combo.

Kendal Williams: That's terrific information. My favorite antidepressant, we're about to talk about, and that is Welbutrin or buproprion. And that falls into a class of atypical antidepressants. They don't fit in the other two categories.

There are two new ones that we're not going to talk about. They're still branded and expensive. But if Eleanor thinks they have a value, I'll have you mentioned them.

But, otherwise, I'm going to focus on Welbutrin otherwise known as buproprion and mirtazepine.

Welbutrin in particular seems to be a highly useful drug that has some advantages over the SSRI and even the SNRI classes in the right patient. What do you think about that?

Dr. Anderson: I think it's a wonderful medication. I think it really does seem to boost a patient's hedonic capacity. So I do tend to reach for that if the patient is just down and blue and nothing seems to be able to reach them the way their previously loved activities used to. You know, they just can't enjoy things the way they used to, then I'll certainly reach for buproprion first actually, sometimes even before an SSRI.

And it's got the indication for smoking cessation and it's a dopaminergic lightly, so it's a little bit stimulant-like in its effect. So if someone's like, "Maybe I have some mild ADHD, I'm not sure," you know, sometimes that can actually help them feel better because dopamine is really a salient neurotransmitter. It can help you focus and kind of say, "Hey, this is important. This food and whatever."

Things are supposed to give us pleasure. The bupropion can really restore that capacity.

Kendal Williams: It works a little more quickly than the SSRIs as well, right? And I think that sometimes can be helpful?

Dr. Anderson: Yeah. And the same with the mirtazepine. I would say the mirtazepine, I would expect more like two to three or three to four weeks as opposed to five to six weeks for the SSRIs.

One thing I would caution is that buproprion, I always ask my patients whether they have a seizure disorder, make sure they don't have a history of brain tumor and so forth, just because the lore is that buproprion is associated with lower seizure threshold and that's probably on the boards as well.

Some studies have indicated that there's no increase with buproprion that's more than the antidepressants as a class. So antidepressants as a class can lower the seizure threshold and so can depression itself lower the seizure threshold. So some are saying, "Hey, buproprion is getting a bad rep. We should just prescribe it if it's the right medication psychiatrically." But a number of my psychiatry colleagues have had experience where they've tried buproprion and a previously healthy person has a seizure. And I think that's been enough to scare a lot of us away from it.

So that's my only caution. Obviously, any active eating disorder or alcoholism, I might avoid bupropion also for the same reason.

Dr. Teel: Yeah. I don't know, Kendall, you know, as your favorite medication, I usually will also turn to this and we didn't really talk about, you know, in terms of side effects, but certainly some of the sexual side effects that we see with the SSRIs seem to be a little less common or augmented with buproprion.

So certainly I think it's a pretty common thing I'll mention. And we talk to patients about if they're having difficulty with sexual dysfunction and trying to think of alternatives perhaps if necessary for their SSRI.

Dr. Anderson: That's an excellent point. And I recommend the same if somebody has sexual side effects from a previous medication is looking to avoid that.

Kendal Williams: The role of mirtazepine seems to be, you know, with the patient that's losing weight, not sleeping, maybe more anxious. I don't know, Eleanor, if that's true, because its side effects are that it causes weight gain and it causes you to be sleepy, So that seems to be its role in this pharmaceutical universe. Is that right?

Dr. Anderson: Yeah, that's roughly right. I think the weight gain thing is a bit overblown. It's really only in like 10 percent of patients that actually have weight gain. So I've seen, when I was in the cancer center as their psychiatrist, I would see people losing weight and get put on mirtazepine in the hopes that that would help them gain weight.

I don't think it's really that powerful. For a subset of patients, absolutely, it does have weight gain with it. But yeah, I think just clinically for its effect, I definitely use it more in my anxious patients. It is effective for depression by itself, but if there is that anxious quality to the depression, mirtazepine is really good.

And the other thing to note about it is that it's got anti-nausea effects also. So we use it a lot of the time in the hospital as well. If someone's really miserable, they're anxious, they can't sleep and they're not touching their food because they're so nauseated, then mirtazepine can really help them.

Kendal Williams: I've just started to use that drug and had a patient that has been finding it very effective. So that's been new for me.

You mentioned this before, Eleanor. Let's say you start one of these medications. but after eight weeks, I don't know how long you give it, I'll have you fill that in, it's not working. And now, you have to either switch to another one or add on. I know that there often is a situation where Welbutrin is added to an SSRI. But we all faced a situation where whatever we've tried hasn't worked.

What do you do then?

Dr. Anderson: The question is sort of, do you keep increasing the dose or do you do an adjunctive medication? And I try to keep it simple where possible. I think if someone again has had a partial response, then I'll increase the dose and try to max that out. And if they're on the max dose and they're almost there, but they still got some residual effects, that's when I would reach for an adjunctive medication. And Welbutrin has evidence for being an effective adjunctive medication.

So let's say I go up to Lexapro 20 and they're feeling better, but they're still having some anhedonia, then I would add Welbutrin. I usually use the extended-release version and I'll just start at, you know, 150 in the morning.

Kendal Williams: What if they're not feeling better? What if you're eight weeks into it, and they noticed no improvement?

Dr. Anderson: I would say probably at that point, I would max out the Wellbutrin just to make sure that we've got these two big guns on And let's really give each of them a shot to work. And if that's not working, then I may go to a different SSRI if they haven't tried it or I would switch to an SNRI.

Dr. Teel: I'm actually curious because I've seen this come up in primary care where say someone is presenting to you for the first visit or even, you know, they've been in your care, but you're addressing their depression at that visit. And they say I've been on this and I've been on that.

And I'm always curious and I don't know what your thoughts are of like how many SSRIs do you try before you just say, "You know what? We're not going to try this again. The class doesn't seem to be effective for this person."

Dr. Anderson: Well, I never really throw out the class. I think of it as members of a family. You know, you may get along with one member of a family, but her sister is not really your type. So it may just be that they haven't really found the right medication in the class.

But generally, I would say after two SSRIs, it's time to try a different class. And honestly, if someone's really having several bouts of depression and several trials of medication that aren't working, I would say it's time to really get help from a psychiatrist colleague. That's when it starts to edge into the more treatment-resistant or treatment-refractory depression, and it might be time to even explore other newer medications or non-pharmacological modalities such as transcranial magnetic stimulation and so forth.

Kendal Williams: So let's say we've been successful and the patient is on one or maybe even two medications. They several months later are feeling much better. How long do you leave them on the medications before you think about tapering off?

Dr. Anderson: To me, that depends on whether they've had previous episodes of depression.

If they've had two or more episodes of depression, I keep them on because the evidence shows that they will face a high risk of relapse if they're off medications. And Andrew Solomon wrote about this in the Noonday Demon, that he thinks of it as just life-sustaining medication like insulin. He has to take it. And it's not any marker of strength to get off of it.

So it really depends on their risk profile and how much they've had depression in the past.

And if it's their first episode though and they've done well right off the bat and they're continuing to do well, I would say after six months to a year, if they'd like to try taking down the dose and monitoring, that I would support them in that.
I just would try to give them some anticipatory guidance about when will you know that it's time to come back and when will you know that it's time to maybe go back on the medication again. And they're usually able to say, "Okay. Well, when I stop showering" or "When I stop watching football" or something.

They have something when they look back at their depressive episode, they've got a behavior or two that really clued them or their family in to the fact that they really weren't themselves.

So I'd say if you're considering tapering somebody off, have an agreement ahead of time what is going to trigger a return.

Kendal Williams: Do you warn patients as you taper them, that they may feel like the withdrawal itself is a return of the symptoms?

Dr. Anderson: : Ideally, you're tapering them in such a way that they don't feel discontinuation syndrome or withdrawal symptoms. So that's certainly true for anxiety. If you've got someone on an anxiety medication that's effective and the tapering is causing more anxiety, then I certainly warn them about that. If it's depression, I think that tends to be a bit more insidious and they tend not to feel suddenly more depressed from withdrawal symptoms.

Kendal Williams: How long do you taper them? Is it two to four weeks? That's what I saw in the literature.

Dr. Anderson: Yeah, I think that's about right. It also depends on kind of what it means to the patient. Often, they're not willing to go that fast. And maybe more of my anxious patients want to hang on to it a little bit longer. So I'll even say, "Okay, like how about two months?" And there's no rush really.

So if they don't care, then, yeah, I would say two to four weeks.

Kendal Williams: So finally, I want to discuss a common scenario I experience. I'm new to the practice I'm in, and now I'm inheriting some patients from other colleagues who have left. And it's actually very common for me to have a patient who was started on an SSRI for an episode of depression from which they've evidently fully recovered. And now, they're still on the SSRI.

I have this discussion with them. It often is the case that it's been just that one episode that led them to be on it. And they ask me, "What's the problem staying on it?" And, you know, studies have highlighted the increased risk of upper GI bleeding, osteoporosis, some hyponatremia in the elderly. I am usually encouraging them to taper off if they no longer feel they need it.

And Eleanor, you highlighted that it's really for two or more episodes of depression that we would keep somebody on more lifelong therapy, right?

Dr. Anderson: Right. But I agree, especially as someone gets into the older age range, that I'm more cautious about the SSRI in general.

They did a study in Pennsylvania of prescriptions of SSRIs to the elderly and the mean PHQ-9 for starting an SSRI was 6, somewhere around 6. So I think overprescribing certainly can be an issue.

And as someone gets older, I think, the risk of hyponatremia, especially in women, becomes much more of a concern, I think, for me. But I certainly think if they just had the one episode and they're doing well and they'd like to come off and you think it would make sense, and certainly if they've got other medications and you're really trying to avoid polypharmacy, it certainly makes sense to try off of it.

But I would recommend relatively frequent follow up just to make sure that you're not missing a recurrence.

Dr. Teel: Eleanor, I appreciate the Pennsylvania data first, just about the fact that people were starting SSRIs and antidepressants in older individuals with fairly low PHQ-9 scores.

I think back to a point Kendall made earlier in the conversation about now that he's in a practice with a PIC program and a collaborative care model, that he doesn't feel like the only answer he has to give someone is medication and that he feels more comfortable having a holistic approach. I'm curious if anybody's ever looked at that, that if you can institute collaborative care models, if almost in some ways our prescribing goes down or becomes "more appropriate".

And so it's interesting with sort of wrapping up from the beginning of our conversation, how you know, this new model of care may be able to even provide more appropriate care to our populations.

Dr. Anderson: There is evidence that collaborative care decreases unnecessary prescribing, and I think it really goes to show that when you have more options to treat depression, especially on the more mild side of things, then you can avoid prescribing medications.

Kendal Williams: So, you know, we usually conclude by asking each of you if there's anything more you'd like to say to the primary care audience at Penn, insights you've gained.

Dr. Anderson: Just the question "Are you depressed?" Is very sensitive and specific for depression. So don't be worried that you need anything special to start the conversation with your patient.

I would also say that if you're going to start an SSRI, I would ask about any kind of history of hypomania or mania to rule out bipolar disorder, because the index episode of bipolar disorder can be depressive. So you may miss it if you don't ask.

And I would ask about family history in that case, as well as age of onset as younger age of onset is more likely to be bipolar disorder.

And then, as always, I would give patients anticipatory guidance about possible suicidality. They're going to see that black box of the suicidality warning on the medication. And so I tell them that it was for a very specific age group and that, while they did find an increase in suicidal ideation, it wasn't an increase in suicidal behavior per se as prescriptions for antidepressants have gone down then suicides have gone up.

So, ultimately, it's a complex issue. Depression is a multifaceted condition, and it really usually requires a multi-pronged approach.

Dr. Teel: And I think my only conclusion is I do hope just across our system, and obviously across other systems, that this model of care and having an integrated approach can hopefully make our colleagues in primary care from physicians to NPS and PAs and others just feel more empowered to be able to manage their patients, and not feel like they, through lack of confidence or lack of support or backup, have to hand everything off to a specialist, that we can really keep the medical home as broad as possible.

Kendal Williams: And on behalf of the primary care community, I want to thank you both for the work you've done with this program.

If you are in a practice that is a part of it, you know its benefits and can appreciate their work. And if you have not yet experienced it, I hope you soon will.

With that, I want to thank everyone for joining the Penn Primary Care Podcast. And please join us again soon.

Disclaimer: Please note that this podcast is for educational purposes. Only for specific questions, please contact your physician. And if an emergency, please call 911 or go to the nearest emergency department