Editor’s note: The issue of emergent suicidal thinking has weighed on us since the 2004 black box warning—amplified by the re-analysis of the now infamous study 329, which found the emergence of previously under-reported suicidal thinking. How do we offer hope for effective use of medication while balancing the potential risks? Dr. Danella Hafeman has made a close study of it all and sheds light on the problem.

CCPR: Can you tell me about your current work?
Dr. Hafeman: I’m a child psychiatrist at the Child and Adolescent Bipolar Spectrum (CABS) clinic at the University of Pittsburgh. While we are primarily a clinic for bipolar disorder, we see a lot of kids with unipolar mood disorders, kids who have seen many providers. I am also a researcher, and my studies focus on children and adolescents at risk for bipolar disorder.

CCPR: What is the current state of knowledge about antidepressants in children? What is the research telling you?
Dr. Hafeman: I think our shared clinical experience is that these medications work for many children and adolescents. So, it was surprising that a recent meta-analysis by Cipriani and colleagues found that antidepressants did not generally fare better than placebo (Cipriani A et al, Lancet 2016;388(10047):881–890). Fluoxetine did a bit better, but not much. This was concerning, but it doesn’t tell the whole story. Another researcher, John Walkup, explains that Cipriani’s study included multiple small industry-sponsored trials of kids who had more minor depressive episodes. Those are kids who tend to respond to whatever we do, and so it’s no surprise that the placebo rate was upwards of 50% (Walkup JT, Am J Psychiatry 2017;174(5):430–437).

CCPR: What can you tell us about the more severe cases that we tend to see?
Dr. Hafeman: The National Institutes of Health (NIMH)-funded Treatment of Adolescent Depression Study (TADS) showed fluoxetine was very helpful in about 60% of adolescents with more significant levels of depression (Silva MJ et al, JAMA 2004;292(7):807–820). The positive effect of treatment was even better for those who also had cognitive behavioral therapy (CBT). In that study, the placebo response rate was about 35%.

CCPR: Wow, what a difference. What else have you learned about antidepressants in younger children?
Dr. Hafeman: The TADS study only included children 12 and older. However, a prior randomized controlled study on fluoxetine with kids as young as 7 showed a positive effect with no differences in how the children or adolescents responded (Emslie GJ et al, Arch Gen Psychiatry 1997;54(11):1031–1037). So, despite the Cipriani meta-analysis, the rigorous NIMH-funded trials really do support using SSRIs. All the trials also show that the optimal treatment is SSRIs plus an evidence-based therapy, such as CBT or interpersonal therapy.

CCPR: How do you select the right SSRI for a child?
Dr. Hafeman: Fluoxetine has the most data. It was around when the large NIMH trials were happening. We don’t have large studies with other medications. Fluoxetine has a very long half-life, which can be helpful if kids or adolescents aren’t taking it every day. Also, kids metabolize medications more quickly, so some of the medications with a quicker half-life, such as paroxetine, can lead to withdrawal. On the other hand, sometimes you might actually want a medication with a shorter half-life if you think kids might have activation, sleep problems, or elevated mood, or if they have a family history of bipolar disorder. The shorter half-life means that, if they respond poorly, it won’t take a long time for those side effects to go away.

CCPR: Any other important factors in choosing medications?
Dr. Hafeman: One other thing to consider when choosing an SSRI is drug interactions. If the child has other medical conditions and is on other medications, you might consider something such as escitalopram, because there aren’t as many drug-drug interactions. But aside from avoiding paroxetine due to its propensity for withdrawal syndromes, there’s really not much evidence to choose one versus another SSRI. I also ask what other family members have responded to, and while studies have shown that it doesn’t matter substantially, parents who have responded to a specific medication might be more comfortable having their child use it.

CCPR: What about SNRIs?
Dr. Hafeman: There’s not as much evidence for SNRIs in adolescents, but the TORDIA trial reported no difference between another SSRI and an SNRI. SNRIs had more cardiac side effects, including elevations in blood pressure, so sticking with SSRIs, at least at first, makes sense (Brent D et al, JAMA 2008;299(8):901–913). You might consider an SNRI after the first 2 trials of SSRIs, though there are limited data for this.

CCPR: Can you help us distinguish between antidepressant-induced mania and behavioral activation?
Dr. Hafeman: These are pretty different adverse events. If children develop mania on SSRIs, the clinical consensus is that they are at much higher risk for bipolar disorder (Barbuti M et al, J Affect Disord 2017;219:187–192). In fact, if the SSRI-induced mania lasts after the SSRI is out of the system, then the patient meets criteria for bipolar disorder based on DSM-5. Depending on the situation, it probably makes sense to switch to a mood stabilizer at that point. And if you’re trying another SSRI, you have to go really slowly and be very careful. Activation manifests differently, with hyperactivity and bounciness. It tends to happen early and sometimes resolves a little bit. Some kids have been very anxious and subdued, and then you give them medication and they’re more outgoing. It’s more likely to happen for younger kids and in kids with a family history of bipolar disorder, but even that is unclear. Starting low and going up slowly helps avoid activation and other side effects. Activation can vary greatly among SSRIs, so I’ll consider switching.

CCPR: You spoke about faster metabolic rates in kids. Can you tell us more?
Dr. Hafeman: You also see that with other medications: for example, with stimulants. I can’t tell you the number of kids that we see that definitely have depression and have been on perhaps 3 different antidepressants—like Prozac 20 mg, Zoloft 25 mg, and Celexa 10 mg. Going up to higher doses makes sense, but only if they are still symptomatic and not having side effects.

CCPR: What about bupropion?
Dr. Hafeman: Bupropion doesn’t have as much evidence for treatment of depression in kids. It can be helpful for kids who have sub-threshold depression and difficulties with attention. Anecdotally, when kids have sort of a more sluggish depression, a lot of them will feel better on bupropion. I think that’s the activating norepinephrine/dopaminergic effect, although not in the sort of activation we were talking about with the SSRIs.

CCPR: Any other antidepressants that you use? How about mirtazapine?
Dr. Hafeman: Mirtazapine can be helpful for sleep, appetite, and anxiety. The evidence for depression in kids is pretty minimal. I don’t use it first line, even if there are those other problems that you’re pretty convinced are related to depression. An SSRI is the place to start. But if there are other reasons to avoid SSRIs, mirtazapine can be a good choice. We use it a fair amount with medically sick kids, who have a lot of those problems along with a lot of anxiety.

CCPR: What are your thoughts on the black box warning about potential suicidality?
Dr. Hafeman: An FDA meta-analysis found slightly higher rates of suicidal ideation and attempts as compared to placebo groups (Friedman RA, NEJM 2014;371(18):1666–1668). There weren’t actually any completed suicides. These were short-term studies, and perhaps the finding was due to side effects, such as behavioral activation. After the black box warning, not only did the rates of prescriptions dramatically go down for SSRIs, but the rates of diagnosis of depression went down too. Suicide rates as a whole went sharply up. Even so, the possibility of suicidal ideation arising from use of the medication, for whatever reason, is a serious concern.

CCPR: So, what do we tell families?
Dr. Hafeman: I tell my patients and families that antidepressants treat depression, and that depression is the biggest cause of suicide. But it’s also really important for families to have a safety plan to be aware of potential suicidality and closely monitor.

CCPR: What does good follow-up look like, and how do we overcome the barriers?
Dr. Hafeman: Ideally, we would see people once a week after starting an SSRI. That is difficult for everyone: patient, family, and provider. I try to see them within a month, and if they’re higher risk, I see them once a week or have them talk with the nurse. It’s important to have the door open for people to call with any emergent side effects or anything that might come up.

CCPR: Can you talk more about your team approach to treating children with depression?
Dr. Hafeman: In our clinic, usually the doctor and the nurse see patients together. It helps us get a good diagnosis*, comorbidities, and the family situation in the very first step. Then you have follow-up to look at side effects and monitoring. But the smarter follow-up is an ongoing assessment of this child and the family—the longitudinal assessment. We are also in touch with the therapist, who has a different picture of the child, and the school. Those conversations make my visits with the patient go faster, because I know what’s going on. The time it takes to communicate with all those people is difficult, but it’s important for good care. *(Ed note: Some medical disorders also drive symptoms—see table below on endocrine disorders that can mimic depression.)

CCPR: What is your path for deciding whether to increase, change, or augment a medication?
Dr. Hafeman: Make sure the patient is on a good dose of the medication before calling it a failed trial. That takes time, and families want their kid to be better right now, which is understandable. Sometimes, kids with anxiety will need even higher doses.

CCPR: What if you just feel the medication is not working?
Dr. Hafeman: Revisit the diagnosis and make sure that it really is a major depression as opposed to ADHD or substance use or mania. If it’s the first trial of an SSRI, I will switch to another SSRI. That’s what’s supported by the TORDIA study. Revisit therapy—more kids will respond when something such as CBT is added. Avoid multiple medications, with exceptions. For example, I’ve had children who don’t want to get off their SSRI because it’s helped their anxiety. Because it has a totally different mechanism, consider adding bupropion. There isn’t great evidence in kids on augmentation. Case studies show that quetiapine can be helpful (Pathak S et al, J of Child and Adol Psycho 2005;15(4):696–702). Synthroid, other atypicals, and even lithium have been used for treatment-resistant depression in kids. There is a risk-benefit analysis for each option.

CCPR: How do you talk with parents regarding their concerns about antidepressant medication?
Dr. Hafeman: I ask them, “What are your thoughts about medication?” If the parents want to start with an evidence-based therapy first and hold off on the medication—aside from severe circumstances—that is a reasonable approach and it is supported by research. The response rate isn’t as high for CBT, but it’s higher than placebo (Arnberg A et al, Behav Therapy 2014;43(4):275–288). Also, make sure not to confuse baseline problems with side effects. Kids get headaches, and if they attribute the next headache they get to the SSRI, then that will be problematic. I’ll always say, “Anything can cause stomachaches or headaches,” and I’ll always go over the black box warning. I’ll go over the possibility of activation. I tell families how they can get in touch with me, and I emphasize that I want them to be in touch with me as needed. You can talk about the pros and cons of medication, but it’s important to talk about how not medicating depression can be absolutely deadly. Medication can be life-saving, and it’s important to highlight that with families. Even when there is no suicidality, adolescence and pre-adolescence are extremely important times for a child’s learning, development, social interactions, for developing a sense of self. If you’re really depressed, you just cannot do these things. It’s very important to not underestimate the detrimental effects of depression.

CCPR: Thank you for your time, Dr. Hafeman.