In response to several recent alerts about antidepressants in children and adolescents (Table 1 [8]), physicians have been asking anew: How should we treat depressed kids? This question first arose when physicians in three countries were notified that paroxetine (Paxil) is now contraindicated as a new treatment for individuals younger than 18. A review of data found paroxetine to be both ineffective in child and adolescent depression and associated with a small increase in suicidal ideation and attempts, as well as hostility.[1-3] Within 2 months, this warning was followed by similar notifications about venlafaxine (Effexor).[4] Subsequently, the US Food and Drug Administration (FDA) issued a general alert regarding a possible increase in suicidality associated with all of the selective serotonin reuptake inhibitors (SSRIs) in children.[5] Next, physicians were notified that nefazodone (Serzone) was being removed from the market because of reports of hepatotoxicty.[6] Most recently, the British Medicines and Healthcare Products Regulatory Agency (MHRA) declared that based on a review of all evidence, published and unpublished, the risk-benefit ratio for antidepressant use in children is not acceptable for any antidepressant except fluoxetine (Prozac).[7] These developments have been followed by labeling changes for SSRIs that warn about suicidality, agitation, aggression, impulsivity, and other emotional and behavioral adverse effects.[8,9]

Physicians were caught off guard by these developments and many have been puzzled by these advisories. The reason for the surprise was that the studies with negative results remained unpublished and the impression from published studies and practice guidelines[10,11] was that antidepressants were probably effective in young people. However, reviews of published data[12-15] have concluded that evidence for efficacy is limited to symptom improvement with fluoxetine. The FDA’s recent reanalysis of the full data set, published and unpublished, concluded that only 3 of 15 trials assessing SSRI use for pediatric major depressive disorder (MDD) showed any benefit over placebo.[16] Meanwhile, a pattern of increasing prescribing rates in clinical practice[17] has been reinforced by the very high placebo response in pediatric depression. The recent alerts and prohibitions have led physicians to wonder how they should now be managing child and adolescent depression in the primary care setting.

Despite these discouraging developments, there is some good news on the horizon. As part of the implementation of the Child and Youth Mental Health Plan for British Columbia,[18] the Ministry of Children and Family Development (MCFD) has begun a planning process that will see evidence-based treatment for children in the province’s community mental health system, and support for primary care physicians and families in improving early intervention and treatment. This includes the provision of more access to other treatments, including cognitive behavioral therapy (CBT).

Realistically, of course, family physicians will still be the initial contact and primary resource for treating depressed teens. The prevalence of adolescent depression, 5% to 8% cross-sectionally with a 20% to 25% cumulative risk by the age of 18,[19] means that much of the care will be provided by those working in primary care settings. The lower prevalence and complexity of prepubertal depression makes it likely that specialized mental health teams and professionals will become involved more readily.

It should be noted that while the prevalence of clinical MDD is about 5% to 8%, the report of depressed mood, on specific questioning, occurs in almost half of young adolescents as the result of a variety of stressors and developmental factors. Thus, all “depressed mood” is not clinical depression. Furthermore, in teens, depressive episodes are relatively mild, appear to remit more quickly than in adults (50% within 2 months),[20] and are more responsive to placebo.[21] The approach described below takes this into account, but also recognizes that even brief episodes of depression are a risk factor for recurrent episodes, and these teens often have multiple risk factors for depression. Furthermore, depression leaves scars in the form of more negative attitudes, poorer coping skills, and generally poorer health behaviors.[22] Given the chronicity and recurrence of depression over a lifetime, intervention should focus on reducing risk and increasing resilience. Risk factors, detection, and assessment have been addressed in another recent article.[23]

Practical management

Primary care physicians should take a practical approach when supporting and treating teens who present with depressive symptoms (Table 2).

Be sensible

Don’t immediately assume that a teen who reports being depressed is clinically depressed. Recognize that depressed mood is common on any given day (about 40% of 14-year-olds will say they are unhappy if asked), and that quasi-depression (loneliness, social anxiety, perfectionism, feeling things deeply) is common as well. Determine the personal meaning of the word “depression,” which teens now readily use to describe their mood in response to daily stressors. One common problem that can mimic depression in teens is regular cannabis abuse or dependence. Cannabis abuse can produce lack of motivation and interest, impaired concentration, and irritable or depressed mood. Irritability may be exacerbated during withdrawal. Cocaine and amphetamine abuse, particularly the use of crystal methamphetamine, produce profound symptoms of depression. If teens are asked in a permissive way (“How often do you use?” rather than “Do you use?”) they are more likely to reveal the extent of use.

Be practical

Take advantage of research on the effectiveness of psychosocial treatments for depression,[24,25] while focusing on interventions that will improve functioning fairly quickly. This includes active problem-solving regarding real issues (school, communication with parents, relationships with peers, drug use). Much of this can be done with brief counseling sessions that support the teen in taking positive steps to enact change. For example, to emphasize the importance of good health habits, you might offer a “prescription” for sleep hygiene, balanced eating, and regular exercise using a prescription pad as a note pad. You might also explain that these “interventions” have an effect on brain and body physiology, and that exercise in particular can have a genuine effect on improving depression. Ideally, several problems can be addressed with one strategy, such as going for a brisk evening walk or swimming daily with a parent or a friend. The family physician should not operate in isolation: enlist parents or school counselors to enlarge the network of practical support. Also, consider some training in motivational interviewing and principles of cognitive behavioral therapy to augment your counseling skills.

Be there

Keep in mind that you are a powerful antidepressant. A relationship with a caring, interested adult is very good medicine for teens, and may contribute to the strong placebo effect seen in pharmacological studies. In primary care, given the limited funds available for counseling, the solution can be to set a series of brief appointments once or even twice a week. Lay out a plan for several appointments at the first visit to give the teen a future connection to you; this can help combat hopelessness. If a complete checkup has not been done recently, use one of these visits for a full physical examination to rule out contributory problems, and to touch on other health and risk behaviors. Set a small practical goal for each appointment. This approach fosters hope, which is an antidote to suicidal thinking. Place the current crisis in a developmental context, and focus on fostering the teen’s age-appropriate coping skills using a coaching and joint problem-solving model. Encourage engagement in positive family, school, and community activities (sports, recreation centre activities, ethnic or spiritual centre events) to increase self-confidence and social support.

A referral to local mental health resources for individual or group cognitive behavioral therapy would be ideal, but services may not be available. It is in this area of access that we hope to see great improvements in all British Columbia communities, as mental health clinicians are trained in CBT for child and adolescent disorders, a plan being implemented during the next year through MCFD.

Be aware

Medication may be indicated, but requires close monitoring and an awareness of potential psychiatric and behavioral adverse effects, as well as setting realistic endpoints for response.

Role of medication

Primary care physicians should consider the evidence supporting the use of both medication and psychosocial interventions when managing adolescent depression (Table 3).

When to prescribe

If symptoms persist for several weeks, medication may be indicated. Particularly noteworthy symptoms are clear and consistent disturbance of sleep (not lack of sleep from being on MSN with friends late at night), changes in eating, and impaired functioning due to poor concentration. A comorbid anxiety disorder, especially obsessive compulsive disorder, is also an indication.

What to prescribe

Fluoxetine is the only medication with any evidence for efficacy, but it must be noted that any improvement seen in trials was slight and only the clinicians thought the patients were better; self-ratings, parent ratings, and global function ratings were no better with medication than with placebo.[26] Remission rates were no higher with fluoxetine compared with placebo, and a reanalysis of some of these data suggests that the improvements were accounted for by comorbid anxiety.[27] In a second trial, fluoxetine was effective on only one of the two main improvement outcome measures.[28] The recently reported large trial with sertraline (Zoloft) showed a marginal effect[29] and reanalysis suggests that the results were negative.[17,30] Paroxetine and venlafaxine are no longer indicated for initial treatment as they were found to be ineffective in large well-designed trials. In nearly a dozen controlled trials, albeit with small sample sizes, tricyclic antidepressants were found to be no better than placebo.[20] Mirtazepine (Remeron) failed in two unpublished trials, while one of two unpublished trials for citalopram (Celexa) had positive results.[18,30] We have no data on fluvoxamine (Luvox) in depression, although fluvoxamine does appear to be helpful in anxiety disorders.[31] Several SSRIs are approved for treating obsessive compulsive disorder in the United States (sertraline, fluovoxamine, and fluoxetine), but none are approved for this use in Canada. Other classes of antidepressant such as buproprion (Wellbutrin) and reboxetine (Edronax) have not been investigated for use in young people. We suspect that older adolescents (age 17 to 18) will respond to these medications more like adults.

To summarize what we do know from the research, if you prescribe any antidepressant in a placebo-controlled trial, 40% to 60% of subjects will show some improvement, but not necessarily a great deal of improvement. In an open trial, which resembles clinical practice, up to 80% of subjects will improve, and in the majority of cases this will be a placebo effect enhanced by supportive interventions and the expectations of the physician and patient. Patients with anxiety symptoms may improve more, at least in the anxiety area. Nevertheless, a small improvement from the medication can add to improvements gained by other practical steps to further enhance functioning.

Monitoring adverse effects

Children and teens on antidepressants need to be monitored closely as they appear to be more sensitive to certain behavioral side effects; the dose should also be increased more slowly. Recent concerns have highlighted issues that are not new but have often been underplayed: SSRIs can have psychiatric and behavioral side effects in up to one-quarter of children in clinical settings[32-34] and also in clinical trials.[35] Although SSRIs are better tolerated in terms of physical side effects than tricyclic antidepressants, a new profile of psychiatric adverse effects characteristic of this class of medications has added to confusion in assessing young people on antidepressants. Adverse effects such as agitation, suicidal ideation, and emotional lability often overlap with symptoms of the condition being treated. Sometimes these symptoms, if not recognized as adverse effects, may lead physicians to increase the antidepressant dose or add other medications such as atypical neuroleptics or mood stabilizers, when the real solution would be to reduce or discontinue the SSRI. While it is reassuring that the actual rate of suicide attempts is low, it is concerning to know that at least a doubling of the rate of suicidality (ideation and attempts) has been documented with paroxetine, sertraline, and venlafaxine, especially in depression trials but also in anxiety disorder trials.[1-5,7,29,30] New labeling and advisories[8] emphasize these behavioral and emotional adverse effects, and recommend close monitoring early in treatment and when the dose is adjusted up or down. An SSRI information and monitoring form is available to facilitate the process of informing families and of monitoring both response and adverse effects. This form can be downloaded from the web site of the College of Physicians and Surgeons of British Columbia.[36]

Monitoring is most important during the initiation phase when many of these effects may occur. This means seeing the patient within 1 or 2 weeks of initiating medication and inviting teens or their parents to contact you by telephone or e-mail if concerns occur within the first week. E-mail is a good way to stay in touch, as even shy teens are comfortable using it, and you can print out a copy of your correspondence for the file. Teens and the parents of young children in particular should be informed about agitation and behavioral activation, as these responses may be an indication of the need to reduce the dose or go more slowly. Teens in particular need to know about the sexual side effect of anorgasmia, which may occur in up to 60% of those treated for a year, and may be too embarrassing for them to bring up themselves. You should be alert, especially in anxious individuals, to the potential for later development of amotivational and apathetic symptoms that patients themselves may not notice, but teachers and parents will observe. You should also remind teens that marijuana and other recreational drugs used in combination with antidepressants may produce atypical, panicky, or even paranoid reactions.

Conclusions

We have been given a wake-up call and shown that our expectations for the benefits of SSRIs and other antidepressants in young people were overly optimistic. Medications had the advantage of being relatively inexpensive compared with the services of a private psychologist or counselor and readily available to the primary care physician. We are now being forced to take a broader approach—one that must realistically take into account the high prevalence and recurrence rate of depression. Fortunately, there is mounting evidence for the effectiveness of CBT-based school and community programs in depression prevention as well as for CBT-based depression treatment. This will lend more support to the demand for publicly funded cognitive behavioral treatments for depression and help us respond to the disappointing results in the pharmacological research.

Competing interests

Currently, Dr Garland holds no consultancies and is not receiving honoraria from industry sources. In the past, Dr Garland received research funding from Pfizer and GlaxoSmithKline, and she is now participating in a CIHR-funded study that has requested supplemental funding from Lundbeck.

Table 1. Antidepressant advisories in 2003.

Table 2. Practical management of adolescent depression.

Table 3. Summary of evidence-based treatments.

E. Jane Garland, MD, FRCPC

Dr Garland is clinical professor of psychiatry at UBC and clinical head of the Mood and Anxiety Disorders Clinic at BC’s Children’s Hospital.