Epidemiology

The prevalence of major depression has been estimated at 1%–3% in prepubertal children and 3%–9% in adolescents, although rates vary with the population, the diagnostic criteria, and the methods of assessment. In the Oregon Adolescent Depression Project (Lewinsohn et al. 1994), the lifetime prevalence of at least one episode of major depression by late adolescence was 20%–25%, and the lifetime prevalence of dysthymic disorder was 3%. Many more youth have depressive symptoms than have full major depression or dysthymia. Before puberty, depression is equally common in boys and girls, with a change in adolescence to the female predominance found in adults.

Etiology

Etiological factors for mood disorders in children are similar to those in adults. Depression in a parent is a significant risk factor via both genetic transmission and interpersonal influence (i.e., parent’s modeling, emotional unavailability, decreased capacity for parenting). Abuse and neglect may be significant precipitants, especially in very young children.

Course and Prognosis

Mood disorders in childhood are serious and potentially fatal problems. In a 15-year follow-up of children with major depression, 4% had died by suicide (Wolk and Weissman 1996). A 10- to 15-year follow-up of a clinical sample of subjects with adolescent onset of major depression found that nearly 8% had committed suicide (Weissman et al. 1999). Risk of suicide increases with a past history of suicide attempt, comorbid substance abuse, access to lethal means (such as a gun), impulsivity, conduct disorder, family history of suicide attempts, or physical or sexual abuse. Over the full age range, earlier age at onset implies a lengthier and more severe course and greater genetic loading. Over the past 25 years, the age at onset of major depression appears to have decreased. In a prospective natural history observation study of clinically referred prepubertal children, the median length of the major depressive episode was 9 months. The median length of time for recovery from dysthymia was 4 years. Dysthymia often progressed to a major depressive episode before resolving. Risk for subsequent episodes of major depression or dysthymia is high (Kovacs 1996). In a sample of high school students, episodes of major depression lasted as long as 10 years (mean = 26 weeks; median = 8 weeks). Of those who recovered, one-third had another episode within 4 years. Suicidal ideation was associated with earlier onset, longer episodes, and earlier relapse of depression (Lewinsohn et al. 1994). Follow-up studies of depressed youth consistently find increased risk of repeated mood disorders and other psychiatric disorders, personality disorders, substance abuse, and impairment in subsequent educational and vocational achievement and peer and family relationships.

Childhood-onset depression is more likely than adult-onset depression to evolve into bipolar disorder. Among adolescents with major depression, subsequent bipolar disorder was predicted by precipitous onset of symptoms, psychomotor retardation, psychotic features, psychopharmacologically precipitated hypomania, and family history of bipolar disorder (Strober and Carlson 1982).

Evaluation and Differential Diagnosis

The clinician should ask children direct questions about depression. Young children have more difficulty recognizing and verbalizing their feelings and may use idiosyncratic words to describe dysphoria or anhedonia. Both parent and child reports are essential. Some children report their mood states more accurately than their parents can, although young children may have limited insight into other symptoms. A trained clinician should observe children for depressed affect. The Children’s Depression Inventory (CDI; Kovacs 1985) may be a useful self-report screening instrument.

Assessment of the degree of potentially dangerous behavior is crucial. The clinician can question children directly regarding suicidal ideation, suicide plans, and suicide attempts. No evidence indicates that inquiries about wishes or attempts to die increase the risk of self-destructive behavior. Substance abuse increases the risk of suicide. The clinician should ask specifically about emotional or physical abuse or neglect of the child, access to lethal weapons, family members or friends who have attempted suicide, nonsuicidal self-injury, prior suicide attempts, and any potential protective factors (relationships or spiritual beliefs).

Even in nonclinical epidemiological samples, anxiety disorders, ADHD, and disruptive behavior disorders frequently coexist with depression and dysthymia, with onset before or after the onset of the mood disorder. Alcohol or drug abuse may cause a secondary depression or may represent an attempt to “self-medicate” dysphoria. Diagnosis of a mood disorder may be impossible without observing the patient in a drug-free state. Separation anxiety disorder may resemble major depression or dysthymic disorder or may coexist with it. Children younger than 4 years may develop a clinical picture similar to major depression when separated from their parents. Children with reactive attachment disorder secondary to parental abuse or neglect who present with lethargy, apathy, and withdrawal may appear depressed.

The strong family clustering of mood disorders suggests that parents and siblings should be evaluated routinely for mood and anxiety disorders, and treatment should be provided or arranged as necessary.

Treatment

If risk-taking behavior or suicidal ideation is present, close parental and psychiatric supervision is needed. Psychiatric hospitalization may be required for youngsters who are psychotic or seriously suicidal or who do not respond to outpatient treatment.

Up to 60% of patients with mild to moderate depression respond to supportive treatment. However, early-onset mood disorders often have devastating effects on development. Long-term treatment is often needed. Even after spontaneous remission or successful treatment, reduced coping skills, cognitive patterns associated with depression, and impaired interpersonal relationships with peers and family members may require individual, group, or family therapy to address developmental deficits or sequelae of the depression. Whatever the treatment, involvement of the family is even more crucial than for adult patients. Both patients and families benefit from psychoeducation (provision of information about the disorder and its treatment) (see Appendix, “Resources for Parents”) and from instruction in relapse prevention (taking medication as prescribed, recognizing early symptoms of relapse, and avoiding precipitants of relapse such as sleep deprivation and drug abuse).

For nonpsychotic depression, psychotherapy is typically the first treatment. Medication is added if symptoms do not improve in 4–6 weeks (see Chapter 17, “Psychopharmacology”). For more severe depression, medication is indicated as initial treatment. Fluoxetine (age 8 years and older) and escitalopram (age 12 years and older) are currently approved by the U.S. Food and Drug Administration (FDA) for the treatment of pediatric major depressive disorder.

Cognitive-behavioral therapy (CBT) and interpersonal therapy (IPT) techniques developed for the treatment of depression in adults have been adapted for use in children and adolescents (Mufson et al. 2004) (see also “Additional Reading”). Social withdrawal and limited peer relationships may respond to behavioral activation and social skills training. Remedial education or tutoring may be needed when the illness has interfered with learning in school.

For adolescent depression, a National Institute of Mental Health (NIMH)–funded multisite study (Treatment for Adolescents With Depression Study Team 2003) compared randomized assignment to fluoxetine, CBT, the combination, or placebo in 439 adolescents with MDD. Summary findings were that on the Child Depression Rating Scale, fluoxetine (with or without CBT) was superior to placebo in efficacy with an acceptable risk-benefit profile, and that at week 12 of treatment, combined treatment and fluoxetine alone were equal in benefit and had greater benefit than CBT, which was not significantly different from placebo (March et al. 2004). Response was faster with combined treatment than with CBT alone, but the results of all active treatments converged by 18 weeks (Kennard et al. 2009). Further analysis showed (somewhat counterintuitively) that combined treatment had an advantage over fluoxetine alone for mild to moderate depression but no significant advantage for moderate to severe depression.

The subsequent NIMH-funded six-site Treatment of Resistant Depression in Adolescents (TORDIA) study included 334 adolescents with MDD or dysthymia who had not responded to an adequate trial of a selective serotonin reuptake inhibitor (SSRI) for at least 8 weeks. Subjects were randomly assigned to one of four conditions: switch to a different SSRI (predominantly citalopram) or to venlafaxine or add CBT to either of those medication switches. Adding CBT to either medication switch was more effective than a medication switch alone. There was no difference in efficacy between switch to another SSRI or venlafaxine, but venlafaxine produced more side effects (Brent et al. 2008). The initial trajectory by 6 weeks of treatment predicted whether the patient would remit, suggesting that clinicians should not wait longer than 6–8 weeks to switch or add treatments if the patient is not responding.

Even with evidence-based treatments, many youth remain impaired and vulnerable to relapse. Research is ongoing to find strategies to improve remission rate and reduce or delay relapses. For severe (especially if psychotic) adolescent depression unresponsive to adequate trials of pharmacotherapy, electroconvulsive therapy may be considered, although there are no research studies in adolescents. Transcranial magnetic stimulation appears to be useful in depressed adults, but there is no completed controlled study in adolescents.

Children and adolescents who have seasonal affective disorder may respond to morning treatment with bright white light (Swedo et al. 1997).