brand logo

Am Fam Physician. 2000;62(10):2297-2308

A more recent article on depression in children and adolescents is available.

See related patient information handout on depression in children, written by the authors of this article.

Depression among children and adolescents is common but frequently unrecognized. It affects 2 percent of prepubertal children and 5 to 8 percent of adolescents. The clinical spectrum of the disease can range from simple sadness to a major depressive or bipolar disorder. Risk factors include a family history of depression and poor school performance. Evaluation should include a complete medical assessment to rule out underlying medical causes. A structured clinical interview and various rating scales such as the Pediatric Symptom Checklist are helpful in determining whether a child or adolescent is depressed. Evidence-based treatment guidelines from the literature are limited. Psychotherapy appears to be useful in most children and adolescents with mild to moderate depression. Tricyclic antidepressants and selective serotonin reuptake inhibitors are medical therapies that have been studied on a limited basis. The latter agents are better tolerated but not necessarily more efficacious. Because the risk of school failure and suicide is quite high in depressed children and adolescents, prompt referral or close collaboration with a mental health professional is often necessary.

Just 40 years ago, many physicians doubted the existence of significant depressive disorders in children, primarily because they believed that children lacked the mature psychologic and cognitive structure necessary to experience these problems. However, a growing body of evidence has confirmed that children and adolescents not only experience the whole spectrum of mood disorders but also suffer from the significant morbidity and mortality associated with them. Suicide has become a growing public health concern as successive generations have shown a parallel increase of suicide and depression in the pediatric age group.1,2 Childhood depression, like the depression of adults, can encompass a spectrum of symptoms ranging from normal responses of sadness and disappointment in stressful life events to severe impairment caused by clinical depression that may or may not include evidence of mania.35


In the United States, the prevalence of major depressive disorder is approximately 1 percent of preschoolers, 2 percent of school-aged children and 5 to 8 percent of adolescents.2,6 The prevalence of depression appears to be increasing in successive generations of children, with onset at earlier ages.1,7

The gender ratio is equivalent in prepubertal children and increases to a 2:1 female-to-male ratio in adolescents.3 Risk factors include a family history of depression, previous depressive episodes, family conflict, uncertainty regarding sexual orientation, poor academic performance and comorbid conditions such as dysthymia, anxiety disorders and substance abuse disorders.3,6

Dysthymic disorder has a prevalence of 0.6 to 1.7 percent in prepubertal children and 1.6 to 8 percent in adolescents.8 It is considered a “gateway” disorder because of its relatively early age of onset and increased risk of subsequent affective disorders.4,9

Clinical Presentation

The criteria for the diagnosis of major depressive disorder are outlined in Table 1.35 However, the clinical expression of these criteria varies by age.

Sadness variationTransient, normal depressive responses or mood changes to stress.Emotional support
BereavementSadness is related to a major loss that typically persists for less than two months after the loss. (These responses do not include guilt about things other than actions taken or not taken by the survivor at the time of death.) Thoughts of death and morbid preoccupation with worthlessness are also present.Emotional support; counseling
Sadness problemSadness or irritability that begins to resemble major depressive disorders in a milder form (see criteria for major depressive disorder). However, these behaviors lack the severity to qualify for a depressive disorder. The symptoms are more than transient and do have a mild impact on functioning. Bereavement lasting for more than two months may also qualify.Support; counseling; consider medication
Adjustment disorder with depressed moodDevelopment of emotional or behavioral symptoms in response to an identifiable stressor which occur within three months of the stressor. Symptoms include depressed mood, tearfulness and hopelessness, and occur in excess of what would usually be expected from exposure to the stressor and cause significant impairment in social and occupational/academic functioning. Once the stressor (or its consequences) has terminated, the symptoms resolve within six months. Acute episodes last for less than six months, and chronic episodes last six months.Psychotherapy; medication (SSRIs or TCAs)
Major depressive disorderSignificant distress or impairment manifested by 5 to 9 of the criteria listed below occurring almost daily for two weeks. A depressed/irritable mood or diminished interest/pleasure must be among these criteria and must represent a change from previous functioning:Psychotherapy; medication (SSRIs or TCAs)
  • Depressed/irritable

  • Recurrent thoughts of death and suicidal ideation

  • Diminished interest or pleasure

  • Weight loss/gain

  • Psychomotor agitation/retardation

  • Fatigue or energy loss

  • Feelings of worthlessness

  • Diminished ability to think/concentrate

  • Insomnia or hypersomnia

Dysthymic disorderDepressed/irritable mood for most of the day, for more days than not (by subjective account or observation by others) for one year, including the presence of two of the following symptoms:Psychotherapy; medication (SSRIs or TCAs)
  • Poor appetite/overeating

  • Insomnia/hypersomnia

  • Low energy or fatigue

  • Poor concentration/difficulty making decisions

  • Feelings of hopelessness

These symptoms are less severe than those in a major depressive disorder but are more persistent.
Bipolar I disorder, most recent episode depressedThe patient is currently in a major depressive episode with a history of one manic or mixed episode. The mood episodes are not better accounted for by schizoaffective disorder and are not superimposed on schizophrenia, schizophreniform disorder, delusional disorder or a psychotic disorder not otherwise specified.Medication (lithium, anti-convulsants, psychiatric consultation), psychotherapy
Bipolar II disorder, recurrent major depressive episodes with hypomanic episodesPresence or history of one major depressive episode + one hypomanic episode (similar to manic episode but only needs to last four days and is not severe enough to cause marked impairment in function). There have never been any manic or mixed episodes. The symptoms are not better accounted for by schizoaffective disorder, delusional disorder, schizophrenia or psychotic disorder. The symptoms overall cause clinically significant distress or impairment in social, occupational or other important areas of functioning.Medication (lithium, anti-convulsants, psychiatric consultation), psychotherapy

Infants and preschoolers do not have the ability to express feelings of sadness in language. Therefore, depressive symptoms must be inferred from overt behavior, including apathy, withdrawal from caregivers, delay or regression of developmental milestones and failure to thrive that has no organic cause.2,3 Because of the difficulties of diagnosing any psychiatric disorder in this age group, clinicians must rely heavily on parental history, evaluation of parent-child interactions and play interviews by appropriately trained professionals.10

School-aged children are cognitively able to internalize environmental stressors (e.g., family conflict, criticism, failure to achieve academically) and display low self-esteem and excessive guilt. However, much of this inner turmoil is expressed through somatic complaints (headaches, stomachaches), anxiety (school phobia, excessive separation anxiety) and irritability (temper tantrums and other behavioral problems).8,11,12 Because these children are in school, their teachers can also serve as a valuable source of information and should be included in the evaluation process if possible. It is important to note that some depressed children attempt to compensate for their low self-esteem by trying to please others and be accepted. Because in this effort they may excel academically and behave well, their depression may go unnoticed.2

Adolescents experience many developmental challenges as they strive to separate from their parents, become autonomous and establish their own identities. In this process they depend increasingly on their peer groups. This period of biopsychosocial maturation creates the conditions for adolescents to experience a greater sense of hopelessness and despair at a time when their ability to complete suicide is greater than when they were younger. They also exhibit more anhedonia, hypersomnia, weight change and substance abuse than younger children.11

Because of their developmental struggle with autonomy from authority figures, adolescents pose a particular challenge to clinicians. The establishment of good rapport is essential for building trust. The issue of confidentiality is key and, from the outset, the clinician should explain the ground rules governing when a parent or other third party will be informed of what is discussed with the patient.10

Despite the differences outlined above, many similarities are found between school-aged children and adolescents in terms of clinical symptoms of major depressive disorder. This includes the frequency and severity of most symptoms such as depressed mood, guilt, anger, irritability, suicidal ideation and attempts, and poor self-esteem.11 In fact, many of these criteria are those used to diagnose depression in adults, revealing a life-span perspective of this disorder.

Dysthymic disorder manifests in similar but milder symptoms compared with major depressive disorder. However, its chronic nature can severely impair a child's development of the social skills needed to combat these same symptoms. While the mean duration of episodes of major depressive disorder in clinically referred samples is seven to nine months, the duration of dysthymic disorder is three years. While 90 percent of children with major depressive disorder reach remission by 18 months to two years after onset, about one half of children diagnosed with dysthymic disorder are still struggling with their symptoms at this point.6,9

These disorders are associated with significant comorbidity, including the superimposition of major depressive disorder and dysthymic disorder on each other in what is called “double depression.” In major depressive disorder, 40 to 70 percent of patients have comorbid psychiatric disorders, whereas dysthymic disorder has a 50 to 70 percent comorbidity rate.6 Thus, it is important for physicians to maintain a high index of suspicion for the presence of other disorders when depression is uncovered. Literally speaking, “Misery loves company.”

These comorbidities include the “double” depressions, anxiety disorders and disruptive disorders (conduct disorder, attention-deficit/hyperactivity disorder). Major depressive disorder is also associated with substance abuse and personality disorders. Comorbidities are negative prognostic factors because they increase the duration and severity of episodes of major depressive disorder, and the likelihood of recurrence and suicidal tendencies.2,6,8,11,13

After recovery from an episode of major depressive disorder, many children continue to show sequelae. These include poor self-esteem, impairment of interpersonal relationships, increased risk-taking leading to smoking and early pregnancy, subclinical depressive symptoms and impairment of global functioning.8 Family physicians are in an ideal position to follow these patients over the lifespan—not only to screen for these disorders but to participate in the evaluation, diagnosis, treatment and monitoring of risk factors, symptoms, sequelae and recurrences.


Some studies have shown that only one third of parents who had psychosocial concerns about their children planned to discuss them with their pediatrician. When parents did initiate this discussion, only 40 percent of pediatricians responded, and the response rate was even lower when the parents were less educated.14,15 This finding indicates that the vast majority of psychosocial disorders in children are being missed, potentially at a time when appropriate evaluation and treatment could prevent or ameliorate the significant subsequent morbidity and mortality associated with them. It is important that screening measures be implemented to help identify children in need.

Many of the structured interview formats, rating scales and questionnaires that are now used primarily for research may be useful in the diagnosis, monitoring and treatment follow-up of childhood depression.16,17 However, most of them require specific computer programs and trained staff to appropriately analyze the data, and therefore are often not practical for the primary care physician.

Because of the ever-expanding managed care environment with its focus on efficiency and cost-containment, it is difficult for clinicians to spend enough time during well-child visits to assess for potential psychosocial dysfunction. One alternative is to use questionnaires such as the Pediatric Symptom Checklist,18 as a routine screening measure (Figure 1).

The rightsholder did not grant rights to reproduce this item in electronic media. For the missing item, see the original print version of this publication.

The Pediatric Symptom Checklist is a 35-item checklist designed to be filled out by parents of six- to 12-year-old children to assess their impressions of their children's psychosocial function. It can be completed and scored in less than five minutes. This screening tool has been found to be well accepted by patients in an office setting. Its specificity ranges from 68 percent in middle-class samples to 100 percent in samples of lower socioeconomic status. The sensitivity ranges from 80 percent in lower socioeconomic status samples to 95 percent in middle-class samples, compared with more detailed assessments made by mental health professionals.15,18 Because of its relatively good sensitivity and specificity, and also because of its ease of administration, this checklist can be a valuable tool to help physicians better screen patients and focus their limited time on those most likely to need help.

Once a patient has been identified by such a screening measure or, even more directly, by parental or clinician concern, a detailed psychosocial history should be obtained to assess the need for treatment or referral. More recently, a set of high-yield questions was published that may be a good starting point for the primary care physician (Figure 2).19

The rightsholder did not grant rights to reproduce this item in electronic media. For the missing item, see the original print version of this publication.

The diagnosis of depressive disorders necessitates a thorough medical and psychiatric evaluation. A number of medical disorders can mimic depression (Table 2).5 The history and physical examination can help to determine which laboratory studies may be appropriate. Several baseline laboratory studies should be considered, when indicated, as part of the evaluation of most children suspected of having depression (Table 3).2

Infectious mononucleosis
Human immunodeficiency virus infection
Neurologic disorders
Addison's disease
Cimetidine (Tagamet)
Alcohol abuse
Drug abuse and withdrawal
Oral contraceptives
Electrolyte abnormality
Wilson's disease
Complete blood cell count with differential—to rule out anemia or infection
Determination of electrolytes, creatinine level, BUN—to rule out electrolyte disorders or renal disease
Liver function studies—to rule out hepatitis and drug effects
Thyroid function tests—to rule out thyroid disorders and to serve as a baseline if lithium treatment is being considered
Electrocardiogram—as a baseline if a TCA is being considered
Electroencephalogram—to assist in ruling out a seizure disorder

The psychiatric evaluation entails a detailed history, including characteristics of the symptoms: onset, duration, frequency, intensity, severity and pervasiveness (home and school). Other medical and psychiatric illnesses should be screened for because depression is associated with many psychiatric comorbidities and is more frequent in children with chronic medical conditions. Lastly, a developmental history, social history, family history of psychiatric disorders and assessment of home and school environment and potential stressors should be obtained.10,14

The clinical interview will vary according to the age group and developmental status of the child. The interview format may range from unstructured play interviews and assessments of child-parent interactions with infants and toddlers, to play interviews with open-ended questions in school-aged children, to more discussion-based interviews with adolescents. Neuropsychologic testing is also an important component of the evaluation, to help rule out neurologic or learning disorders and assess the child's developmental capabilities.10


The evidence-based literature on the treatment of major depressive disorder in children is limited, and most of the available treatment strategies are based on extrapolation from data obtained in studies of adults. As with adults, optimal treatment rests on a multidisciplinary approach, including psychotherapy, pharmacotherapy and education of the patient and family.

Psychotherapy can be useful as initial therapy for children and adolescents with mild to moderate depression and as an adjunct to medications for children with more severe depression.6 It encompasses a multitude of approaches including play therapy, psychodynamic therapy, supportive therapy, interpersonal therapy, family therapy, group therapy and cognitive behavior therapy. It is important to determine the child's cognitive and emotional developmental level in deciding the most appropriate approach. For example, play therapy and parental training would likely be the most appropriate for use in depressed preschool-aged children, while psychodynamic or cognitive behavior therapy would be more appropriate in older children and adolescents.5

A preponderance of the available research has dealt with cognitive behavior therapy in adolescents (10 years and older). The basis of this treatment modality is helping patients change the negative cognitions about themselves and the world that contribute to their depression. In one meta-analysis20 it was found that cognitive behavior therapy has both short- and long-term effectiveness in the treatment of adolescents with depressed or dysphoric mood. Because of the relatively high levels of cognitive distortions in depressed adolescents, it was not surprising to find that cognitive behavior therapy had greater efficacy in the treatment of depressed adolescents than did nondirective supportive therapy and systemic-behavioral family therapy.21 Additional research is necessary to delineate the effectiveness, or lack thereof, of psychotherapies, especially in younger children.

Optimal treatment also involves education of the patient and his or her family. This not only enhances understanding of the disorder but can also improve patient compliance with therapy. It can decrease the tendencies of the parents to blame themselves or their child for the disorder and may alert parents to depressive tendencies in themselves or other family members.6 Given the high prevalence of affective illness in first-degree relatives of patients with major depressive disorder, family education can be an integral component of improving the patient's environment and support structure by allowing for the treatment and healing of the whole family. Optimally, this education and collaboration should occur within the patient's community as well, especially the school, to provide improved support in all significant settings.

Medical therapy in depressed children is fraught with controversy and limited information. In addition, the high placebo response in children confounds many studies of the effects of antidepressants in children. This rate is about 50 percent in studies of tricyclic antidepressants versus placebo.8

Tricyclic antidepressants (Table 4) were the first available pharmacotherapy for depression in children. Although some open studies reported an improvement rate of 60 to 80 percent in children with major depressive disorder, subsequent randomized controlled trials, which accounted for the high placebo effect, have shown equivocal results.22 Most of these trials involve small numbers of patients over short periods. A meta-analysis23 of 12 randomized controlled trials revealed no significant improvement over placebo. In fact, an inverse relationship was noted between the quality of the studies and the estimated treatment effect, indicating that the best available studies showed the least effectiveness of the tricyclic antidepressants. Some researchers expressed concern that these poor results may be secondary to the variability of plasma levels of tricyclic antidepressants in children. However, a double-blind, placebo-controlled trial of nortriptyline, which controlled for plasma levels, showed no significant response over that of placebo.24

The rightsholder did not grant rights to reproduce this item in electronic media. For the missing item, see the original print version of this publication.

Because of the concern about anticholinergic and potentially cardiotoxic side effects of the tricyclic antidepressants, especially in the case of overdose, these antidepressants have become second-line medical therapy, giving precedence to the newer selective serotonin reuptake inhibitors (SSRIs).

SSRIs (Table 5)25 dominate the pharmacotherapy of depression by primary care clinicians because of their relative safety. They are as efficacious as tricyclic antidepressants and have almost none of the anticholinergic or antihistaminergic side effects associated with tricyclic antidepressants secondary to their relative selectivity of 5-HT sites. However, much of the rationale for their use in children revolves around data available from studies of adults.

The rightsholder did not grant rights to reproduce this item in electronic media. For the missing item, see the original print version of this publication.

Only a limited amount of evidence-based research has been done on SSRIs in children. The first double-blind, randomized, placebo-controlled trial of fluoxetine (Prozac) involving 40 adolescents (13 to 18 years of age) revealed a nonsignificant advantage in the fluoxetine-treated group.26 A subsequent double-blind, randomized, placebo-controlled trial of 96 children and adolescents (seven to 17 years of age) revealed a significantly improved response of fluoxetine over placebo (56 percent versus 33 percent).27 However, this study needs to be replicated before its conclusions can be generalized to the greater population of depressed children. An open-label, prospective study of 13 depressed adolescents, showing promise for sertraline (Zoloft), also needs to be corroborated by further investigation.28

Tricyclic antidepressants have been found to be effective in the treatment of attention-deficit/hyperactivity disorder (considered second-line to stimulants) and enuresis. They may have a role in the treatment of obsessive-compulsive disorder. Given the multiple comorbidities in children with major depressive disorder, tricyclic antidepressants may benefit children with mild to moderate major depressive disorder in conjunction with attention-deficit/hyperactivity disorder or enuresis.22 Some evidence indicates the efficacy of SSRIs in children with obsessive-compulsive disorder and anxiety disorders, thus offering dual therapy in a single medication in children with these comorbidities.25

The side effect profile of tricyclic antidepressants includes the usually mild anticholinergic symptoms and weight gain, and the more ominous cardiotoxic effects. Although a few cases have been reported of sudden death in children taking therapeutic dosages of tricyclic antidepressants, most lethal arrhythmias occur in cases of overdose.22 Because of this lethal potential, serum levels of tricyclic antidepressants should be carefully monitored. This is especially important in young children, who generally have a more rapid hepatic metabolism of medications. However, some children display slow hydroxylation and clear tricyclic antidepressants more slowly.22 Because of their cardiotoxic potential, tricyclic antidepressant therapy may be contraindicated in children with heart disease (i.e., significant structural or rhythm abnormalities).

Baseline studies before initiating treatment with tricyclic antidepressants should include electrocardiography and a check of resting blood pressure, resting pulse and weight. Close clinical follow-up is important, especially when the dosage is changed or if symptoms of toxicity, such as lightheadedness and tachycardia, occur.6

SSRIs have side effects, including mild gastrointestinal upset, sedation and activation symptoms. In addition, they may induce hypomanic or manic symptoms in vulnerable persons, unmasking bipolarity. They can also lead to later-onset frontal lobe symptoms. These latter symptoms, which include disinhibition, apathy and indifference, may mislead clinicians into thinking that the dosage is subtherapeutic when, in fact, these patients need a decrease in dosage. Serotonin syndrome must also be considered because of its lethal potential, especially when SSRIs are prescribed in conjunction with tricyclic antidepressants or monoamine oxidase inhibitors.25

Tricyclic antidepressants and SSRIs both have discontinuation syndromes. Patients can experience flu-like symptoms on the withdrawal of tricyclic antidepressants, and dizziness, nausea, headaches and sensory disturbances on the withdrawal of SSRIs. Therefore, both of these agents require tapering if discontinued.22,25

Limited data exist on the appropriate dosing of these medications in children. As in geriatrics, the adage, “Start low and go slow,” finds prudent application in this age group.

Even less information is available regarding the use of other antidepressant agents in children. Some promising results have been reported with nefazadone (Serzone), but no published information is available pertaining to the use of trazadone (Desyrel) or mirtazapine (Remeron).29 A small trial of venlafaxine (Effexor) resulted in no significant difference from placebo, but this study followed patients for only six weeks.30 Bupropion (Wellbutrin) has demonstrated some benefits in the treatment of attention-deficit/hyperactivity disorder, but it has not been studied specifically in relation to depression.29


Guidelines for the appropriate referral of depressed children and adolescents are sparse in the literature. Common sense dictates that children who are the most difficult to diagnose and treat should be the ones to refer as early as possible. These would include patients with severe depression, especially those involving psychosis or high suicide risk; very young patients (infants, toddlers); those refractory to psychotherapy/medical therapy; and those with significant comorbidities. The level of comfort and experience of the primary care physician should play a role in the decision to refer, and the option of comanagement of a patient is a viable alternative.

Because in the context of the managed care environment the responsibility of the identification and management of these children is increasingly falling on the primary care physician, close collaboration with mental health professionals (child psychiatrists, psychologists, counselors, social workers) is essential for their appropriate management. Because of the high familial clustering of depressive disorders, early identification of affected children can also assist in the identification of affected family members and in the healing of the whole family. Given the lifetime and familial nature of these disorders, family physicians are in an excellent position to screen for, treat, follow and monitor these patients, and refer them when appropriate.

Continue Reading

More in AFP

More in PubMed

Copyright © 2000 by the American Academy of Family Physicians.

This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP.  See permissions for copyright questions and/or permission requests.