The following treatment recommendations are based on published treatment guidelines, including the Practice Parameter produced by the American Academy of Child and Adolescent Psychiatry (AACAP),[70] the Guidelines for Adolescent Depression in Primary Care (GLAD-PC) produced by expert consensus and empirical evidence,[71] the antidepressant treatment algorithm based on the Texas Children's Medication Algorithm Project (CMAP) consensus conference,[72] the clinical guideline on depression in children and young people by the National Institute for Health and Care Excellence (NICE) in the UK,[73] and the guidelines on depression in adolescents and young adults by the Australian National Health and Medical Research Council (NHMRC).[74]

It is crucial to make an accurate diagnosis, based on a comprehensive assessment and review of the history, with input from multiple sources. The safety of the child and others, and the duration and severity of depression, need to be evaluated carefully to help determine the appropriate level of care and treatment modality. Ongoing assessments of safety, symptoms, and treatment response are important in limiting adverse events and in optimising and guiding the treatment approach.

Confidentiality should be discussed with the patient and his or her family, including the limits of confidentiality (e.g., the need to inform parents or legal authorities if there is an imminent risk of harm to the patient or to others [in keeping with clinicians’ local legal frameworks]).[57]

Phases of treatment

The treatment of depression can be divided into 3 phases.

  • Acute: lasting for 6-12 weeks; the goal of this phase is to achieve remission (i.e., for the patient to become asymptomatic or to have minimal symptoms only).

  • Continuation: immediately follows the acute phase, and lasts for 6-9 months; the goal of this phase is to prevent relapse. The same dose used for acute treatment should be continued. Without continuing treatment, relapse rates are very high, as demonstrated in both psychotherapy and medication studies. In the case of fluoxetine, even with continuation of treatment, many children and adolescents relapse. It has also been found that adding 6 months of continuation-phase cognitive behavioural therapy (CBT) after acute CBT significantly lowered relapse rates compared with historical controls (6% vs 50%).[76]

  • Maintenance: follows the continuation phase in some people; 1-2 years of the maintenance treatment may be recommended for young people at risk for recurrence (e.g., those with high genetic loading, chronic depression, multiple episodes, and severe episodes). One small paediatric depression maintenance study has been reported. Although a larger study is needed, maintenance treatment is recommended by treatment guidelines.

Establishing a good therapeutic alliance with both the child and the parents, providing psychoeducation, and including family in the treatment decision-making process may improve adherence and promote positive outcomes throughout the treatment phases.

Urgent measures for children at risk of harm to themselves or others

Children and adolescents who are depressed with severe suicidality and without being able to maintain safety, or with significant psychosis, require urgent referral to the emergency department. Hospitalisation may be necessary to:

  • Carry out an urgent mental healthcare assessment

  • Ensure safety for the patient and/or for others

  • Stabilise the patient.

Initial step in all patients: brief active monitoring

For uncomplicated depression, a brief period (≤4 weeks) of active monitoring, with supportive care including psychoeducation for the child and parents, is recommended. Mild depression often resolves with non-specific treatment. Children tend to have a high placebo response.[78] It may be necessary to cut the usual period of 4 weeks of monitoring short if the symptoms are severe. If the depression becomes severe, or suicidality or psychosis develops, immediate treatment and high levels of care (e.g., inpatient treatment) may be required.

A lifestyle assessment and recommendations for changes in diet and exercise may facilitate treatment and achieve better outcomes. There is growing evidence in support of the use of physical exercise to prevent and treat depressive disorders. Several controlled studies have demonstrated that exercise has an efficacy comparable with antidepressant therapy, and superior efficacy compared with placebo, in reducing depressive symptoms in adults.[79] One study indicated that regular exercise significantly reduced the risk of developing dysthymia in adults.[80] Although there are no current published paediatric studies available, several paediatric depression studies are under way.

Usually no further therapy is required, apart from the management of comorbid disorders and specific resistant individual symptoms, unless the depression increases in severity or symptoms persist.

Adjunctive treatment

Typically, adjunctive medications are instituted to manage symptoms associated with depression (e.g., insomnia and agitation) and are discontinued when the target symptom resolves. Adjunctive therapy may be required in any of the treatment steps for depression of all severities. One important issue is the duration of treatment. Because some depressive symptoms may take a long time to resolve, certain patients may need adjunctive medications early on during the acute treatment phase.

Insomnia is a frequent symptom of depression and also a frequent residual symptom.[81] Antidepressants may not sufficiently resolve insomnia, so adding a sleep aid (e.g., melatonin, antihistamine) or using an evidence-based psychosocial intervention for insomnia may be helpful to resolve insomnia early during acute treatment. In a post-hoc analysis of 2 combined studies of fluoxetine versus placebo, adolescents who reported marked insomnia showed no more improvement with fluoxetine compared with placebo, while those without insomnia showed greater benefit with fluoxetine over placebo.[82] No controlled studies have been done to show whether adding a sleep agent would improve depression treatment outcome in the paediatric population. However, in the TORDIA study, adolescents treated with trazodone were 6 times less likely to show a response to treatment compared with those not using a sleep medication, while those using other sleep medications showed similar response to adolescents not using a sleep medication.[83]

For psychotic depression or agitation, an atypical antipsychotic medication may be used concomitantly with an antidepressant. This would be necessary only if the antidepressant therapy was not controlling these symptoms adequately. For mild and low-risk psychotic symptoms or agitation an antidepressant may be sufficient, because the psychotic symptoms may resolve as depression improves.

Some of these drugs are considered off-label in some countries and are only to be prescribed by a specialist experienced in the treatment of paediatric psychiatric disorders in those countries. Guidelines for prescribing are available for North America[70][71] and the UK.[73]

Comorbid disorders

This is an important consideration at all steps of treatment. Anxiety disorders and ADHD are common comorbid disorders. Psychotherapy for an anxiety disorder may be started concomitantly with antidepressant treatment. However, it is not recommended to start both an antidepressant and another medication to treat a comorbid disorder simultaneously. The primary illness, or the disorder causing the most impairment of function, needs to be treated first. Then, sequentially, a second and third treatment to target comorbid disorders may be initiated if required (e.g., a stimulant may be started sequentially with an antidepressant).

Mild depression: inadequate improvement with active monitoring

For mild depression that does not respond to active monitoring, a course of specific evidence-based psychotherapy, such as CBT or interpersonal psychotherapy (IPT), if available and appropriate, may be used.[84][85] CBT has also been shown to have long-term effects on prevention of depression onset.[59][60]

In the US, if there is an inadequate response with specific psychotherapies, selective serotonin-reuptake inhibitor (SSRI) therapy may be used instead, or an SSRI could be added onto existing specific psychotherapy. Usually no further therapy is required, apart from the management of comorbid disorders and specific resistant individual symptoms, unless the depression increases in severity.

Moderate or severe depression: inadequate improvement with active monitoring

Specific psychotherapies may also be used for children and young people with depression of moderate or greater severity. The UK NICE guideline recommends psychotherapy as first-line treatment for all young people with depression of moderate or greater severity.[73] The AACAP practice parameter[70] and the GLAD-PC guidelines[57][71] recommend any of the following approaches for young people at moderate or greater severity:

  • Specific psychotherapies (e.g., CBT, IPT, dialectical behavioural therapy)

  • SSRI medication

  • A combination of specific psychotherapy and SSRI therapy.

If monotherapy with either specific psychotherapy or an SSRI is the initial approach and there is an inadequate response, treatment could be augmented and continued with combined SSRI and psychotherapy.

The UK NICE guideline recommends an initial trial of psychotherapy for all young people with depression of moderate or greater severity, and recommend antidepressants only when used in conjunction with therapy.[73] The Australian NHMRC guidelines have similar recommendations but allow antidepressants when psychotherapy is not effective.[74]

In one randomised controlled trial of moderately to severely depressed adolescents, CBT plus SSRI was no more effective than SSRI alone.[87] There was no increase in disinhibition, irritability, and violence from baseline with either treatment.[87] However, long-term results and safety outcomes from one study indicate that combination therapy may decrease suicidal ideation more than medication alone.[88] In a meta-analysis of combined medication and psychotherapy versus medication alone, there was greater improvement in global functioning with combination treatment, but no difference was reported between the groups on depressive symptom reduction.[89]

SSRIs: general considerations

SSRIs are the drug of choice, if a drug is required, and are often used after non-pharmacological steps have failed. [ Cochrane Clinical Answers logo ] A meta-analysis including 36 trials (6778 participants) found that SSRIs seem to be more beneficial than placebo for treating children and adolescents with depression, but that the effect size was small. The results suggest that, for children and adolescents with depression, the placebo effect plays a significant role in the efficacy of SSRIs.[90] The US Food and Drug Administration (FDA) issued a black box warning on suicidality associated with paediatric use of antidepressants in 2004.[91]


  • The only SSRI that has >2 positive controlled trials and has positive data in both children and adolescents.[92][93][94][95] This finding has been replicated in an adolescent-only trial.[96]

  • Doses may be increased incrementally after initial starting regimen if there is not sufficient improvement (<50% of reduction in depression severity) of depressive symptoms.[97] Adolescents may be more likely to need a gradual increase to a higher dose than children.[98] Occasionally, fluoxetine may be increased to 60 mg daily if there is not sufficient response at other doses, especially in children or adolescents who also have comorbid anxiety disorders.

  • Fluoxetine may have greater and more rapid efficacy in premenstrual dysphoric disorder.[99]

  • Fluoxetine has the longest half-life of the recommended SSRIs (4-6 days) and is a potent inhibitor of cytochrome P450 enzyme 2D6 and 2C19. When concomitant medications that are metabolised by these enzymes are used, a reduced dose may be considered.

  • There is evidence to suggest that fluoxetine during the first trimester of pregnancy may be associated with a slightly increased risk of congenital heart malformations in the baby. In consideration of this finding, the physician should also take into account the established conferred risk to the fetus and newborn when the mother is depressed.[100]


  • May be used as a first-choice antidepressant for adolescents. One trial that included children and adolescents demonstrated positive results for escitalopram over placebo in the adolescent subgroup.[101]

  • It is the active S-enantiomer of citalopram, with double the potency of citalopram. It has a half-life of 27-32 hours. Similar to citalopram, it is a weak 2D6 inhibitor with minimal drug-to-drug interactions.


  • Has been found to be more effective than placebo only when the results of 2 trials were pooled.[102]

  • It has a 26-hour half-life and is a moderate P450 enzyme inhibitor, although at high doses it is a potent inhibitor of 2D6. At high doses, drug-to-drug interactions need to be considered. Stronger evidence for efficacy in anxiety and obsessive-compulsive disorder.


  • Has only 1 positive trial.[103]

  • It has minimal effect on the P450 enzyme system, thus limited drug-to-drug interactions.

Paroxetine is not recommended for children or adolescents as first-line treatment due to lack of efficacy. It has been the subject of 3 controlled trials and all were negative, except for 1 partial positive trial in adolescents.[104][105][106] This medication is not effective and not well tolerated in young children; thus it is not recommended for children. It has a short half-life of 21 hours, and it is a potent inhibitor of 2B6 and 2D6 as well as a moderate inhibitor of 3A4, so drug-to-drug interactions need to be considered.

There are no paediatric depression trials for fluvoxamine.

Adverse effects and safety of SSRIs

Overall, most of the SSRIs and non-SSRIs (considered for use in later steps) are well tolerated by children and adolescents, and adverse effects are mild and short-lived. However, potential adverse effects and precautions need to be discussed thoroughly with the child and parents, to ensure safety and improve adherence. Children and adolescents treated for depression with SSRIs (and serotonin–noradrenaline reuptake inhibitors [SNRIs]) appear to experience a greater number of adverse effects (including severe adverse effects) than those treated with placebo, indicating a need for a careful and individualised approach to prescribing that considers the relationship between anticipated clinical benefit and possible side effects.[90] The FDA issued a black box warning on suicidality associated with paediatric use of antidepressants in 2004.[91] Common adverse effects of SSRIs include:

  • Headaches

  • Nausea

  • Diarrhoea

  • Abdominal pain

  • Insomnia

  • Sedation

  • Tremors

  • Increased bleeding time.

Sexual adverse effects related to SSRIs occur frequently in adults and adolescents (up to 40% of patients), but these adverse effects are not always discussed at patient review. Most of these effects may be mitigated by simple strategies, such as:

  • Initiating medication at a low dose

  • Titrating slowly

  • Taking medication with food, to avoid nausea or gastric discomfort

  • Taking sedating medications at bedtime and alerting medication in the morning.

Uncommon but more concerning adverse events include:

  • Activation

  • Bipolar switching

  • Suicidal thoughts and behaviours.

Clinicians need to be aware of activation and bipolar switching. A careful review of family history, first-degree relatives' response to medication, previous medication history, and concurrent use of other medications for potential drug-to-drug interactions may better prepare the patient, family, and clinician to avoid negative consequences.

Suicidal thoughts and behaviours are part of the presentation of depression, and may occur prior to or during treatment. Although there is insufficient evidence to support suggestions that antidepressants could cause suicidal ideation and behaviour, both the FDA analysis and a re-analysis of all the controlled trials of antidepressant therapy have indicated an increased risk of suicidal ideation but not suicide attempt with antidepressant treatment versus placebo. The risk is relatively small, comprising only a 1% to 2% increased risk of suicidal ideation with antidepressant use (3% to 4%) compared with placebo (2%).[78][107] Eleven times more youth who are treated with an antidepressant will respond to the medication than will develop suicidal behaviour.[108] It is recommended that children should be monitored closely during the early weeks of initiating antidepressant treatment and during dose adjustments. Rating scales may be used to assess and monitor adverse effects and adverse events during depression treatment, such as:

  • Safety Monitoring Uniform Report Form[109]

  • Toronto Side Effects Scale[110]

  • Liverpool University Neuroleptic Side Effects Scale[111]

  • Mental Health Therapeutic Outcomes Tool.[112]

Children need to be monitored closely with a more frequent follow-up schedule when initiating a new treatment, or during dose change, as adverse effects and adverse events are more likely to occur during those periods. A minimum of 1-2 surgery visits every 4 weeks during the initial months of antidepressant treatment is required.

Moderate or severe depression: inadequate improvement with first SSRI

If, after 8 weeks of treatment with an SSRI at an adequate dose (either as a monotherapy or combined with specific psychotherapy), there is no response (no change in depression severity or functioning impairment), or only partial response (less than a significant reduction of depression severity or improvement of functioning), switching to another SSRI is recommended. Choices of a second SSRI include fluoxetine, sertraline, citalopram, and escitalopram (escitalopram is for adolescents only), depending on which was used initially; addition of CBT along with a switch will result in the best outcome.[113]

Moderate or severe depression: management of inadequate response to second SSRI by switching to a non-SSRI

If there is an inadequate response after the second SSRI, then the medication can be switched to a non-SSRI, or the SSRI could be augmented. Current evidence to support switching to a non-SSRI or augmenting is quite limited. Before initiating this step, a careful reassessment is important to verify the diagnosis and to rule out other contributing factors, such as unrecognised or newly emergent comorbid illness (e.g., substance abuse), inadequacy of psychosocial intervention, unresolved stress, or a new trauma. If there is no indication of these factors as the cause of treatment resistance, and if a second SSRI has produced minimal to no response, switching to a non-SSRI is recommended. These agents should only be commenced by specialists who are experts in managing depression in childhood. Choices for switching include:

  • Venlafaxine

  • Duloxetine

  • Mirtazapine

  • Bupropion.

SNRIs seem to be more beneficial than placebo for treating children and adolescents with depression, although the effect size is small. One meta-analysis found that children and adolescents treated with SNRIs reported a greater incidence of adverse effects than those treated with placebo, including serious adverse effects such as suicidality, emphasising the need for a careful and individualised cost-benefit analysis prior to considering treatment.[90]


An SNRI. To assess the efficacy of this treatment, 3 controlled studies have been conducted, and all were negative. However, when the data from 2 studies was pooled together, the adolescent subgroup showed that venlafaxine was superior compared with placebo in reducing depression.[114][115] Venlafaxine is not recommended as a first-line choice. However, the only controlled trial for treatment-resistant depression indicated that venlafaxine was effective in treating depression in adolescents who did not respond to one SSRI.[116] In the FDA assessment, compared with other antidepressants, venlafaxine did have significantly more suicide-related adverse events than placebo.[107] A retrospective cohort study of 36,842 patients aged 6 to 18 years showed no evidence that risk of suicide attempts differed for SSRI and SNRI antidepressants.[117] Extended-release formula is recommended, which still has a relatively short half-life of 10.3 hours.


Also an SNRI. A negative trial has been published.[118] Unlike venlafaxine, which does not significantly influence the P450 enzyme system, duloxetine is a potent inhibitor of 1A2 and 2D6. It has a similarly short half-life of 12.5 hours.


A serotonin receptor 2 (5-HT2) antagonist, with a half-life of 20-40 hours. The only 2 multicentre paediatric mirtazapine trials were negative, possibly due to the high placebo response in the 2 studies.[119] The drug does not inhibit any P450 enzymes. At lower dose, mirtazapine has a strong antihistamine effect, and may also cause sedation and weight gain.


A potent 2D6 inhibitor with a half-life of 21 hours for the sustained-release formulation. There have been no controlled trials for bupropion, although it is frequently prescribed.[91]


There is one positive trial but the medication is not marketed for children due to hepatotoxicity (though risk is low).

Moderate or severe depression: management of inadequate response to second SSRI by augmentation strategies

An alternative approach to switching to a non-SSRI medication is to use augmentation strategies, especially if there is a partial response to the second SSRI. Psychotherapy (e.g., CBT or IPT) or medication may be used to augment.

In adults, lithium,[120][121] triiodothyronine,[122][123] atypical antipsychotic medications,[124][125] and bupropion[126] have been studied most frequently, with some indications of efficacy. Atypical antipsychotics and bupropion have been used more frequently in the paediatric population as augmenting agents compared with other agents. However, paediatric controlled studies have not been done. These agents should only be commenced by specialists who are experts in managing depression in childhood. The only study of paediatric treatment-resistant depression found that adolescents who had more than 9 CBT sessions in addition to treatment with a second SSRI or venlafaxine were more likely to have a positive response than adolescents who received fewer CBT sessions.[127]

Bupropion is one of the more frequently used augmenting agents, although no paediatric trials have been conducted on it. Sustained-release bupropion may be used to augment.

Atypical antipsychotics have been used clinically to augment antidepressant effect. Quetiapine, aripiprazole, and ziprasidone are used more frequently in clinical settings, because risperidone and olanzapine may cause significant weight gain and metabolic effects. No controlled trials have been conducted in depressed children, although a chart review of 10 cases indicated the efficacy of adding quetiapine to treat depressed adolescents who have not responded to an adequate trial of an SSRI. For all other atypical antipsychotics except ziprasidone, weight, lipid profile, and fasting glucose need to be monitored.

Lamotrigine has been studied in the management of unipolar depression and bipolar depression, though studies are limited. In one study that included chart reviews of 42 adolescent patients with treatment-resistant depression, 22 showed improvement with lamotrigine.[128] More studies have been completed in bipolar depression.

Lithium has been studied as an augmenting agent most frequently in adults, with more than 10 controlled trials. In most of the studies, lithium was used as an augmenting agent for tricyclic antidepressants (TCAs). Only 2 open paediatric trials also used lithium to augment a TCA.[129][130] Blood levels need to be monitored to avoid toxicity. Thyroid and renal function also need to be monitored regularly.

Levothyroxine has been studied, but its efficacy is inconclusive according to adult depression data. Neither controlled nor open trials have been done in the paediatric population. Thyroid-stimulating hormone levels need to be monitored to avoid negative biofeedback.

Novel alternative approaches for children with treatment-resistant depression

If response remains poor, despite all of the possible treatments outlined up to this phase, novel and alternative treatments may be considered. These should only be commenced by specialists who are experts in managing depression in childhood. They include:

  • Other antidepressants (e.g., TCAs and monoamine oxidase inhibitors [MAOIs])

  • Biological treatments (e.g., light therapy and electroconvulsive therapy [ECT])

  • Emerging therapies (e.g., the transdermal selegiline patch), as part of a clinical trial.

Other antidepressants, such as TCAs or MAOIs, may be used for children and adolescents who have not responded to SSRIs or non-SSRIs. TCAs have not proved to be effective in treating paediatric depression, and tend to produce more adverse events.[131] [ Cochrane Clinical Answers logo ] Because of the adverse effects and the difficulty of managing diet in children and adolescents, MAOIs have not been recommended for use in paediatric depression. However, the patch form of a selective MAOI, selegiline, may bypass the concern and become an alternative treatment for young people with depression resistant to other treatment. A study comparing the selegiline patch and placebo in adolescents with major depressive disorder demonstrated safety of the active medication, although response rates were similar for both groups (58.6% versus 59.3%).[132]

Biological treatments include light therapy and ECT. Light therapy is recommended for seasonal affective disorder (SAD), and its efficacy has been demonstrated in a few case series and one controlled study in young people with SAD.[133] A recent review and meta-analysis indicates that light therapy may be effective for non-seasonal depression in adults.[134]

Case reports on the efficacy of ECT in paediatric depression have appeared for more than 60 years.[135] However, there have not been any controlled trials conducted in the paediatric population. Series reports indicate that the best response is in youth with catatonia, psychosis, and bipolar depression. There is a negative perception regarding ECT in some countries, including in several US states, where the use of ECT in children and adolescents has been banned. This, and the fact that generally medication and psychotherapy are readily available, has led to infrequent use of ECT as a treatment modality, even in children and young people with treatment-resistant depression.

Complementary and alternative medicine treatment (CAM)

Although CAM is used frequently around the world for treating paediatric depression, including in the US, there is extremely limited empirical evidence for its efficacy.[136] The 2005 UK NICE guideline opposes the use of St. John's wort in children, due to insufficient evidence of efficacy and known drug interactions. Other guidelines (AACAP, CMAP, GLAD-PC, Australian NHMRC) have not discussed the use of natural remedies in the management of paediatric depression.

Among the natural remedies, the most frequently used and studied remedies for depression include:

  • St. John's wort

  • Omega-3 fatty acids

  • S-adenosyl methionine (SAMe).

St. John's wort is the most frequently used herbal remedy for depression. It is the number one antidepressant prescribed for children in Germany.[137] There are many adult-controlled studies that indicate inconsistent efficacy in treating adult depression.[138][139] No controlled studies have been conducted in depressed children. Several open-label studies indicate that St. John's wort is well tolerated by children, and that it is effective in treating depression in children.[140][141][142] However, it is not recommended to treat moderate to severe cases of depression, due to unclear efficacy. St. John’s wort has also been associated with longer coagulation time; in addition, it increases metabolism of contraceptives and can result in unwanted pregnancy. Caution is necessary with concurrent use of other medications, due to potential drug-to-drug interactions. This is a particular concern with concurrent use of other antidepressants, because of the risk of serotonergic syndrome.

Omega-3 fatty acids are suggested to be beneficial in many health problems. A combination of eicosapentaenoic acid and docosahexaenoic acid from fish oil, rather than omega-3 fatty acids from plants, is effective. Results from adult depression studies indicate benefit in reducing depression.[143] Only one small controlled study of depressed children has been done. This demonstrated that 70% of children who received 1000 mg daily dose of omega-3 fish oil, versus 0% of children who received placebo, had a reduction in their depression severity.[144] Fish oil is well tolerated in general, but high doses are not recommended due to inhibition of platelet aggregation and concern for potential bleeding. Fish oil may also interact with anticoagulants, so it is not recommended for use concurrently with those drugs. There is a concern about contamination of fish oil from heavy metals and pesticides. Algae omega-3 may be a purer alternative source of omega-3 fatty acids compared with fish oil.

SAMe is important in the synthesis of neurotransmitters, such as serotonin, noradrenaline (norepinephrine), and dopamine. Adult depression trials of SAMe indicated superior efficacy to placebo and comparable efficacy to TCAs.[145] SAMe may also be effective as an augmenting agent.[146] No paediatric depression studies are available, except for case reports that indicate efficacy in treating depression in children and adolescents.[147] Interactions of SAMe with other medications appear to be infrequent.[139]

Use of this content is subject to our disclaimer