1. The Diagnosis and Treatment of Pediatric Depression
Jess P. Shatkin, MD, MPH
Vice Chair for Education
NYU Child Study Center
New York University School of Medicine

2. Presentation Outline We will review the following:
Pediatric Depression Disease State
Medication Treatment
Tricyclic Antidepressants
Serotonin Specific Reuptake Inhibitors
Other antidepressants
Augmenting medications & ECT
Psychotherapy Treatment
Manualized therapies (CBT, IPT)
Psychodynamic psychotherapy
Current FDA Controversy
Evidence based treatment recommendations

3. Learning Objectives Participants will be able to:
Describe the primary differences between pediatric and adult depression.
Identify evidence-based pharmacologic and non-pharmacologic treatments for pediatric depression.
Make rational treatment recommendations for children and adolescents with depression.

4. A Brief History of Depression in Children and Adolescents
Case reports on childhood depression date to the early 17th century
Melancholia in children was first reported in the mid-19th century
In general, however, the existence of depression prior to 1960 was seriously doubted because it was felt that children’s immature superego would not permit the development of depression
Research from Europe and NIMH funded American studies in the 1970’s increased the awareness & acceptance of childhood depression

5. Psychoanalytic Perspective
Psychoanalytic theory posits that depression results from an intrapsychic conflict between the ego and a persecutory superego
Psychoanalysis held that the superego was formalized only after resolution of the Oedipal Conflict, which occurred by late adolescence
By this theory, then, children could not experience intrapsychic conflict and, ergo, could not develop mood disorders

6. Epidemiology
Varying rates have been reported; no large, well accepted epidemiologic studies
Generally accepted 1-year incidence is:
*Preschool age 1%
*School age 2%
*Adolescent age 4 - 8%
Gender ratio of 1:1 in childhood and 2:1 (female to male) by adolescence
Lifetime prevalence of MDD among adolescents is 15 – 20% (similar to adults); 15.3% per NCS
Kashani & Sherman, 1988; Fleming & Offord, 1990; Lewinsohn et al, 1993 & 1994; Kessler & Walters, 1998

7. Epidemiology (2)
Studies on Dysthymia suggest a wide range in point prevalence: children from 0.6 – 1.7%; and adolescents from 1.6 – 8.0%
Garrison et al, 1992; Kashani et al, 1987; Lewinsohn et al 1993 & 1994
Studies in specialized populations show increased incidence, such as 40% among children on neurology wards with unexplained headaches (Ling et al, 1970); 7% of general pediatric inpatients (Kashani et al, 1981); 28% of children in psychiatric clinics (Carlson & Cantwell, 1980); 59% of child psychiatry inpatients (Petti, 1978); and 27% of adolescent inpatients (Robbins et al, 1982)

8. Epidemiology (3)
Prevalence increases during adolescence, possibly due to:
Biological factors (e.g., sexual maturation)
Environmental factors (e.g., increased social/academic expectations, more chance of exposure to negative events)
Psychological & cognitive factors (e.g., increased autonomy and abstract thinking)
Since 1940, each successive generation has been at higher risk for MDD

9. Ordinary People
Conrad Jarrett is 16 and has survived a boating accident in which his brother, Buck, died.
The book takes place as Conrad tries to readjust to life back at home after a psychiatric hospitalization for a suicide attempt.
He sees a psychiatrist, Dr. Berger, who tries to address Conrad’s chief desire upon presenting (e.g., he wants to be in control).
This scene takes place as Berger is asking Conrad about quitting the swim team.

10. Etiology: Theories of Depression
Psychodynamic: anger turned inward; severe superego
Attachment: insecure early attachment
Behavioral: inability to obtain reinforcement
Cognitive: depressive mindset
Self-Control: deficits in self-monitoring, self-evaluation, and self-reinforcement
Interpersonal: characteristic to individual, roles and events
Socioenvironmental: stressful life circumstances exacerbate vulnerabilities
Neurobiological: neurochemical, endocrine, and receptor abnormalities

12. Genetics (1)
Findings from twin studies suggest a moderate genetic influence on depression in community samples with heritability ranging from 35 – 75% across studies (Eley, 1999; Glowinski et al, 2003)
Twin/adoption studies have not been conducted, so the extent to which clinical depression in children and adolescents is genetically driven is not known
Still, children with a parent who suffered from depression as a child are up to 14x more likely than controls to become depressed prior to age 13 (Weissman et al, 1988)
Children of parents with depression have about 2-3x the risk of having depression (Beardslee et al, 1998; Weissman et al, 1997)

13. Genetics (2)
Children of depressed parents have an earlier age of onset for their depression by 3 years (Weissman et al, 1997)
The lifetime history of MDD in mothers of children with MDD is also high, about 50 – 75% (Kovacs, 1997)
A family history of depression is more common in 1st degree relatives of children with MDD than in children without MDD (Wickramaratne et al, 2000)
Adults with one or two copies of the short allele of the 5-HT Transporter gene have been shown to exhibit more depressive symptoms, diagnosable depression, and suicidality in relation to stressful life events (Caspi et al 2003)

14. More about Mothers
A 20-year follow-up of offspring of depressed and non-depressed parents found that the risks for anxiety disorders, major depression, and substance dependence were ~3x higher in the offspring of depressed parents vs. non-depressed parents; social impairment was also greater. The time of greatest incidence was 15 – 20 y/o; higher rates of medical problems and mortality in the offspring of depressed parents were beginning to emerge as the offspring enter middle age (Weissman et al, 2006)

15. Even More about Mothers
Effective treatment of mothers with MDD is associated with a reduction of anxiety, depressive, and disruptive disorders and symptoms in their offspring (Weissman et al, 2006; STAR*D trial which follows 151 mother/child pairs at 3 month intervals)
Remission of maternal depression after 3 months of medication treatment was significantly associated reductions in children’s diagnoses and symptoms; diagnoses dropped by 11% vs. an 8% increase in diagnoses among those children whose mothers did not respond to medication treatment
Of the children with MDE at baseline, remission occurred in 33% of those whose mother’s depression remitted vs. 12% of those whose mother’s depression did not remit
17% of children (without dx at baseline) whose mothers remained depressed developed a psychiatric diagnosis at 3 month f/u vs. none of those whose mothers responded to treatment

16. And Fathers?
Medical Expenditure Panel Survey data (N = 22K US children, ages 5 – 17) of both mothers and fathers; data generated by parents only.
Scales used: PHQ-2, SF-12, MCS, CIS, PCS
Risks of child emotional/behavioral problems are much greater if mothers, rather than fathers, have such problems:
Paternal MH problems are independently associated with a 33 – 70% increased risk
Maternal MH problems are associated with a % increased risk
Weitzman et al, 2011
PHQ-2 = patient health questionnaire – 2 (abbreviated self-admin form of primary care eval of mental disorders to evaluate depressive symptoms in parents)
SF-12 = Short-Form – 12  MCS and PCS are from the SF-12
MCS = Mental Component Score (parent scale of mental health problems)
PCS = Physical Component Scale (parent scale of physical problems)
CIS = Columbia Impairment Scale (measures parent report of mental health impairment in the children)
Note: Epidemiological survey study, not clinician admin interviews

17. Serotonin Gene
Among those with pervasive suicidal thoughts and intent, levels of the major serotonin metabolite (5-HIAA) are lower in the cerebrospinal fluid.
Adults with one or two copies of the short allele of the 5-HT Transporter gene have been shown to exhibit more depressive symptoms, diagnosable depression, and suicidality in relation to stressful life events (Caspi et al 2003)

18. Biogenic Amine Hypothesis
The biogenic amine or catecholamine hypothesis suggests that too much neurotransmitter causes mania and too little causes depression.
Too simplistic, but supported by the observation that medications that increase dopamine, norepinephrine, and serotonin improve depression and worsen mania.
Many limitations to this hypothesis, including the fact that L-dopa and tryptophan, direct precursors of amines, have no effect on mood; and cocaine and amphetamines which block amine reuptake do not generally improve depression.

19. Neuroendocrine Markers
70% of adults do not show normal suppression of cortisol secretion following administration of dexamethasone (DST), suggesting an alteration in stress response.
Blunting of normal growth hormone release in response to insulin challenge.
Blunted production of thyroid stimulating hormone in response to thyroid releasing hormone

20. Annie Hall
Woody Allen’s 1976 breakthrough film about his relationships with women
Won the Academy Award for Best Picture
Early in the film he reflects upon his childhood in Brooklyn, which is filled with exaggerations of how he “remembers” his childhood
He also demonstrates the increased abstract thinking which can sometimes overwhelm children as they enter adolescence

22. Clinical Presentation
DSM-IV Criteria do not differ for children & adolescents
Generally, children show fewer neurovegetative signs than adults
Irritability may substitute for depressed mood

23. Diagnosis (1) The DSM-IV requires 5 of 9 symptoms for the diagnosis
At least two straight weeks in duration with symptoms present pretty much every day or most of every day
Not better accounted for by another illness
MDE = Major Depressive Episode
MDD = Major Depressive Disorder (2 or more episodes, lifetime)
Specifiers
Severity, psychosis, chronic, atypical, postpartum, melancholic

24. Diagnosis (2) Depressed Mood
*in children, can substitute “irritable” mood
Anhaedonia (diminished interest & pleasure)
Significant decrease in weight (5%)
*in children, failure to make expected weight gains
Insomnia or hypersomnia
Psychomotor agitation or retardation
Fatigue or loss of energy
Feelings of worthlessness or excessive guilt
Diminished ability to think or concentrate or indecisiveness
Recurrent thoughts of death or suicidal ideation

25. Developmental Variants of MDD
Adolescents:
More cognitive components to their depression than children
Guilt and hopelessness become apparent
More sleep & appetite disturbances, delusions, suicidal ideation & attempts
Compared to adults, still more behavior problems and fewer neurovegetative difficulties
Children:
More symptoms of anxiety (e.g., phobias, separation anxiety), somatic complaints, and auditory hallucinations
Depression is expressed as temper tantrums & behavior problems
Fewer delusions and serious suicide attempts
By middle childhood, preoccupations w/death, lowered self-esteem, social withdrawal/rejection, & poor school performance

26. Clinical Variants of MDD
Unipolar Depression
Psychotic Depression
Bipolar Depression
Atypical Depression
Seasonal Affective Disorder
Subclinical or subsyndromal Depression
Treatment-Resistant Depression

27. Comorbidities
Most children with MDD have a comorbid psychiatric diagnosis:
*40 – 90% have a second psychiatric disorder
*20 – 50% have two or more comorbid disorders
Dysthymia and Anxiety Disorders (30 – 80%); Disruptive Disorders (10 – 80%); and Substance Use Disorders (20 – 30%)
MDD usually manifests after the onset of other psychiatric disorders, except substance abuse
Conduct problems may develop secondary to depression and persist after the depression is effectively treated
Separation anxiety is more common in children, whereas SUDS, conduct disorder, social phobia, and GAD are more common in adolescents
-Birmaher et al, 1996; Goodyer et al, 1997; Kovacs, 1996; Rohde et al, 1991; Biederman et al, 1995; Weissman et al, 1997

29. Thirteen
Tells the story of Tracy, who is a straight-laced, geeky, 13 y/o A student growing up in LA with her brother, Mason.
Her divorced mother is a recovering drug addict living with her former cocaine addict boyfriend; her absent & generally unsuccessful father is struggling with earning enough money to support the kids and develop something for himself.
She becomes friends with Evie, a cool kid, by acting out and as teen stress mounts in her life, she begins to cut to cope.
In this scene, her father, who has not been paying close attention, struggles to figure out what’s going on with his daughter.

31. Differential Diagnosis
Adjustment Disorder with Depressed Mood
Bereavement
General Medical Conditions (e.g., hypothyroidism, cancer, lupus, anemia, HIV, diabetes, epilepsy, etc.)
Chronic Fatigue Syndrome
Medication induced (stimulants, neuroleptics, corticosteroids, contraceptives)

32. Differential Diagnosis (2): Nonaffective Psychiatric D/O’s
Anxiety Disorders
ADHD
Externalizing Disorders
Learning Disorders
SUDS
Eating Disorders
Personality Disorders
Premenstrual Dysphoric Disorder

33. Clinical Course
Median duration: Clinically referred: 7 – 9 months; community: 1 – 2 months
Predictors of increased duration: depression severity, comorbidity, negative life events, parental psychiatric disorders, poor psychosocial functioning
90% of MDD episodes remit w/in 1-2 years after onset (where remission is 2 weeks – 2 months with only 1 clinically significant symptom)
≈ 50% relapse
6 – 10% of MDD are protracted
Clarke et al, 1992; Goodyer et al, 1997; Kovacs, 1996; Lewinsohn et al, 1994 & 1997; Reinecke et al, 1998; Sanford et al, 1995; Warner et al, 1992

34. Relapse Relapse is an episode of MDD during a period of remission
40 – 60% of youth with MDD experience relapse after successful treatment of acute episode (indicates the need for continual treatment)
Predictors of relapse: natural course of MDD, lack of compliance, negative life events, rapid decrease/discontinuation of therapeutic treatment
Emslie et al, 1997; Kovacs, 1996; Lewinsohn et al, 1994; Vostanis et al, 1996; Wood et al, 1996

35. Recurrence
Recurrence is the emergence of MDD symptoms during a period of recovery (asymptomatic period of more than 2 months)
Clinical & community samples show probability of recurrence 20 – 60% w/in 1-2 years post-remission and 70% after 5 years
Predictors of Recurrence:
Earlier age at onset
Increased number of prior episodes
Severity of initial episode
Psychosis
Psychosocial stressors
Dysthymia & other comorbidities
Treatment noncompliance
Emslie et al, 1997; Kovacs et al, 1996 & 1997; Lewinsohn et al, 1994

36. Risk of Bipolar Disorder
20 – 40% of depressed children & adolescents develop bipolar disorder within 5 years of index episode of MDD
Predictors of Bipolar I Disorder onset:
Early onset MDD
Psychomotor retardation
Psychotic features
Family history of bipolar disorder
Family history of psychotic depression
Heavy familial loading for mood disorders
Pharmacologically induced (hypo)mania
Geller & Luby, 1997; Kovacs, 1996; Strober & Carlson, 1982

37. Treatment
Opinions vary as to whether one should start with psychotherapy or medication or both
Psycho-education of patient, family, and teachers is critical
Parental (and other family members’) mental health issues should be addressed
Certainly, the least restrictive treatment and setting should be a starting point

38. Typical Exclusion Criteria for Pediatric Depression Studies
ADHD
PTSD
Bipolar Disorder
Pervasive Developmental Disorders
Mental Retardation
Externalizing Disorders
Psychosis
Any recent medication treatment (within 2-4 weeks)
EtOH/drugs
Eating Disorder
Recent initiation of psychotherapy
Potentially suicidal patients (attempts in past year)

39. What’s in a Study?
The “gold standard” for any type of clinical intervention study (medication, therapy, community intervention, etc.) is that it be:
Randomized
Double-Blind
Blinded to subject
Blinded to treatment team
Placebo Controlled

40. Research Study Instruments
Kids are often not very good informants about their own mood state
They often underestimate medication effects and side effects
As a result various rating scales and surveys have been designed to assess their responses to treatment
Some of these are clinician administered:
Children’s Depression Rating Scale (most commonly used)
Clinical Global Impression
Children’s Global Assessment Scale
Hamilton Depression Rating Scale
Some of these are child self-administered:
Children’s Depression Inventory
Beck Depression Inventory

41. Placebo Effect Typically very high in most medications
Studies of antidepressants in both children and adults reveal approximately a 30% placebo rate
Overall response rates to antidepressants are about 65% at highest; consequently, about half of that is due to placebo
True antidepressant response rate is about 35%
Remember, the average length of a depressive episode (not chronic) is 6 – 9 months with or without treatment

42. Tricyclic Antidepressants (TCAs): History
The TCA story begins with the synthesis of chlorpromazine in 1950 from synthetic antihistamines first produced in the 1940s. Chlorpromazine was thought to be an antihistamine, but in 1952 it was found to have profound psychiatric effects. By 1955 chlorpromazine was widely used and rapidly revolutionized the world of inpatient psychiatry as the first effective antipsychotic.
Imipramine, the first TCA, is an analogue of chlorpromazine, which was not designed for the treatment of depression but rather for psychosis. The drug's tendency to induce manic effects (and generally worsening psychosis in schizophrenics), however, was noted, and the paradoxical observation of a sedative inducing mania lead to testing with depressed patients. The first trial of imipramine took place in 1955, and the first report of its antidepressant effects was published in 1957.
Merck introduced the second member of the TCA family, amitriptyline (Elavil), in 1961.

43. Tricyclic Antidepressants
The “original” antidepressants
Examples:
Desipramine
Amitriptylene
Imipramine
Clomipramine
Nortriptylene
Putative Mechanism of Action: Block the reuptake of norepinephrine, dopamine, and serotonin by neuronal presynaptic receptors
Unfortunately, while effective for adult depression, they have shown little utility in the treatment of pediatric depression

44. TCA Mechanism of Action

45. Tricyclic Antidepressants (2)
Open trials of TCAs have found that 60 – 80% of depressed children and 44 – 75% of depressed adolescents respond positively
At least 11 randomized DBPC trials each demonstrated no difference between placebo and active TCA treatment (5 in adolescents, 6 in children)
Dulcan et al, 1998; Ryan & Varma, 1998
One study (Preskorn, 1987; n = 22) of depressed children (ages 6 – 14) treated with imipramine was positive
Meta-Analysis (Hazell et al, 1995) found no effect

46. Tricyclic Antidepressants (3)
Problems with child & adolescent TCA studies:
Small sample sizes
Diagnostic heterogeneity (e.g., mild, mod, severe depression) & included patients with secondary depression (higher placebo response)
Studies of limited duration (6 – 8 weeks)
Lower doses were used because of cardiac safety concerns
Noradrenergic (secondary amines) TCAs were exclusively used (receptors not fully developed in children); except imipramine study
High prevalence of comorbid conditions
More adolescents transition into Bipolar D/O than adults (and BP depression may be less responsive to TCAs)
More efficient hepatic metabolism of drugs in children

47. Amines
Tricyclics are sometimes classified as secondary or tertiary amines. In general, the tertiary amines boost serotonin as well as nor-epinephrine (adrenergic) and produce more sedation, anticholinergic effects, and orthostatic hypotension. The secondary amines act primarily on nor-epinephrine and tend to have a lower side-effect profile.
Tertiary amines include: Amitriptyline, imipramine, trimipramine, doxepin, clomipramine, and lofepramine.
Secondary amines include: Nortriptyline, desipramine, protriptyline, and amoxapine.

48. More About Amines Desipramine (Secondary) Amitriptyline(Tertiary)
Amines are organic compounds whose functional group contains a nitrogen atom with a lone pair of electrons.
A primary amine has one of the 3 hydrogen atoms replaced by a carbon group.
A secondary amine has 2 hydrogen atoms replaced by carbon groups.
A tertiary amine has 3 hydrogen atoms replaced by carbon groups.
Desipramine (Secondary)
Amitriptyline(Tertiary)

49. Clinical Use Depression Anxiety (particularly serotonergic TCAs) ADHD
Analgesia
Migraine headache prevention
Neuropathic pain (PNS)
Enuresis

50. TCA Side Effects
Most common TCA side effects are related to antimuscarinic (anticholinergic) activity, including:
Dry mouth (salivary secretion is affected)
Dry nose
Blurred vision (accommodation in the eye is affected)
Decreased gastro-intestinal motility and secretion (constipation)
Urinary retention or difficulty with urination
Hyperthermia
Tolerance to these adverse effects often develops if treatment is continued
Side effects may also be less troublesome if treatment is initiated with low dose and then gradually increased, although this may delay the clinical effect.
Other side effects may include drowsiness, anxiety, restlessness, cognitive and memory difficulties, confusion, dizziness, akathisia, increased appetite with weight gain, sweating, decrease in sexual ability and desire, muscle twitches, weakness, nausea and vomiting, hypotension, tachycardia, and rarely irregular heart rhythm.

51. TCA Side Effects

52. Safety Concerns with TCAs
Concern related to at least 8 reported cases of “sudden” death in children and adolescents using TCAs for the treatment of depression
QT prolongation and subsequent torsade de pointes is the suggested cause of death
Level of risk remains unclear

53. Monoamine Oxidase Inhibitors (MAOI)
Monoamine oxidase inhibitors (MAOIs) are a class of powerful antidepressants
They work by decreasing the function of monoamine oxidase, an intracellular enzyme which metabolizes neurotransmitters
Due to potentially lethal dietary and drug interactions, MAOIs had been reserved as a last line of defense, used only when other classes of antidepressant drugs have been tried unsuccessfully.
Recently, however, a patch form of the drug selegiline (Emsam) was developed (2006).
When applied transdermally the drug does not enter the gastro-intestinal system as it does when taken orally, thereby decreasing the dangers of dietary interactions associated with MAOI pills.

54. MAOIs Continued Isocarboxazid (Marplan) Phenelzine (Nardil)
Tranylcypromine (Parnate)
Selegiline (Eldepryl & Emsam)

55. Clinical Use
In the past MAOIs were prescribed for those resistant to TCA therapy, but newer MAOIs are now sometimes used as first-line therapy.
Depression
Social Anxiety
Smoking Cessation
Atypical Depression
NO data in children & adolescents

56. Side Effects
Hypertensive crisis (when foods containing tyramine are consumed) or hyperserotonemia (if foods containing tryptophan are consumed). MAO typically degrades these products.
Assumed that tyramine displaces norepinephrine from the storage vesicles, which may trigger a cascade in which excessive amounts of norepinephrine can lead to a hypertensive crisis.
Examples of foods and drinks with potentially high levels of tyramine include fermented substances, such as red chianti and other aged red wines and aged cheeses.
The most significant risk associated with the use of MAOIs is the potential for interactions with over-the-counter and prescription medicines, illicit drugs and certain supplements (e.g. St. John’s Wort).
MAOIs should not be combined with other psychoactive substances (antidepressants, illicit drugs, painkillers, stimulants, etc.) except under expert care.

57. Serotonin Specific Reuptake Inhibitors (SSRIs)
The “new” antidepressants; much safer and easier to prescribe and tolerate
Examples:
Fluoxetine (Prozac®)
Sertraline (Zoloft®)
Paroxetine (Paxil®)
Citalopram (Celexa®)
Fluvoxamine (Luvox®)
Escitalopram (Lexapro®)
Putative Mechanism of Action: Block the reuptake of serotonin by neuronal presynaptic receptors
Very useful for pediatric anxiety disorders; generally less effective (but often useful) for pediatric depression

58. Clinical Use Depression Anxiety Eating Disorders (especially Bulimia)
OCD, Panic, Social, Generalized Anxiety
Eating Disorders (especially Bulimia)
Chronic Pain
Premature Ejaculation

59. Chemical Structure
SSRIs may look different from one another, but all of them block the reuptake of serotonin in the synapse between two neurons
Lexapro and Prozac

60. SSRI Mechanism of Action

61. Serotonin Specific Reuptake Inhibitors
Numerous open label studies report a 70 – 90% response rate to SSRIs in adolescents
Ambrosini et al, 1999; Apter et al, 1994; Masi et al, 1997; McConville et al, 1996; Rey-Sanchez & Gutierrez-Casares, 1997; Rodriguez-Ramos et al, 1996; Simeon et al, 1998
Simeon et al (1990) performed the first randomized DBPC study of SSRIs in 32 adolescents (13 – 18 y/o) using 60 mg fluoxetine (Prozac®) vs. placebo
Rating scales included Ham-D and CGI
Results did not reach clinical significance
One historical case-control study (Strober et al, 1999) found fluoxetine superior to imipramine in a severely ill inpatient adolescent population

62. SSRIs (2): Fluoxetine (Prozac®)
Two randomized DBPC studies by Emslie et al demonstrated the superiority of fluoxetine (Prozac®) over placebo, leading to FDA approval of fluoxetine for the treatment of pediatric depression (ages 7 – 17):
1997 Single Site Study (sponsored by NIMH)
*n = 90, 8-week study, nonpsychotic MDD, 20 mg of fluoxetine vs. placebo; CDRS-R & CGI
56% (fluoxetine) vs. 33% (placebo) showed improvement on CGI; significant differences in weekly CDRS-R also noted (fluoxetine vs. placebo); no difference in complete symptom remission
2002 Multisite Study (sponsored by Eli Lilly)
*n = 219, 9-week study, nonpsychotic MDD, 20 mg of fluoxetine vs. placebo; CDRS-R & CGI
52% (fluoxetine) vs. 37% (placebo) showed improvement on CGI; greater mean improvement on fluoxetine by week #1 (and maintained through study) on CDRS-R; remission rates 41% (fluoxetine) vs. 20% (placebo)

63. SSRIs (3): Paroxetine (Paxil®)
One recognized favorable open label study (Rey-Sanchez & Gutierrez-Casares, 1997)
Keller et al (2001) performed a multisite randomized DBPC trial of paroxetine in 275 adolescents (12 – 18 y/o) vs. imipramine vs. placebo (sponsored by GSK)
A priori primary outcomes (all not significant) included:
Ham-D score ≤8 or a >50% reduction
Statistically significant change in mean Ham-D score
Post hoc analysis of primary and secondary outcomes (statistically significant) included:
Revised Ham-D outcome to ≤8 “only”
Depression item sub-scores on Ham-D and K-SADS-L
CGI (65.6% for paroxetine vs. 52.1% for imipramine and 48.3% for placebo)

64. SSRIs (4): Paroxetine (Paxil®) cnt’d
Berard et al (2006) reported on a prospective international multicenter, RDBPC trial of paroxetine for adolescents (13 – 18 y/o) with depression
Data collected from 286 adolescents in 10 countries (not USA)
Response rate (at least 50% reduction from baseline) for paroxetine vs. placebo was not statistically significant for Montgomery-Asberg Depression Rating Scale (MADRS) scale nor K-SADS
CGI was statistically better for paroxetine than placebo (69% vs. 57%), a secondary endpoint, with older adolescents generally doing better than younger (>16). Generally tolerated well with a greater incidence of suicidal behavior (4.4% vs. 2.1%) in the paroxetine treated group, but not statistically significant

65. SSRIs (5): Paroxetine (Paxil®) cnt’d
Emslie et al (2006) completed a randomized, multicenter, double-blind, placebo-controlled trial of Paroxetine
206 patients, aged 7 to 17 years old with major depressive disorder, received paroxetine (10-50 mg/day) or placebo for 8 weeks from 2000 to 2001
The primary efficacy measure was change from baseline in the Children's Depression Rating Scale-Revised total score at week 8 (LOCF)
Safety was primarily assessed by spontaneous reporting of adverse events
104 patients received paroxetine vs 102 who received placebo
CDRS-R total score adjusted mean changes from baseline for patients receiving paroxetine and placebo were (SE 1.47) and points (SE 1.60), respectively (0.80, 95% confidence interval to 4.69, p = 0.684); thus, paroxetine was not shown to be more efficacious than placebo
Side Effects included increased cough (5.9% versus 2.9%), dyspepsia (5.9% versus 2.9%), vomiting (5.9% versus 2.0%), and dizziness (5.0% versus 1.0%). The incidence of adverse events of suicidal behavior and/or ideation while taking study medication (excluding taper) was 1.92% (2/104) for paroxetine versus 0.98% (1/102) for placebo.

66. SSRIs (6): Sertraline (Zoloft®)
One recognized favorable multicenter open label study (Ambrosini et al, 1999)
Wagner et al (2003) reported on two multisite-international separate controlled trials; data were aggregated (sponsored by Pfizer)
N = 376; age range 6 – 17 years
Primary outcome measures were mean change from baseline in CDRS-R, and CGI & CGAS
Changes in mean CDRS-R & CGI between drug & placebo were significant
Based on a 40% decrease in adjusted CDRS-R total score at study endpoint, 69% vs. 59% were responders
The treatment effect was only noted for adolescents (when broken down by age groups)
When the trials are considered separately, no effect was noted, possibly due to very high placebo rates (59% for CDRS, 53% for CGI)

67. SSRIs (7): Citalopram (Celexa®)
Chart review by Bostic et al (1997) at CMHC of 21 adolescents showed favorable results on CGI by 76% of patients
Multisite DBPC study by Wagner et al (2004) randomly assigned 178 children and adolescents (7 – 17 y/o) to 20 – 40 mg/d citalopram or placebo for 8 weeks
Primary outcome measure was CDRS-R; secondary measure included CGI
Statistically significant improvement on the CDRS-R was noted by week #1; by week #8 36% of citalopram-treated patients vs. 24% of placebo patients demonstrated a statistically significant treatment response
CGI results were not significant (47% vs. 45%)

68. SSRIs (8): Citalopram (Celexa®)
Von Knorring et al (2006) reported on a randomized, double-blind, multisite (Europe) placebo-controlled study of citalopram in adolescents with major depressive disorder
244 adolescents, 13 to 18 years old, with major depression were randomized to treatment with citalopram (n = 124) or placebo (n = 120)
No significant differences in improvement of scores from baseline to week 12 between citalopram and placebo were found. The response rate was 59% to 61% in both groups according to the K-SADS and Montgomery Asberg Depression Rating Scale (MADRS)
Remission (MADRS score < or =12) was achieved by 51% of patients with citalopram and 53% with placebo.
A post hoc analysis revealed that more than two thirds of all patients received psychotherapy during this study. For those patients not receiving psychotherapy, there was a higher percentage of Kiddie-SADS-P responders with citalopram (41%) versus placebo (25%) and a significantly higher percentage of MADRS responders and remitters with citalopram (52% and 45%, respectively) versus placebo (22% and 19%, respectively).
Side effects were mild. Suicide attempts, including suicidal thoughts and tendencies, were reported by 5 patients in the placebo group and by 14 patients in the citalopram group (not significant) with no pattern with respect to duration of treatment, time of onset, or dosage. In contrast, the suicidal ideation (Kiddie-SADS-P) single item showed worsening more frequently in the placebo (18%) than in the citalopram group (8%).

69. SSRIs (9): Escitalopram (Lexapro®)
A RDBPC trial by Wagner et al (2006) examined efficacy of escitalopram in 131 children and adolescents (6 – 17 y/o dosed flexibly 10 – 20 mg/d) vs. 133 treated with placebo
82% of patients completed treatment with no major AEs (HA & GI pain more common in active treatment group) and no induction of SI/SA
Active treatment did not statistically separate from placebo at endpoint by CDRS-R with LOCF
Post-hoc analysis of adolescent completers (12 – 17 y/o) did statistically separate active drug from placebo by CDRS-R

70. SSRIs (10): Escitalopram (Lexapro®)
8 week RDBPC trial of 10 – 20 mg escitalopram per day in adolescents 12 – 17 y/o
155 active treatment, 157 on placebo (n=312)
Statistically significant separation between drug and placebo at end of trial (LOCF, 83% completion rate) with a 22.1 point reduction in CDRS-R on active treatment versus 18.8 point reduction on placebo (p = 0.22), Effect Size = 0.27 (Emslie et al, 2009)
16-week double-blind extension of Emslie study found that statistical separation was maintained for escitalopram treated group (Findling et al, 2008)
The FDA review concluded that maintenance efficacy could be extrapolated from data in adults, although not systemically evaluated in adolescents.

71. Side Effects
When present, most notable during the first 1-4 weeks while the body adapts to the drug (with the exception of sexual side effects, which tend to occur later in treatment but often improve with time). Almost all SSRIs are known to cause one or more of these symptoms:
nausea, vomiting, diarrhea
drowsiness
headache
clenching of teeth
extremely vivid and strange dreams
dizziness
changes in appetite
weight loss/gain (measured by a change in bodyweight of 7 pounds)
decreased sexual interest and/or anorgasmia
increased feelings of depression and anxiety
tremors
Autonomic dysfunction including orthostatic hypotension or sweating
Akathisia
hyponatremia
liver or renal impairment
suicidal ideation
photosensitivity (increased risk of sunburn)

72. Bupropion (Wellbutrin®)
Daviss et al (2001) treated 24 adolescents (11 – 16 y/o) w/ADHD and either MDD or Dysthymia in an open label fashion with buproprion SR
After a 1-2 week single-blind lead in, all subjects were dosed with buproprion SR to a target dose of 3 mg/kg BID for up to 10 weeks
Clinician rating was the CGI
30% improvement during placebo lead in, followed by an 88% improvement in clinician rated depression by CGI

73. Venlafaxine (Effexor®)
Mandoki et al (1997) treated 33 children & adolescents in a randomized DBPC fashion for 6-weeks (8-17 y/o) with MDD with either venlafaxine plus CBT or placebo plus CBT. The dose in the 8-12 year olds was 37.5 mg/d whereas in the year olds was 75 mg/d
Rating scales included HAMD for those 12+ y/o and CDRS for those < 12 y/o, parent ratings (CBCL) and patient ratings (CDI)
Improvement noted in many subjects, but results were not statistically significant
The authors suggest that the negative findings are due to low dosages, high rates of hepatic metabolism in pediatric populations, short duration of treatment, and the fact that CBT may have distorted any medication effect
CBT was beneficial regardless of active medication treatment

74. Venlafaxine (Effexor®) cont’d
Emslie et al (2007) reported on the use of Venlafaxine ER in two multicenter, randomized, placebo-controlled trials in children and adolescents, ages 7 – 17 years, with MDD conducted between October 1997 and August 2001
Patients received venlafaxine ER (flexible dose, based on body weight; intent to treat, n = 169) or placebo (intent to treat, n = 165) for up to 8 weeks. The primary measure was the change from baseline in the CDRS-R at week 8
There were no statistically significant differences between venlafaxine ER and placebo on the CDRS-R. A post hoc age subgroup analysis of the pooled data showed greater improvement on the CDRS-R w/venlafaxine ER than with placebo (-24.4 versus -19.9; p = .022) among adolescents (ages 12-17), but not among children (ages 7-11).
The most common adverse events were anorexia and abdominal pain. Hostility and suicide-related events were more common in venlafaxine ER-treated participants than in placebo-treated participants. There were no completed suicides.

75. Nefazodone (Serzone®)
Wilens et al (1997) reported on 7 cases of children & adolescents (average age = 12 y/o) with depression (4 with BP depression) who took nefazodone for an average of 13 (±8) weeks at dosages averaging 350 mg/d
56% of adolescents were “much” or “very much” improved on CGI
2 of the 4 BP patients did well and 2 experienced mild manic activation
Findling et al (2000) studied 23 youth (7 – 17 y/o) in a 8-week open label fashion to explore the pharmacokinetics of nefazodone
Statistically significant improvements were noted on the CDRS-R, CGI, and CGAS
Pharmacokinetics were variable, but the medication appeared “safe”

76. Mirtazapine (Remeron®)
One published study; a multicenter open label study of mirtazapine in adolescents (12 – 18 y/o) with MDD (Haapasalo-Pesu et al, 2004); n = 24
Rating scales included Ham-D, BDI, & CGI
Doses of mirtazapine varied from 30 – 45 mg/d
Statistically significant improvement noted on all rating scales (Ham-D = 78%; CGI = 74%)

77. Currently FDA Approved Antidepressants & Indications
Major Depressive Disorder
Fluoxetine (Prozac®) 8 – 17 y/o
Escitalopram (Lexapro®) 12 – 17 y/o
Obsessive Compulsive Disorder
Fluoxetine (Prozac®) 7 – 17 y/o
Sertraline (Zoloft®) 6 – 17 y/o
Fluvoxamine (Luvox®) 8 – 17 y/o
Clomipramine (Anafranil®) 11 – 17 y/o

78. Antidepressant Augmentation
Lithium:
Strober et al (1992) examined the effect of LiCO3 augmentation (300 mg TID) on imipramine in 24 treatment refractory adolescent MDD (DSM-III or DSM-III-R) patients in a 3-week open label trial. Mild beneficial effects noted
Walter et al (1998) noted effective LiCO3 augmentation of venlafaxine (Effexor XR®) in two adolescent cases
Ryan et al (1988) found LiCO3 augmentation in adolescents with a partial response to imipramine effective in a chart review

79. Electroconvulsive Therapy
Case reports in children and adolescents dating to 1942; most cases suffer from lack of diagnostic clarity, small samples, and heterogeneous diagnoses
Since 1990 numerous studies (all retrospective) have reported success with ECT in adolescents with a variety of psychiatric disorders (but primarily unipolar or bipolar mood disorders)
Response rates vary from 51 – 100% in these studies, with higher response rates noted among those with mood disorders (Ghaziuddin et al, 2004)
Only one study (Kutcher & Robertson, 1995) compared treated patients with those who refused treatment
Significant improvements noted among those who received ECT
Treated patients had shorter hospital stays (74 vs. 176 days)

80. ECT (2) Use estimates vary worldwide Safety
NIMH Study (Thompson & Blaine, 1987) revealed about 1.5% of all ECT performed in 1980 in the USA (or about 500 cases) were between 11 – 20 y/o
No mandatory reporting system currently exists
Safety
Guttmacher & Cretella (1988) noted that ECT was ineffective in 4 cases and that prolonged seizures (>4 minutes) were caused
This finding has not been replicated; all other studies have found ECT effective and with no greater side effects than those routinely found in adult studies

81. Algorithm for Treatment of Depression in Children and Adolescents
Fluoxetine*
Alternate SSRI or SNRI*
TCA*
MAOI
ECT
*May augment with: lithium, T3, stimulant, buspar, pindolol, antipsychotic, 2nd antidepressant, benzodiazepine

82. Psychotherapy Studies
7 of 9 studies indicate that CBT is more efficacious than a wait-list condition or than a non-CBT alternative psychotherapy (Curry, 2001)
Harrington et al’s (1998) systematic review of CBT in depressed children & adolescents indicated a beneficial effect in 62% of treated patients vs. 36% in placebo groups
CBT is associated with more rapid remission of symptoms than is family or supportive therapy (Brent et al, 1997)
Long term follow-up indicates high rates of remission or recovery among adolescents with MDD but no superiority of CBT over other psychotherapies (Birmaher et al, 2000)
No single type of CBT has been shown to be more efficacious than any other
IPT has been shown more efficacious than a wait-list condition or minimal clinical management in two acute treatment studies (Mufson et al, 1999; Rossello & Bernal, 1999)

83. Ordinary People
Based on a novel by Judith Guest about an affluent family’s painful adjustment to tragedy, Mary Tyler Moore and Donald Sutherland play a seemingly happy couple who lose the older of their two sons in a boating accident.
Robert Redford’s Oscar winning directorial debut, and Tim Hutton’s film debut in 1980
After Tim Hutton, the younger son, tries to kill himself, he is sent to a psychiatrist, Judd Hirsch

85. New Data
NIMH sponsored “Treatment of Adolescents with Depression Study” (TADS)
Multicenter controlled clinical trial
12 – 17 y/o with MDD
Aims to compare the efficacy of fluoxetine, CBT, combination, and placebo at 36 weeks with 1 year f/u
NIMH sponsored “Treatment of Resistant Depression in Adolescents” (TORDIA)
12 – 17 y/o treatment resistant adolescents
Aims to compare the efficacy of fluoxetine, paroxetine, or venlafaxine, either alone or in combination with CBT for 24 weeks with 1 year f/u

86. Medication + Therapy: The TADS Study
Multisite study of adolescents, aged 12 – 17 y/o; n = 439; tested for short (12 weeks) and longer term (36 weeks) effectiveness with durability (1 year naturalistic follow-up) [March et al, 2004]
Participants were randomly assigned to fluoxetine alone (10 – 40 mg/d), CBT alone, fluoxetine with CBT, or placebo; medications blinded, all CBT conditions unblinded
Rating scales included CDRS-R and CGI
Rates of response on the CDRS-R indicate that combined treatment (fluoxetine + CBT) was statistically superior to fluoxetine alone and CBT alone
Fluoxetine alone was superior to CBT alone, which did not separate from placebo
Rates of response on CGI for fluoxetine + CBT (71%), fluoxetine alone (61%), CBT alone (43%), and placebo (35%)

87. TADS (2) Effect sizes at week 12 on the CDRS-R:
Combined = 0.98; Fluoxetine = 0.68; CBT = -0.03
Rates of Remission (<28 on CDRS-R):
Total by week 12 = 23% (37% COMB, 23% FLX, 16% CBT, 17% PBO)
Total by week 36 = 60% (60% COMB, 55% FLX, 64% CBT)
By 36 week extension, CBT had “caught up” with fluoxetine and response rates were 69% for fluoxetine and 65% for CBT
Combined CBT + fluoxetine reached maximum benefit at week 18 (85% response rate), 3 months earlier than CBT or fluoxetine alone (all Rx converged at week 36, with med + CBT at 86%, med and CBT alone each at 81%)
Younger, less chronically depressed, higher functioning, less hopeless w/less SI, fewer melancholic features and comorbid dx, and those with greater expectations for improvement were more likely to benefit from treatment
24 suicide related events occurred in the 12 week study; only fluoxetine had more suicide related events than placebo; 5 total attempts; no suicide completion

88. TADS (3)
Treatment consisted of 3 stages: (1) acute (12 weeks), (2) continuation (6 weeks more to 18th week), and (3) maintenance (18 week to 36th week)
242 FLX, CBT, and COMB patients in their assigned treatment at the end of stage 1 were included in this study
Stage 2 treatment varied based on stage 1 response. Stage 3 consisted of 3 CBT and/or pharmacotherapy sessions and, if applicable, continued medication
Sustained response was defined as 2 consecutive Clinical Global Impression-Improvement ratings of 1 or 2 ("full response")
Among 95 patients (39.3%) who had not achieved sustained response by week 12 (29.1% COMB, 32.5% FLX, and 57.9% CBT), sustained response rates during stages 2 and 3 were 80.0% COMB, 61.5% FLX, and 77.3% CBT (difference not significant)
Among the remaining 147 patients (60.7%) who achieved sustained response by week 12, CBT patients were more likely than FLX patients to maintain sustained response through week 36 (96.9% vs 74.1%; P = .007; 88.5% of COMB patients maintained sustained response through week 36)
Total rates of sustained response by week 36 were 88.4% COMB, 82.5% FLX, and 75.0% CBT
Thus, most adolescents with depression who had not achieved sustained response during acute treatment did achieve that level of improvement during continuation and maintenance therapies
Rohde et al, 2008

89. TADS (4) 196 patients followed 5 years out
96% eventually recovered from the index episode of MDE within 3.5 years
Nearly half (46%) of those who recovered from MDE became depressed again within 5 years, regardless of the initial treatment they received
Girls were more likely than boys to have a repeat episode (58% versus 33%)
Those with an anxiety disorder were also more likely to have a recurrence (62% versus 42%)
Curry et al, 2010

90. TORDIA
RCT of 334 patients 12 – 18 years with MDD who had not responded to 2 months of initial treatment with SSRI (CGI of 2 or less & 50% decrease on CDRS)
12 weeks of:
Switch to a second SSRI (Paxil, Celexa, Prozac)
Switch to a second SSRI + CBT
Switch to Effexor ( mg)
Switch to Effexor + CBT
CBT + a switch to either medication regimen showed a higher response rate (55%) than med switch alone (41%)
No difference in response rate between a second SSRI and Effexor; more SEfx with Effexor (BP, rash)
No differential effects on self harm or SI
Less severe depression, less family conflict, and absence of NS-SIB predicted better treatment response. COMB treatment was more evident among youths who had more comorbid disorders (esp ADHD and anxiety disorders), no abuse history, and less hopelessness.

91. TORDIA (2) Patients were reassessed 48 and 72 weeks from intake
Remission defined as ≥ 3 weeks with ≤ 1 clinically significant symptom and no associated functional impairment and relapse as ≥ 2 weeks with probable or definite depressive disorder
By 72 weeks, 61% had reached remission
The group assigned to an SSRI had a more rapid decline in self-reported depressive symptoms and SI than those assigned to venlafaxine (p>0.3)
Those with more severe depression, greater dysfunction, and EtOH or drug use at baseline were less likely to remit
Remitters diverged from nonremitters by 6 weeks of treatment
Of 130 in remission by week 24, 25% relapsed within the next year
Summary: Most achieved remission but more than 1/3 did not and ¼ of remitters experienced a relapse
Vitiello et al, 2010

92. ADAPT Adolescent Depression Antidepressant and Psychotherapy Trial
RDBPC trial of 208 adolescents, 11 – 17 y/o (74% female), at six outpatient clinics in England
Fluoxetine + CBT or Fluoxetine alone
10 mg/d x 1 week, 20 mg/d x 5 weeks; if no response by week 6, 40 mg/d; if no response by week 12, 60 mg/d
CBT delivered for 12 weeks plus 1 session week 28
Groups did not differ at F/U at 12 and 28 weeks
Goodyer et al, 2007

93. CBT + Sertaline
Melvin et al (2006) evaluated the effects of CBT and sertraline, alone and in combination, for 73 adolescents (12 – 18 years) in the Netherlands
Diagnoses included MDD, DD, or Dep NOS
Randomly assigned to one of 3 treatments (med, CBT, combined) at 3 community clinics
Measures included the Reynolds Adolescent Depression Scale, Revised Children’s Manifest Anxiety Scale, Suicidal Ideation Questionnaire, self report and CGI
CBT demonstrated a superior acute treatment response to sertraline at 12 weeks (OR = 6.86, CI 1.12 – 41.82), but all groups showed improvement maintained at 6 months; of those w/MDD 71% achieved partial remission (CBT = 71%, Med = 33%, Comb = 47%)
Medication doses were lower than in prior sertraline studies and a slow titration schedule was utilized

94. Black Box Warning

95. Recent FDA Antidepressant Controversy
9 drugs included in FDA review (fluoxetine, sertraline, paroxetine, fluvoxamine, citalopram, bupropion, venlafaxine, nefazodone, mirtazapine)
Approximately 4400 patients
25 placebo controlled studies (ranging from 4 – 16 weeks in duration):
16 in MDD
4 in OCD
2 in GAD
1 in social anxiety disorder
2 in ADHD

96. Recent FDA Antidepressant Controversy (2)
Pooled analyses of these studies found an excess of SI and attempts noted in children and adolescents taking antidepressants (roughly 4% in those taking medication vs. 2% in those taking placebo)
No suicides occurred in these trials
FDA could not rule out an increased risk of suicidality for any of these medications
Data was adequate to establish effectiveness in MDD only for fluoxetine based upon Emslie et al’s two studies
Black Box Warning to apply to all antidepressants

97. FDA Recommended Guidelines
“After starting an antidepressant, your child should generally see his/her healthcare provider:
Once a week for the first 4 weeks
Every 2 weeks for the next 4 weeks
After taking the antidepressant for 12 weeks
After 12 weeks, follow your healthcare provider’s advice about how often to come back
More often if problems or questions arise”
FDA Medication guide (rev 1/26/05)

98. How Real is the Concern?
12.5% (11 of 88) adolescents enrolled in a week psychotherapy for depression trial (randomly assigned to CBT, systemic behavioral family therapy, or nondirective supportive therapy) reported suicidality at some point during treatment (no meds used) even though they denied suicidality on initial intake interview. 
The detection of suicidality was improved by specific and systemmatic assessment, whereas in prior clinical trials of depression adverse events were reported by patients or observed.  Self-reported suicidality in the week prior to intake predicted the onset of emergent suicidality to a much greater extent than did interview-rated suicidality, treatment assignment, cognitive distortions, and depression severity.
Bridge et al 2005:  Emergent Suicidality in a Clinical Psychotherapy Trial for Adolescent Depression

99. The Reality Is…
The vast majority of teen suicide completers are not on an SSRI at the time of the event
The risk of suicide increases greatly for those with chronic MDD, as opposed to those suffering a single MDE.

100. The FDA and Adult Suicide
The FDA has recently reported that antidepressants double the risk of suicidal behavior in young adults (18 – 25 years) from about 0.25% among adults who took placebos to 0.5% among adults who took an antidepressant
The analysis found no increased risk of completed suicides in patients taking the medications
The risk appears to decline with age

101. Antidepressant Sales
Prescriptions for antidepressants have dropped by 20% for those 18 y/o and younger since 2004 when FDA initial warnings were published
After concerns were raised in the Netherlands about the suicide risk, there was a 22 percent drop from 2003 to 2005 in antidepressant prescriptions for patients under 18 years and a corresponding 50 percent increase in suicides (the number of suicides increased from 34 to 51)
Sales of antidepressants among adults were down 14% in 2005
Sales are climbing again in 2006

102. Antidepressant Sales & Suicide Rates
Recent data suggest that antidepressant sales of Prozac, amongst others, may be tied to a decrease in overall suicide rate
The U.S. suicide rate held fairly steady for 15 years from 1973 – 1988 at 12.2 – 13.7/100,000
Subsequent to the introduction of Prozac in 1988, the suicide rate gradually declined to 10.4/100,000 in This drop is associated with an increase in Prozac prescriptions from 2.47 million in 1988 to million in 2002.
Adopting this data, the decrease in suicides totals 33,600 (Licinio et al, 2006)

103. Uh Oh…
From 2003 to 2004, the suicide rate among Americans younger than 19 years rose 18 percent, the most dramatic one-year change since the government started collecting suicide statistics in Between 2003 and 2005 the number of SSRI prescriptions for pediatric depression (ages 5 – 18) dropped by more than 50%.
The rise followed a sharp decrease in the prescribing of SSRI antidepressants after placement of the Black Box Warning.
The data suggest that for every 20 percent decline in antidepressant use among patients of all ages in the United States, an additional 3,040 suicides per year would occur. About 32,000 Americans commit suicide each year.
If the drugs were causing a high rate of suicide, then the reduction in prescriptions should have caused a decrease in the suicide rate.
The study included the Netherlands, which had a 22 percent decrease in antidepressant use among children between 2003 and The suicide rate among youngsters there increased by 49 percent in that period.
Gibbons, 2007

104. TADS Lessons About Suicide
Remember that 24 suicide related events occurred in the 12 week study; only fluoxetine had more suicide related events than placebo; 5 total attempts; no suicide completion
Findings across 36 weeks demonstrates that patients treated with FLX alone were twice as likely as patients treated with combined FLX + CBT or CBT alone to show both clinically significant SI (on patient report) and to experience treatment emergent suicide events (on clinician report): FLX 14.7%, COMB 8.4%, and CBT alone 6.3%
Thus: (1) There is no increased risk of a suicide event with CBT; (2) there is a protective effect on suicidality from adding CBT to medication

105. The Other Side of Summer
After a decade long decline in the suicide rate among youths, 10 – 19 y/o, there was an 18% increase in suicide rates in 2004, a trend that persisted in 2005
This analysis shows 326 more deaths than expected in 2004 and 292 more deaths than expected in 2005
Bridge et al, 2008

107. Evidence Based Treatment Recommendations: Take Home Points
When treating pediatric depression, start with either CBT and/or medication
If medication is employed, you should seriously consider using fluoxetine first line
You need to develop a monitoring strategy to comply with FDA guidelines