Texas Children's Medication Algorithm Project: Update From Texas Consensus Conference Panel on Medication Treatment of Childhood Major Depressive Disorder

SPECIAL COMMUNICATION

Texas Children_s Medication Algorithm Project: Update From Texas Consensus Conference Panel on Medication Treatment of Childhood Major Depressive Disorder CARROLL W. HUGHES, PH.D., GRAHAM J. EMSLIE, M.D., M. LYNN CRISMON, PHARM.D., KELLY POSNER, PH.D., BORIS BIRMAHER, M.D., NEAL RYAN, M.D., PETER JENSEN, M.D., JOHN CURRY, PH.D., BENEDETTO VITIELLO, M.D., MOLLY LOPEZ, PH.D., STEVE P. SHON, M.D., STEVEN R. PLISZKA, M.D., MADHUKAR H. TRIVEDI, M.D., AND THE TEXAS CONSENSUS CONFERENCE PANEL ON MEDICATION TREATMENT OF CHILDHOOD MAJOR DEPRESSIVE DISORDER ABSTRACT Objective: To revise and update consensus guidelines for medication treatment algorithms for childhood major depressive disorder based on new scientific evidence and expert clinical consensus when evidence is lacking. Method: A consensus conference was held January 13Y14, 2005, that included academic clinicians and researchers, practicing clinicians, administrators, consumers, and families. The focus was to review, update, and incorporate the most current data to inform and recommend specific pharmacological approaches and clinical guidance for treatment of major depressive disorder in children and adolescents. Results: Consensually agreed on medication algorithms for major depression (with and without psychosis) and comorbid attention-deficit disorders were updated. These revised algorithms also incorporated approaches to address issues of suicidality, aggression, and irritability. Stages 1, 2, and 3 of the algorithm consist of selective serotonin reuptake inhibitor and norepinephrine serotonin reuptake inhibitor medications whose use is supported by controlled, acute clinical trials and clinical experience. Recent studies provide support that selective serotonin reuptake inhibitors in addition to fluoxetine are still encouraged as first-line interventions. The need for additional assessments, precautions, and monitoring is emphasized, as well as continuation and maintenance treatment. Conclusions: Evidence and expert clinical consensus support the use of selected antidepressants in the treatment of depression in youths. The use of the recommended antidepressant medications requires appropriate monitoring of suicidality and potential adverse effects and consideration of other evidence-based treatment alternatives such as cognitive behavioral therapies. J. Am. Acad. Child Adolesc. Psychiatry, 2007;46(6):667Y686. Key Words: medication algorithm, major depressive disorder, suicidality, childhood psychopharmacology, child and adolescent depression.

Accepted January 1, 2007. Drs. Hughes, Emslie, and Trivedi are with the Department of Psychiatry, University of Texas Southwestern Medical Center, Dallas; Dr. Crismon is with the College of Pharmacy, University of Texas at Austin; Dr. Posner is with Columbia University, New York; Drs. Birmaher and Ryan are with the Western Psychiatric Institute and Clinic, Pittsburgh; Dr. Jensen is with the Center for the Advancement of Children_s Mental Health, Department of Psychiatry, Columbia University, and the Office of Mental Health, New York; Dr. Curry is with Duke University, Durham, NC; Dr. Vitiello is with the National Institute of Mental Health, Bethesda, MD; Drs. Lopez and Shon are with the Texas Department of State Health Services, Austin; and Dr. Pliszka is with the Department of Psychiatry, University of Texas Health Science Center at San Antonio. The Texas Consensus Conference Panel is coordinated and funded by the Texas Department of State Health Services. No pharmaceutical company

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:6, JUNE 2007

provided support or was in any way involved with this consensus conference. The Texas Department of State Health Services, the Department of Psychiatry, University of Texas Southwestern Medical Center, and the University of Texas at Austin College of Pharmacy receive funding from the Texas State Legislature. The opinions and assertions contained in this report are the private views of the authors and are not to be construed as official or as reflecting the views of the National Institute of Mental Health, the National Institutes of Health, or the U.S. Department of Health and Human Services. Correspondence to Dr. Carroll W. Hughes, Department of Psychiatry, University of Texas Southwestern Medical Center, Dallas, 5323 Harry Hines Blvd., Dallas, TX 75390-8589; e-mail: [email protected]. 0890-8567/07/4606-06672007 by the American Academy of Child and Adolescent Psychiatry. DOI: 10.1097/chi.0b013e31804a859b

667

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

HUGHES ET AL.

A second consensus conference for childhood major depressive disorder (MDD) was held January 13Y14, 2005, in Dallas, TX, to revise and update the medication algorithms for childhood depression (Hughes et al., 1999). It was deemed important to incorporate new research findings, address recent controversial issues regarding the use of antidepressants in the pediatric population, U.S. Food and Drug Administration (FDA) warnings about possible increased suicidality with antidepressant use in children and adolescents, and data from a recent feasibility trial for the childhood MDD algorithm. The latter study indicated clear improvement in clinical symptoms, functioning, and global response when the guidelines were implemented in Texas community mental health settings in comparison to a historical cohort (Emslie et al., 2004b). Treatment algorithms for adults as a component of disease management have been successfully completed in a large-scale comparative trial in adults with MDD in the public sector (Rush et al., 2006; Trivedi et al., 2004). The rationale and methods for developing algorithms to guide clinical practice and their importance for improving clinical outcomes have been reviewed elsewhere (Crismon et al., 1999; Gilbert et al., 1997; Rush and Kupfer, 1995; Rush et al., 1999). Recent reviews of the efficacy and safety of antidepressants in depressed youths summarize the findings for all of the recent pediatric antidepressant studies including both published and unpublished data (Cheung et al., 2005; Hammad et al., 2006; Mann et al., 2006). The reviews focus on the results of double-blind, randomized, placebo-controlled, acute clinical trials included in the recent FDA analyses (autumn 2004) of the efficacy (Cheung et al., 2005; Mann et al., 2006) and safety (Hammad et al., 2006) of selective serotonin reuptake inhibitors (SSRIs) and non-SSRI antidepressants (nefazodone, venlafaxine, and mirtazapine) for major depression in youths. Lengthy reanalyses of all of the suicide-related data from randomized, controlled acute trials (RCTs) including a re-examination of the rating scales that had been used, resulted in the FDA issuing a box warning in October 2004 (Cheung et al., 2005; Hammad et al., 2006). The warning described an increased risk of suicidality (suicidal behavior and ideation) for all antidepressants used in individuals under the age of 18. The incidence of suicidal ideations and behaviors in these pooled

668

reanalyses of all RCTs of antidepressants in children was about 4% for those youths receiving antidepressants compared with 2% on placebo (Hammad et al., 2006). It is important to note that no completed suicides were reported in any of these RCTs. Furthermore, methodological limitations of the suicidality data based on post hoc reviews from these RCTs question the strength of the conclusion regarding the causal relationship between suicidality and antidepressant treatment in youths. No systematic assessment of suicidality was performed during the RCTs (Posner et al., 2007a). Only a few prospective treatment studies of adolescent suicidal behavior have been conducted, all underpowered, and none used medication in a standard fashion, despite its administration to most depressed suicidal youths in clinical practice. The first multisite pilot treatment study of adolescents who have recently attempted suicide is in progress (National Institute of Mental [NIMH] Treatment of Adolescent Suicide Attempters [TASA]). Contrary to the FDA_s conclusions on childhood suicidality, epidemiological studies have found an inverse relationship between suicides in the community and prescription rates of SSRI antidepressants for both youths and adults (Gibbons et al., 2005, 2006; Olfson et al., 2003, 2005). Other pharmacological studies have found no association between SSRIs and suicide after controlling for factors that increase the risk of suicide (Ludwig and Marcotte, 2005; Valuck et al., 2004). Other analyses of large medical claim databases have found higher rate of suicide among antidepressant users (Olfson et al., 2006). However, it is important to keep in mind that risk of suicide has been shown to be greatest in the month before commencing antidepressant treatment and continues to decrease after initiating treatment (Simon et al., 2006). A recent FDA meta-analysis of placebo-controlled clinical trials indicated that the risk for suicidality during antidepressant treatment is higher than on placebo until age 25 but that past that age the risk for suicidality for those taking antidepressants becomes less than when taking placebo declines afterward to become actually favorable to antidepressant in older age (Food and Drug Administration, 2006; Stone, 2006). Toxicology studies of completed suicides have found that most teenagers, even those who had been prescribed an antidepressant, did not take antidepressants before the suicide (Gray et al., 2003; Isacsson et al., 2005; Isacsson and Rich, 2005; Leon et al., 2004).

J. AM. ACAD. CHILD ADOLESC. PSYCH IATRY, 46:6, JUNE 2007

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

CHILDHOOD MDD MEDICATION ALGORITHM

Disconcertingly, since the introduction of FDA_s warnings, SSRI prescription rates have declined by 25% (Nemeroff, 2007; Rosack, 2005). Given the welldocumented inverse relationship between suicide and antidepressant prescription rates, there has been increasing concern that these recent effects on prescribing practices would negatively affect completed suicide rates in youths. Recent Centers for Disease Control and Prevention data revealed a 14% increase in suicide rates among 5- to 19-year-olds between 2003 (the introduction of FDA warnings) and 2004 (Centers for Disease Control and Prevention, 2006). This represents by far the largest increase in youth completed suicide rates since the Centers for Disease Control and Prevention began documenting rates in 1979 (R. Gibbons, Ph.D., personal communication, 2007). In addition to reductions in prescription rates, rates of MDD diagnoses in the general pediatric population have also declined significantly since the FDA initiated warnings (Valuck, 2006). Medication algorithms for childhood MDD have been incorporated as pharmacotherapy guidelines for a number of ongoing NIMH-sponsored trials in both children and adolescents (e.g., TASA; Vitiello, 2006) and adults (Fava et al., 2003; Rush et al., 2006; Trivedi et al., 2004, 2006; Yates et al., 2004). Clinicians, administrators, and patients and families have endorsed the use of these treatment algorithms recently. Initial concerns regarding the applicability of the algorithms, concern that treatment algorithms would limit clinical judgment and professional identity as well as concerns about liability have not been borne out (Emslie et al., 2004b). Algorithm recommendations meld available research evidence with expert academic opinion and that of practicing clinicians. When the first childhood MDD algorithm was published (Hughes et al., 1999), placebo-controlled trials of SSRIs in youths were limited to one positive acute study with fluoxetine (Emslie et al., 1997a). The initial double-blind study finding fluoxetine superior to placebo in children and adolescents with MDD (Emslie et al., 1997a) has been replicated in one multisite RCT and supported by the results from the Treatment for Adolescents With Depression Study (TADS; Emslie et al., 2002; Treatment for Adolescents With Depression Study [TADS] Team, 2004). A number of additional RCTs in more than 2,500 child and adolescent patients have been conducted since the original study. A number of issues related to the

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:6, JUNE 2007

treatment of depression in youths, including FDA hearings on antidepressants and suicidality, have arisen. The revised childhood MDD algorithm incorporates the new research findings and addresses FDA warnings about possible increased suicidality with antidepressant use in children and adolescents. METHOD 2005 Consensus Conference Update The Children_s Medication Algorithm Project (CMAP) Depression Consensus Conference Update, sponsored by the Texas Department of State Health Services (DSHS) was held in Dallas, TX, in January 2005. An expert panel consisted of the authors and also included practicing child and adolescent psychiatrists in DSHS facilities and contracted community mental health centers, administrators, and parents/caregivers of depressed youths. The format consisted of a series of presentations of current research evidence and panel discussions, including practitioners within DSHS and parents/caregivers of youths with depression. The consensus panel used previously established rules for evaluating and classifying available evidence (i.e., level A data consist of randomized, double-blind, placebo-controlled studies; level B data consist of open-label and large epidemiological studies; and level C data consist of case series and expert consensus; Crismon et al., 1999; Hughes et al., 1999). The panel (see participant list) critically discussed the available data and drafted the revisions to the treatment algorithms, strategies, and tactics (Hughes et al., 1999). This was followed by multiple postmeeting reviews and revisions to produce the algorithms reported here. General Principles The 10 general principles underlying the algorithm are as follows: 1. Treatment recommendations are based on both empirical science and expert clinical consensus. 2. Careful and accurate diagnosis with careful evaluation of potential comorbid diagnoses is a precursor to treatment. 3. The algorithm is driven by the developmentally unique aspects of depression in children and adolescents. 4. Child- and adolescent-specific assessments and outcome measures are important. 5. Psychoeducation for both caregivers and youths is a critical element of medication treatment, independent of a lack of empirical data. 6. Nonmedication treatment alternatives should be considered and discussed with parents and children. 7. Patient/caregiver preference is important in treatment decisions and the decision about whether to treat with medication, psychotherapy, or both is left to clinician judgment with discussion and consent/assent from the family and patient. 8. The presence of depression in family members as well as family conflict must be addressed and treated (Birmaher et al., 2000; Brent et al., 1998). 9. The role of schools in providing input about a patient_s symptoms and functioning is recognized. 10. Complicated, partial, or nonresponsive cases, whenever possible, should be referred to a child and adolescent

669

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

HUGHES ET AL.

psychiatrist for further assessment and treatment or a second opinion, particularly after initiating adequate trials of the first stage. Reassessment of diagnosis; additional assessments of dose, duration, adherence (Lloyd et al., 1998); and consultations are recommended with any treatment failures. Consultations are particularly important if symptoms persist after an adequate therapeutic trial or the patient has not fully recovered following increased doses and/or medication changes. The algorithm strategies are subdivided into three major phases of treatment: acute, continuation, and maintenance. For acute phase, initial strategies involve implementation of single medications with favorable side effect profiles and level A data. Treatment may progress to the next stage either because of inadequate symptom improvement or medication intolerance. A clinician may choose to begin treatment at any of the stages with a documented rationale (e.g., patient history, family preference). As one progresses down the algorithm sequence, evidence becomes less rigorous, strategies become more complicated, risk of adverse effects may be greater, and closer monitoring by and attention from clinicians are required (Crismon et al., 1999; Rush et al., 1998). RESULTS Updated Algorithm for MDD

The updated algorithm focuses on youths 6 to 17 years of age who meet DSM-IV criteria for MDD and whose depression is deemed to be of sufficient severity to warrant medication (e.g., significant psychosocial impairment in family and peer functioning or school performance and/or at risk of harm to self or others). Insufficient data exist to make guideline recommendations for children younger than 6 years of age (Thomas et al., 1997) and for childhood adjustment disorders with depressed mood. Treatment Strategies

Treatment strategies are divided into a series of stages (Fig. 1) with multiple options within each stage. In addition to the strategies for acute treatment, recommendations are provided for continuation and maintenance treatment. Stage 0: Diagnostic Assessment and Monitoring. The consensus panel identified unique and important aspects of children and adolescents with MDD involving treatment and assessment issues (Vitiello et al., 2003). Therefore, stage 0 summarizes and emphasizes the following areas that need to be addressed before initiating medication treatment in children and adolescents with MDD. These are often addressed over multiple visits before initiating medication. Assessment and monitoring issues. The diagnostic criteria are essentially the same for child or adult MDD,

670

except for the inclusion of either depressed or irritable mood as a key qualifying symptom along with anhedonia. However, the diagnostic process is different and requires synthesizing information from both parents and children separately and evaluation of school performance instead of work. The panel noted that the irritable mood symptom can be mistaken as a symptom of an externalizing disorder or bipolar disorder, and careful assessment is required. Nonspecific treatment intervention. Indirect evidence suggests that depression in children and adolescents is influenced by psychosocial variables such as peers and family as well as environmental events (see Hammen, 1999), and children in particular appear more responsive to nonspecific treatment (e.g., higher placebo response rate in some acute RCTs than in adult trials). This is reflected in the Cheung et al. (2005) review in which the placebo response rates vary from around 35% in the fluoxetine trials to 50% to 60% in other studies. In contrast to earlier reports for children versus adolescents in smaller studies, Mayes et al. (2007) reported in a large study of combined fluoxetine trials that the drug-placebo difference was greater in children compared with adolescents. Although the panel did not specifically recommend a minimum time between initial evaluation and initiating medication (stage 1), the clinician and parent/child should consider this developmental difference in deciding whether to initiate medication at the initial assessment. BWatchful waiting[ (alternatively called Bactive monitoring[ in Guidelines for Adolescent Depression in Primary Care, Cheung et al., in press) with ongoing assessment of clinical status for a few weeks is encouraged when this can be safely done, particularly for preadolescents. Psychoeducation and lifestyle management training may be helpful adjuncts during this period (Cheung et al., 2005; Lopez et al., 2005). Assessment of suicidality. The panel strongly emphasizes not only the importance of initial assessment for suicidality (both present and past) in children and adolescents but also the continued need for monitoring for emerging suicidality and suicide attempts during treatment. For example, history of attempts (e.g., type, lethality) and suicidal ideation are reviewed. Because cognition and energy often improve with pharmacotherapy more rapidly than mood, patients may be at increased risk to act on suicidal ideation during the first few weeks of treatment. Therefore,

J. AM. ACAD. CHILD ADOLESC. PSYCH IATRY, 46:6, JUNE 2007

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

CHILDHOOD MDD MEDICATION ALGORITHM

Fig. 1 Medication algorithm for treating children and adolescents who meet DSM-IV criteria for major depressive disorder. FDA = U.S. Food and Drug Administration. Revised based on Hughes et al. (1999). (This algorithm is in the public domain and may be used and reprinted without special permission. The authors appreciate proper citation. The authors bear no responsibility for the use of these guidelines by third parties.)

careful monitoring and systematic assessment of suicidal thoughts and behavior during medication initiation and dose changes are essential. A helpful tool to track suicidal events throughout any treatment is the Columbia Suicide Severity Rating Scale (available from [email protected]). Medication versus alternative treatment interventions. The decision to initiate medication versus specific psychotherapy is made jointly by the clinician and adequately informed parents (guardians) with assent from the child. Evidence suggests that specific therapies, such as cognitive-behavioral therapy (CBT; Brent et al.,

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:6, JUNE 2007

1997; Clarke et al., 1999) and interpersonal therapy (Mufson et al., 2004) are effective treatments, at least for mild to moderate depression. Specific psychotherapies have shown efficacy with diagnosed depressed adolescents, and CBT significantly reduces depressive symptoms in prepubertal children (Curry, 2001). CBT has yielded response rates similar to those of medication and appears superior to supportive psychotherapy and systemic behavioral family therapy (Birmaher et al., 2000; Brent et al., 1997; Compton et al., 2004). However, in the recent TADS (Treatment for Adolescents With Depression Study [TADS] Team, 2004) of

671

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

HUGHES ET AL.

439 adolescents, CBT did not produce significantly better results than placebo. On the primary categorical outcome measure (responder status based on a final Clinical Global Impressions-Improvement score of 1 or 2), the response rate for combination treatment was the most effective with 71% response, followed by fluoxetine alone (61%), CBT (43%), and placebo (35%). The response rates with combination or fluoxetine were superior to placebo, whereas CBT was not (Treatment for Adolescents With Depression Study [TADS] Team, 2004). Recent additional analyses indicated that at the end of acute TADS treatment (week 12), clinical response was moderated by family income, severity of depression, and cognitive distortions (Curry et al., 2006). CBT alone was as effective as combination treatment for adolescents from higher income/higher parental education families (fluoxetine did not add anything). Conversely, medication alone was as effective as combination treatment for severely depressed adolescents (CBT did not add anything). Teenagers with higher levels of cognitive distortions benefited from adding CBT to medication (combination treatment), whereas those with low levels of cognitive distortion did not benefit (CBT did not add anything). Elements to consider when deciding on medication versus psychotherapy include the severity of the depression, suicidality, how other family members have responded to medication or therapy, recurrence of a depression or its chronicity, lack of response to psychotherapy, family preferences, and consideration of psychosocial stressors. It is important to note that nonspecific therapies (e.g., supportive, psychodynamic, family therapy, psychoeducation) may be appropriate in addition to medication as part of an individualized treatment plan, although no empirical evidence exists to guide the selection of these interventions. No specific recommendations were made about when to use combination therapies versus medication management alone because the focus of the algorithm is medication management. Therapeutic alliance. Stage 0 emphasizes the necessity of a therapeutic relationship between the clinician and the patient and family. A clinician unfamiliar with interviewing children and adolescents cannot conduct medication management. Antidepressant trials in youths often fail because of medication nonadherence (Lloyd et al., 1998), inadequate dosing, or

672

inadequate duration of antidepressant treatment. It is important that the health care delivery system allow for continuity of care by the same clinician whenever possible. Stage 1: SSRIs (Monotherapy). Because medications may differ in their potential benefit and risk profiles and given recent FDA box warnings applying to all antidepressants, parents or guardians should provide informed consent for each separate medication initiated. Entering into stage 1 is based on the clinician and family agreeing that medication is indicated. A patient entering stage 1 is either treatment naı¨ve for the current episode or received an inadequate trial (e.g., inadequate dose or duration, nonalgorithm medication). The recommended monotherapy antidepressants for stage 1 are the SSRIs fluoxetine, sertraline, or citalopram. Three SSRIs have at least one RCT demonstrating efficacy compared with placebo on the a priori determined primary outcome: fluoxetine (Emslie et al., 1997a, 2002; Treatment for Adolescents With Depression Study [TADS] Team, 2004), sertraline (Wagner et al., 2003), and citalopram (Wagner et al., 2004b). One acute trial of citalopram showed greater symptom response compared with placebo (Wagner et al., 2004b). A second study, which was conducted in Europe beginning in 1996, was negative (i.e., unable to distinguish between citalopram and placebo). In the latter study, both inpatients and outpatients were included and more than two thirds of patients were also receiving concurrent psychotherapy (von Knorring et al., 2006). Two identically designed RCTs of sertraline versus placebo were prospectively pooled to gain greater statistical power, and the results were positive for sertraline (Wagner et al., 2003). These two RCTs were reported separately to the FDA to satisfy requirements for a 6-month patent extension, and individually were negative. The FDA requires two positive acute trials for approval of an indication, and the FDA concluded that these studies individually did not support efficacy for sertraline. Because it was prospectively determined that efficacy analyses would be performed on the total sample, the consensus panel accepted this as a positive acute study for sertraline. One acute study of paroxetine failed to demonstrate efficacy on the a priori identified primary outcome, but found efficacy on several secondary outcome measures (Keller et al., 2001). Two other unpublished studies with paroxetine in youths were negative (Berard et al.,

J. AM. ACAD. CHILD ADOLESC. PSYCH IATRY, 46:6, JUNE 2007

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

CHILDHOOD MDD MEDICATION ALGORITHM

2006; Emslie et al., 2006b). Thus, the available evidence provides the greatest support for the efficacy of fluoxetine in treating depression in children and adolescents, and fluoxetine is the only antidepressant to show positive efficacy in more than one acute trial. It is the only antidepressant with FDA-approved labeling for treatment of depression in this population and remains a first-line choice based on a preponderance of evidence (Emslie et al., 1997a,b, 2002, 2006a; Kratochvil et al., 2006; Treatment for Adolescents With Depression Study [TADS] Team, 2004). Several other antidepressants (citalopram, sertraline, paroxetine) show some evidence of efficacy, but results are inconsistent and have not been replicated in other trials. Fluoxetine, as in the original algorithm, remains the antidepressant of choice, unless reasons exist in an individual patient to choose a different SSRI (e.g., potential drug interactions, past poor response, family resistance, prior lack of response with an adequate dose and trial). However, sertraline and citalopram are also alternatives in stage 1 because both have demonstrated efficacy. The pros and cons of selecting the particular SSRI should be reviewed with the family and joint decisions should made. When comparing the SSRIs recommended in stage 1, all of them have relative advantages and disadvantages. The FDA box warning is included in the labeling for all antidepressants. The consensus panel held that adverse effects are not clearly different among SSRIs, and prescribing clinicians should remain sensitive to family preferences. SSRIs should usually be initiated with low doses (e.g., fluoxetine 10 mg/day), titrated upward, and backed down if side effects emerge. Other antidepressants are not recommended for stage 1, given the lack of supporting randomized, placebocontrolled studies of these medications demonstrating efficacy in childhood and adolescent depression. Stage 2: Switching to Alternate SSRI (Monotherapy) (SSRIsVFluoxetine, Sertraline, Citalopram, Escitalopram, or Paroxetine [Adolescents Only If Paroxetine]). Stage 2 is recommended for children who did not experience adequate clinical improvement during stage 1 due to unsatisfactory symptom response or medication intolerance. If depressive symptoms did not adequately improve with the SSRI used in stage 1, then an alternate SSRI is recommended. If the patient is not switching from fluoxetine, then the new antidepressant should usually be cross-tapered

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:6, JUNE 2007

with the initial antidepressant. If the patient experienced intolerable side effects (e.g., nausea, excessive restlessness, agitation) on the SSRI chosen in stage 1, then initiation of an alternate SSRI at a lower dose should be considered. All SSRIs, including escitalopram and paroxetine, are included in stage 2. These two medications were not included in stage 1 because they do not have level A data. However, they are included as additional SSRI options. A recent double-blind study of escitalopram versus placebo in 6- to 17-year-olds did not show a significantly improved Children_s Depression Rating Scale-Revised (CDRS-R) score at endpoint with last observation carried forward analysis (Wagner et al., 2006). However, a post hoc analysis of adolescents only did find a better response to escitalopram than placebo. Efficacy data are mixed with paroxetine. The adolescent study was positive on some secondary outcome variables (Keller et al., 2001); however, two other trials did not demonstrate positive efficacy, even in the adolescent age group (Berard et al., 2006; Emslie et al., 2006b). Level A data do not support the use of paroxetine in preadolescent children (Emslie et al., 2006b), although level B data (open studies) have shown positive efficacy of paroxetine in preadolescents (Rey-Sanchez and Gutierrez-Casares, 1997). In addition, paroxetine appears to be associated with more dropouts caused by adverse effects in prepubertal children (Emslie et al., 2006b), and it may cause more agitation/hostility than other SSRIs (Cheung et al., 2005). It has been suggested that this may be associated with the shorter half-life of paroxetine in children. Therefore, paroxetine is not recommended for prepubertal children. Although paroxetine had a higher mean risk ratio for suicidality in the FDA-supervised analyses of the SSRI RCTs, the 95% confidence interval from its analyses overlapped with those of other antidepressants (Hammad et al., 2006). The panel did not deem the evidence adequate to prove causality and elected to keep paroxetine at stage 2 in the algorithm for adolescents until stronger data exist to make a different recommendation. Stage 2A: SSRI + Augmentation. For children and adolescents showing a partial response to SSRI treatment, stage 2A provides the option of adding an augmenting agent. In children and adolescents only limited data exist to support the efficacy of augmenting agents (Ryan et al., 1988; Strober et al., 1992). No recent double-blind studies of SSRI augmentation in

673

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

HUGHES ET AL.

children and adolescents have been conducted. Augmentation is typically used when a partial response occurs with antidepressant monotherapy or when remission is not achieved despite significant clinical improvement. The potential advantages of augmentation versus switching to another antidepressant monotherapy include no need for discontinuation of initial antidepressant, less lag time for response, partial responders continue to receive treatment without interruption, and treatment of breakthrough symptoms is possible. Most augmentation recommendations are extrapolated from available adult data (Trivedi et al., 2006). Recent RCTs have shown mirtazapine and sustained-release bupropion to be effective augmenting agents in adults who have experienced inadequate improvement with an SSRI, but this has not been studied in children (Carpenter et al., 2002; Trivedi et al., 2006). The best augmenting agent for children and adolescents who fail to respond to SSRIs or are partial responders remains to be determined. Based on adult data and clinical opinion, augmentation may be a useful strategy for youths who have shown initial response with optimal dosing, but who have not achieved remission. Stage 3: Switching to Alternate Antidepressant Monotherapy (Bupropion, Venlafaxine, Mirtazapine, Duloxetine). To move to stage 3, the child should have experienced at least two failed trials of SSRI antidepressants. It is critical to readdress the accuracy of diagnosis, comorbidity, and contributing factors. The adequacy of psychotherapeutic interventions should also be addressed and modified as needed. An alternate antidepressant monotherapy is recommended at stage 3 (bupropion, venlafaxine, mirtazapine, duloxetine). Quitkin et al. (2005) demonstrated in adults that 85% of patients would achieve therapeutic response and about two thirds would achieve remission after three sequential antidepressant trials of adequate dose and duration. Based on adult data, the panel recommends switching to a different class of medication. In contrast to major depression in adults, studies of other newer antidepressants show little evidence of efficacy for the pediatric population. Two acute venlafaxine trials (Emslie et al., 2007) and two acute mirtazapine trials in children and adolescents with MDD have been negative (Cheung et al., 2005, although the pooled data from the venlafaxine trials indicate some efficacy for adolescents, but not for

674

children. (See Cheung et al., 2005, and Mann et al., 2005, 2006 for a thorough discussion of these studies.) A small open-label trial of bupropion in children and adolescents with depression (MDD or dysthymia) and attention-deficit/hyperactivity disorder (ADHD) was conducted (Daviss et al., 2001) that found that 87.5% of subjects met criteria for depression response with bupropion. However, a large trial specific to MDD has not been conducted with bupropion. It should be noted that controlled studies in youths completed on mirtazapine and venlafaxine did not establish efficacy over placebo. One acute trial of nefazodone in adolescents with MDD demonstrated a positive effect of medication over placebo, but only on secondary outcomes. Another acute nefazodone trial was negative. Nefazodone was not included for stage 3 because of a lack of data and potential side effects including hepatotoxicity. No pediatric trials with duloxetine have been performed; however, adult data suggest that it is an effective treatment for depression (Mallinckrodt et al., 2006). No reports are available from controlled studies in youths who have failed to respond to an SSRI, although such a study is nearing completion (Treatment of Resistant Depression in Adolescents [TORDIA], NIMH, David Brent, PI). Meta-analyses of tricyclic antidepressant studies show no statistically significant efficacy in depressed children and adolescents (Hazell et al., 2002; Maneeton and Srisurapanont, 2000) and have unfavorable side effect profiles, including cardiotoxicity. However, the metaanalysis relies heavily on trials with small sample sizes. Overall, the recommended medications for stage 3 are generally well tolerated by youths. Venlafaxine is the only antidepressant for which a statistically significant increased risk of suicidal behavior over placebo has been shown, although this was attributable primarily to ideation, not to attempts. Although these medications are not recommended for initial treatment stages, it is possible that they may be considered for patients who have failed to respond with two previous trials of SSRIs based on adult studies (Trivedi et al., 2006). This is an area where additional data are critically needed. Stage 4: General Treatment Guidance. The panel deemed that inadequate data are available to make specific algorithm recommendations beyond the recommendations above. This does not negate the fact that the clinician must keep working to treat

J. AM. ACAD. CHILD ADOLESC. PSYCH IATRY, 46:6, JUNE 2007

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

CHILDHOOD MDD MEDICATION ALGORITHM

the child_s/adolescent_s depression. Thus, one should not infer that clinicians are expected to give up after multiple medication trials, but rather that inadequate research evidence exists to provide specific guideline recommendations. This being said, general recommendations are provided for options that clinicians may consider. First, assessment of medication adherence as well as the adequacy of the medication trials (dose and duration) should occur. A thorough diagnostic reassessment should occur with close re-examination of the primary diagnosis as well as co-occurring diagnoses (including substance abuse). Reassessment of the family situation, including the child_s/youth_s perspectives on the causes of his or her difficulties and attempts to address such should occur as well as the need for any family-based therapies.

Kutcher and Robertson, 1995; Scheekloth et al., 1993; Strober et al., 1988). ECT is recommended with the awareness that state and institutional guidelines must be followed, and, at least in Texas, state statutes do not permit ECT for patients under the age of 16 years. Recent ECT practice guidelines have been developed for adolescents (AACAP, 2004). It is anticipated that most children and adolescents will benefit from the early stages of the algorithm that are best supported by efficacy data; however, more treatment-resistant depressions may benefit from additional interventions. The actual response rates are not known because the algorithmic hierarchy has not been empirically tested, although a current study of treatment resistance in adolescents is ongoing (TORDIA, NIMH, David Brent, PI).

Considerations for treatment options include: • If the child has been in psychotherapy, then the adequacy of the psychotherapy should be assessed. If a psychotherapy technique other than CBT or interpersonal therapy has been used, then one of these is recommended. If psychotherapy has not been used, then it should be recommended to the child and family. • Although minimal research evidence is available to support antidepressant augmentation (Trivedi et al., 2006) or antidepressant combinations in youths, these strategies may be considered if the clinicians judge that the potential clinical benefits outweigh any potential adverse effects of the chosen regimen. • Consider other pharmacotherapy treatment options based on data from the adult literature (Rush et al., 2006). However, in applying this information to the treatment of children and adolescents, one should be diligent about the challenges in doing this. In particular, interventions showing efficacy in adults have not consistently done so in youths, particularly in preadolescents. Side effect risks may also be different in children and adolescents than in adults. • If the depression is severe and nonresponsive, then electroconvulsive therapy (ECT) should be considered at some point. Good efficacy data exist for adults (Abrams, 1992; Buchan et al., 1992; Fink and Foley, 1999; Prudic et al., 1996) [level A]) and level B and C literature exists for children and adolescents (AACAP, 2004; Bertagnolli and Borchart, 1990;

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:6, JUNE 2007

Approach for MDD With Psychotic Features

The treatment of MDD with psychotic features remains unchanged at this time and is approached the same as nonpsychotic MDD with the addition of the clinician_s choice of antipsychotic medication at each stage. The panel recognized the lack of data to support the choice of antipsychotic medication, but did prefer one of the newer atypicals because of the lower risk of neurological side effects. Clinicians are referred to recent reviews (Patel et al., 2005; Schur et al., 2003) for additional information on the use of antipsychotics in children and adolescents. Emphasis is placed on education to provide families and children with various nonpharmacological options (Pappadopulos et al., 2003), with the additional warning of the potential hazards of not actively treating the psychosis as well as the MDD. It is important in the assessment phase to evaluate the possible role of substance abuse as part of the diagnosis, and if there is no response by end of stage 3, further evaluation for possible bipolar disorder is needed. Again, Bwatchful waiting[ with ongoing assessment of clinical status for a few weeks is an option when this can be safely done, particularly for preadolescents. If an antidepressant is used in conjunction with antipsychotic medication, then the clinician should continue as previously recommended in the algorithm for MDD without psychotic features. Alternatively, the physician may choose to use an antipsychotic alone as initial treatment with the addition of an antidepressant if improvement is not seen. It is important to begin

675

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

HUGHES ET AL.

tapering the antipsychotic after remission of psychotic symptoms. One needs to be aware of weight gain and metabolic abnormalities associated with at least some atypical antipsychotics and of potential drug interactions between SSRIs and most antipsychotics (Crismon and Buckley, 2005). Prolonged use of an antipsychotic is associated with a risk of tardive dyskinesia, with atypical antipsychotics possessing a lower relative risk. MDD plus psychosis is an indicator of possible development of bipolar disorder (Akiskal et al., 1995; Strober and Carlson, 1982), and clinicians should be alert to this possibility, particularly if antidepressants are prescribed.

Treatment of MDD With Anxiety Disorders

A specific medication algorithm to address childhood anxiety disorders comorbid with MDD remains to be developed, although one is available for obsessivecompulsive disorder (March et al., 1998). Anxiety disorders or anxiety symptoms are frequently associated with depressive disorders, and treatment with SSRIs tends to address both sets of symptoms. A number of double-blind, placebo-controlled studies in youths have been completed in recent years demonstrating SSRIs to be superior to placebo in treating general anxiety disorder (Birmaher et al., 2003; Rynn et al., 2001; Walkup et al., 2002), social anxiety disorder (Wagner et al., 2004a), and obsessive-compulsive disorder (Riddle et al., 1992, 2001). In sum, the pharmacotherapy studies of childhood anxiety disorders with SSRIs have been encouraging without the prior concerns of abuse associated with the benzodiazepines (Hughes and Emslie, 1998). As with depression alone, consideration should be given to the use of nonpharmacological interventions (Compton et al., 2004; Manassis and Monga, 2001) because some of these disorders can be addressed with CBT (Cartwright-Hatton et al., 2004). Posttraumatic Stress Disorder (PTSD). The panel emphasized that the possibility of co-occurring PTSD needs to be addressed in the assessment of depressed youths (AACAP, 1998). It is critical that a thorough history is conducted for potential trauma, and that PTSD be recognized and treated if present. Traumafocused CBT has demonstrated efficacy (Cohen et al., 2005). Although only minimal research has been conducted addressing pharmacotherapy in youths with PTSD, adult studies support the use of SSRIs.

676

Algorithm for MDD With ADHD

In the original algorithm the panel recommended treating ADHD first unless the child had suicidal thoughts or behaviors. However, for the updated algorithm (Fig. 2, the panel recommended that clinicians assess the severity of the respective disorders and treat the most severe disorder first. Only children and adolescents with ADHD and unequivocal MDD that is clearly more severe than the ADHD should be treated first with an SSRI. The patient should have depressed mood nearly every day, lasting several hours, and it should be associated with significant neurovegetative signs. Patients with a comorbid diagnosis of ADHD and MDD whose ADHD has responded to a stimulant and are already stable can continue taking it and antidepressant therapy begun. The algorithm was revised in part because of the results from the feasibility study (Emslie et al., 2004b) in which a small number of patients presented with both MDD and ADHD and the depression was the more severe disorder. The treating clinicians chose to treat the depression first in those cases. As such, in the update of CMAP, the panel elected to allow the clinician to make the determination of which disorder to treat first, recognizing that the ADHD is known to respond more quickly and is a safer treatment. Once a treatment algorithm is initiated, symptoms should be reassessed after a few weeks of treatment. If the symptoms begin to remit (e.g., MDD) and if the other disorder_s symptoms remain (e.g., ADHD), then the clinician should begin the appropriate algorithm in conjunction with the first. Because treatment of one disorder often results in improvement of symptoms associated with the other disorder, it is recommended that pharmacotherapy be initiated for only one disorder at a timeVthe disorder judged to be more severe. Once pharmacotherapy is initiated and optimized for the first disorder treated, then symptomatology of the co-occurring disorder can be assessed for the possible need of additional pharmacotherapy (Fig. 2). If a child is started on a stimulant and then develops increased irritability, then the stimulant should be stopped. The clinician should switch to treating the depression and then, if necessary, retrying the stimulants. Treatment Recommendations for Stage 0. Recommendations for stage 0 are as above for MDD.

J. AM. ACAD. CHILD ADOLESC. PSYCH IATRY, 46:6, JUNE 2007

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

CHILDHOOD MDD MEDICATION ALGORITHM

Fig. 2 Medication algorithm for treating children and adolescents who meet DSM-IV criteria for major depressive disorder who also meet criteria for attentiondeficit/hyperactivity disorder (ADHD). Revised based on Hughes et al. (1999). (This algorithm is in the public domain and may be used and reprinted without special permission. The authors appreciate proper citation. The authors bear no responsibility for the use of these guidelines by third parties).

Treatment Recommendations If ADHD Is Treated First. Stage 1. Begin treatment with stage 1 of the ADHD algorithm, either methylphenidate or amphetamine (Pliszka et al., 2006). If both depressive and ADHD symptoms adequately improve, then continuation treatment with the agent used in stage 1 is maintained (Fig. 2). Stage 2. Decision possibility 1: If the patient_s ADHD symptoms respond but the patient_s depressive symptoms remain, then the clinician should begin the MDD algorithm and add a stage 1 SSRI to the stimulant. Decision possibility 2: If neither the patient_s ADHD symptoms nor depressive symptoms respond to a stimulant, then the clinician should discontinue the stimulant and treat according to the MDD algorithm. Once the depressive symptoms respond, the physician should re-evaluate for presence of ADHD symptoms. If they are still present, then the patient should be treated using the ADHD algorithm (Pliszka et al., 2006). Of note is that although noradrenergic antidepressants (e.g., desipramine, bupropion) have been shown to be effective for ADHD (Biederman et al., 1989; Conners et al., 1996), they have not been shown to be effective

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:6, JUNE 2007

for MDD; therefore, initial treatment with an SSRI for the depression, with the possibility of the addition of a stimulant for the ADHD, is preferred over initiating a noradrenergic antidepressant. Treatment Recommendations If Depression Is Treated First. Stage 1. Begin treatment with an SSRI from stage 1 of the depression algorithm. If both depressive and ADHD symptoms adequately improve, then continuation treatment with the agent used in stage 1 is maintained. Stage 2. Decision possibility 1: If the patient_s depressive symptoms adequately improve but the ADHD symptoms remain, then the clinician should begin the ADHD algorithm and add a stage 1 stimulant to the regimen. Decision possibility 2: If the patient_s depressive symptoms do not improve, the clinician should continue with stage 2 of the depression algorithm. If partial improvement has occurred and the clinician deems that inadequate response may be due to the ADHD symptoms, then the clinician may consider adding an

677

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

HUGHES ET AL.

ADHD algorithm stage 1 stimulant. However, it is important that only one medication change (e.g., dose change, switch to another medication) be made at a time so that the effectiveness and tolerance of the change can be evaluated. Comorbid Mood Disorder With Aggression and Other Disruptive Behavior Disorders

A need exists to address individuals who present with mood and disruptive behavior disorders and when SSRI monotherapy is insufficient to address these disruptive behaviors (Jensen et al., 2007). The recommended first step in these cases is careful assessment to rule out bipolar disorder, and if bipolar disorder is diagnosed, then the clinician should consider implementing a recent algorithm developed specifically for childhood bipolar disorder (Kowatch et al., 2005). A mood disorder with irritable and aggressive mood often necessitates additional measures to quickly ensure safety and efficacy (Pliszka et al., 2006). Guidelines have been published to address this important need, with 14 specific treatment recommendations for the use of atypical antipsychotics for youths with aggressive behavior that is comorbid with other psychiatric disorders (Pappadopulos et al., 2003; Schur et al., 2003). For situations in which mood disorders co-occur with aggression, the application of these guidelines is recommended. Treatment Tactics

Although strategies define which treatment to provide and the recommended order (stage) to implement, tactics address how to optimally implement a chosen treatment regimen (i.e., a strategy; Crismon et al., 1999; Rush et al., 1998, 1999). Table 1 outlines proposed tactics with the emphasis that the goal of treatment is to achieve remission of symptoms. For patients showing minimal or no improvement, a change in treatment stage may be considered at 4 to 8 weeks. However, recent studies in adults have shown that many nonresponders at 4 to 6 weeks will actually have symptom remission by 12 weeks of continued treatment (Quitkin et al., 2005). For patients with a partial response, the pharmacotherapy trial may last up to 12 weeks because decisions to increase the dose or augment with lithium may reasonably postpone the

678

critical point at which it would be decided to change stages. Serious consideration should be given to changing treatment stages in patients who are only partial responders at 12 weeks because patients with residual depressive symptoms are at risk of relapse. It should be noted that this includes individuals who demonstrate robust improvement, but who still have not achieved remission. Visit Frequency. The consensus group agreed that patients should be seen more often early in treatment (e.g., every 1Y2 weeks). Seeing patients often allows the clinician to detect potential worsening of depression, emergence of suicidality, and onset of bothersome adverse effects, and to adjust dose; this likely increases patient adherence to treatment. In addition, more frequent contact enhances the engagement of the child and family in treatment and provides additional opportunities for education (Crismon et al., 1999; Depression Guideline Panel, 1993; Lopez et al., 2005; Rush and Kupfer, 1995). The 2004 FDA box warning recommends that patients treated with antidepressants be observed closely, and that Bsuch observation would generally include at least weekly face-to-face contact with patients or their family members or caregivers during the first 4 weeks of treatment, then visits every other week for the next 4 weeks, then at 12 weeks, and as clinically indicated beyond 12 weeks.[ However, visit frequency for specific patients may vary based on multiple factors (e.g., presence of suicidal behaviors, severity of illness, severity of psychosocial stressors, development of therapeutic alliance, level of family support, progress in treatment). Both patients and families should be encouraged to contact the clinician if depression worsens, the patient demonstrates suicidal behavior or verbalizations, or medication side effects occur. The appropriate use of nonphysician personnel who are knowledgeable about the patient population can aid in increasing the frequency of contact between the clinic and the patient/parent. Measurement of Outcome. It is recommended that improvement in depressive symptoms be systematically and objectively monitored (e.g., when treating someone for high blood pressure, physicians measure ongoing changes in blood pressure to determine whether the treatment is effective). Similarly, severity of depression, rather than simply overall functioning, needs to be monitored. Various depressive rating scales are available in both clinician-rated and self-report

J. AM. ACAD. CHILD ADOLESC. PSYCH IATRY, 46:6, JUNE 2007

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

CHILDHOOD MDD MEDICATION ALGORITHM

Assessment Point

TABLE 1 Tactics for Acute Phase Treatment of Major Depressive Disorder Clinical Status

Before pharmacotherapy

Symptomatic

Weeks 1Y3 (critical point 1)

Symptomatic

Week 4 (critical point 2)

Response or remission Partial responsea

Week 6 (critical point 3)

Week 8 (critical point 4)

Week 10 (critical point 5)

Week 12 (critical point 6)

Watchful waiting, particularly for preadolescents Consider psychoeducation and lifestyle management Initiate medicine: adjust dose to lower end of therapeutic dose range or serum level if useful If patient shows rapid remission in first 2Y3 wk, then this may be a placebo response; continue to offer treatment and encouragement Go to continuation phase Satisfactory rate of improvement: observe Rate too slow, tolerating well: increase dose Discontinue, proceed to next stage

Minimal or no response; patient intolerant of lowest therapeutic dose Minimal or no response; patient tolerating medicine Response or remission Partial response

Increase doseb

Minimal response; patient intolerant of higher dose Minimal response; patient tolerant Response or remission Partial response

Minimal response to lithium augmentation for 2Y3 wk Response or remission Partial response

No or minimal response Response or remission Remains partial responder

Plan

Go to continuation phase Satisfactory rate of improvement if previously increased dose: observe Rate too slow, tolerating well: increase dose; if dose already increased to maximum, then consider augmentation (stage 2 or later) Discontinue, proceed to next stage Consider augmentation (stage 2 or later) Go to continuation phase If tolerating regimen, then consider augmentation with lithium (or alternative as above) if not previously done (stage 2 and later) If not tolerating regimen, then go to next stage Discontinue, switch to next level in plan Go to continuation phase Increase lithium dose if not previously done (stage 2 and later) If on higher lithium dose, then go to next stage Go to next stage Go to continuation phase Go to next stage

a

Partial response: Patient is showing some evidence of improvement based on rating scales to the extent that it is unlikely that the clinician would choose to switch to a different medication. However, some change in treatment may be warranted to enhance response. b Minimal response: Patient is showing virtually no change, although 1Y2 symptoms may be showing slight improvement. Some change in treatment, whether dose change, switching, or augmenting, are definitely warranted.

form (Zuckerbrot and Jensen, 2006). Further refinement of measures that would be easier for clinicians to use is needed. Dosing and Side Effects. Although dosing should usually start low to minimize side effects, dose titration should occur in children and adolescents who do not achieve a robust improvement in symptoms (e.g., at 4Y6 weeks of treatment). In patients achieving less than

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:6, JUNE 2007

a robust response and with minimal or no side effects, further dose adjustments should be considered (e.g., at 8Y10 weeks of treatment) before declaring a therapeutic or stage failure. Children should be monitored carefully for the emergence of side effects, including overactivation, during dose titration. If such side effects occur, then the dose should be lowered or an alternate treatment stage should be considered.

679

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

HUGHES ET AL.

The half-lives of the SSRIs and other newer antidepressants are shorter in children (Axelson et al., 2002; Daviss et al., 2005). At least theoretically, based on these data, children and adolescents could experience withdrawal side effects 8 to 12 hours after the last medication dose (except with fluoxetine), and these symptoms could be confused with lack of response or medication side effects. SSRIs are associated with several side effects such as headache, gastrointestinal distress, sexual dysfunction, increased sweating, disruptive sleep, and, rarely, bleeding and serotonergic syndrome (Cassano and Fava, 2004; Masand and Gupta, 2002). Throughout each stage of the algorithm, the treatment of medication-related side effects should be addressed through the lowering of doses where possible, and if no improvement results, changing medications. In particular, although not common, patients should be monitored for the emergence of behavioral symptoms such as aggression, hostility, mania, behavioral activation, or motor restlessness, as well as for suicidal thoughts and behaviors (Cheung et al., 2005). No level A or B data exist to support the addition of another medication as a means of treating or reducing the severity of a side effect in children and adolescents. Drug interactions can result in either increased side effects or decreased therapeutic effectiveness (DeVane and Nemeroff, 2002), and clinicians should be knowledgeable about and vigilant regarding potential interactions when patients are receiving multiple medications. Assessment of Suicidality. The panel also recommends that suicidality be assessed at every visit to include behavior and ideation, including frequency, plans, intention, and potential dangerousness (Posner et al., 2007a). More frequent visits, combined with follow-up calls as necessary, should be considered along with appropriate review of safety plans and contracts with patient and family. A challenge to adequate assessment has been the clinical myth that asking about suicidality will cause increases in suicidal feelings when in fact recent research has shown this not to be the case (Gould et al., 2005). Other major suicide risk factors that should be assessed include substance abuse and conduct disorder (because there are even higher rates of substance abuse in male victims [Shaffer et al., 1996]), life stressors such as legal or disciplinary/school problems and interpersonal losses (Brent, 1993; Gould

680

et al., 1996), family and peer discord, abuse, lack of support, poor interpersonal problem-solving ability, the tendency to respond with hostility or overt aggression to frustration or stress, and hopelessness and cognitive distortions (Beautrais, 2000; Brent et al., 2002; Fergusson et al., 2000; Gould et al., 2003; Molnar et al., 2001;Shaffer et al., 1996). Finally, assessment should include parental depression and substance abuse (Gould et al., 1996). Reassessment of Treatment Nonresponders. See stages 0 and 4, Binadequate response,[ above. Continuation Phase Treatment for MDD

Recommendations for continuation phase treatment for children and adolescents are based on available studies of childhood and adolescent depression (Emslie et al., 2004a; Keller et al., 1998; Pine, 2002) as well as adult guidelines. At baseline and throughout treatment, patients should receive an assessment of their psychosocial needs, including the utility of psychotherapy or other rehabilitative interventions. Continuation of medication(s) for 6 to 12 months is recommended after symptom remission (i.e., after end of successful acute phase treatment) at the full therapeutic dose used in the acute phase (Crismon et al., 1999; Emslie et al., 2004a; Depression Guideline Panel, 1993; Fava and Rosenbaum, 1995; Pine, 2002; Rush and Kupfer, 1995). After reaching remission of depression, patients should be seen by clinicians at least once every 3 months (preferably every 1Y2 months) during continuation treatment to monitor for relapse of symptoms or suicidal ideation. Because bipolar disorder may present with depression as the initial episode, monitoring for the potential occurrence of a manic episode should occur. If this was the first major depressive episode, then the patient should be evaluated for tapering and medication discontinuation at the end of continuation phase treatment. If previous depressive episodes have occurred, then the patient should be evaluated for maintenance treatment. Determining the duration of continuation treatment, frequency of visits, and need for maintenance treatment is based on several factors, including severity of the index episode, ongoing residual symptoms, history of multiple episodes, patient and family preference, and side effects of continued treatment. Clinicians, patients, and families

J. AM. ACAD. CHILD ADOLESC. PSYCH IATRY, 46:6, JUNE 2007

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

CHILDHOOD MDD MEDICATION ALGORITHM

collaboratively evaluate the patient_s need for and intensity of ongoing treatment. The multisite fluoxetine trial (Emslie et al., 2002, 2004a) reports on the pharmacological maintenance treatment of 40 children and adolescents who had responded to 20 to 60 mg fluoxetine during a 19-week, double-blind, placebo-controlled study of fluoxetine. Patients were rerandomized to either continue fluoxetine or switch to placebo for an additional 36 weeks (Emslie et al., 2004a). Relapse was defined as a CDRS-R score 940 with a history of 2 weeks of clinical deterioration or patients who had, in the opinion or the physician, experienced a relapse. Fewer fluoxetine subjects (34%) than placebo subjects (60%) met the criteria for relapse. Furthermore, time to relapse was shorter for placebo subjects than fluoxetine subjects (71.2 T 9.5 days versus 180.7 T 17.0 days, respectively; p = .046). Using the strict CDRS score 940 criterion (excluding physician decision to terminate), 21% fluoxetine versus 47% placebo relapsed, and time to relapse was 37.2 T 2.1 days versus 203.0 T 13.0 days (p = .032). When discontinuing the antidepressant, the dose should be tapered no more rapidly than 25%/week (or as slow as practical with available dose forms). Tapering and discontinuation usually occur over a 2- to 3-month period (Crismon et al., 1999; Depression Guideline Panel, 1993; Pine, 2002; Rush and Kupfer, 1995). Although fluoxetine has a long elimination half-life, it is likely best to taper it as well, if for no other reason than it reinforces the need for the family to observe the patient for possible reemergence of depression with tapering and discontinuation. The time to begin a taper and the tapering schedule should be discussed with and agreed to by the patient and family. Patients and their primary caretaker should be taught to monitor for recurrence of depressive symptoms. Because a new depressive episode is most likely to occur within the first 8 months after medication discontinuation, patients should be seen every 2 to 4 months during this period (Emslie et al., 1997b). If depression recurs, then prompt treatment with the previously effective medication should be reinitiated (i.e., initiate algorithm stage and tactic that previously resulted in remission of depressive symptoms). For patients with MDD with psychotic features treated with an antipsychotic medication, the acute phase antipsychotic dose should be continued for 2 to

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:6, JUNE 2007

3 months following remission of psychotic symptoms and then slowly tapered (Schur et al., 2003). The duration of antipsychotic treatment should be limited as the patient_s clinical situation permits to reduce the risk of neurological side effects. For patients with comorbid ADHD who are stabilized on only a stimulant medication, recommendations for continuation phase treatment follows those suggested by the CMAP Consensus Conference Panel on ADHD (Pliszka et al., 2006). Maintenance Phase Treatment

In adults with an initial episode of MDD, at least 50% are at risk of a second episode, and by the third episode, the recurrence risk is 90% (Depression Guideline Panel, 1993). Similarly, up to 70% of youths experiencing MDD will experience a recurrence (Birmaher et al., 2002). Therefore, all patients, including children and adolescents, experiencing their third depressive episode, and depending on risk factors (e.g., family history, severity of episode, suicidality during the episode) some patients with a second episode, should be considered for maintenance treatment (Crismon et al., 1999; Depression Guideline Panel, 1993; Fava and Rosenbaum, 1995; Rush and Kupfer, 1995). Data suggest that continuity of major depression exists between childhood and adulthood, with recurrence rates of another depressive episode estimated to be approximately 60% to 70% (Harrington et al., 1990; Rao et al., 1995). Maintenance medication should continue at full therapeutic doses, and, as in the continuation phase, the regimen associated with symptom remission is recommended. The optimal duration of maintenance medication has not been established in adults or children and adolescents, but depending on risk factors, is generally believed to be between 3 years and lifetime. Active discussions regarding the initiation and duration of maintenance treatment are an important element in the clinicianYpatientYfamily collaboration. The patient and family_s personal preference as well as the risk factors for recurrence must be considered in the decision process. No published maintenance studies of antidepressant use in children are available, and these recommendations are borrowed from the adult MDD literature, which may or may not be applicable to children and adolescents.

681

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

HUGHES ET AL.

CONCLUSIONS

A consensus panel of experts, clinicians, administrators, and consumer family members revised and updated the Texas medication treatment algorithm for MDD in children and adolescents. The final product resulted in revised recommendations for antidepressant treatment, with an emphasis on adverse event monitoring and specialized assessment for suicidality and comorbid disorders. The implementation of a treatment algorithm as a component of disease management for children with depression was superior to a historical treatment as usual cohort of children and adolescents (Emslie et al., 2004a) as well as in a large-scale comparative trial in adults with MDD in the public sector (Trivedi et al., 2004, 2006). Medication algorithms for childhood MDD have been incorporated as pharmacotherapy guidelines for a number of ongoing NIMH-sponsored trials in both children and adolescents (e.g., NIMH Treatment of Adolescent Suicide Attempters; Vitiello, 2006) and adults (Fava et al., 2003; Rush et al., 2006; Trivedi et al., 2004; Yates et al., 2004). Clinicians, administrators, patients, and families have recently endorsed the use of these treatment algorithms. Initial concerns regarding the applicability of the algorithms, concern that treatment algorithms would limit clinical judgment and professional identity, and concerns about liability have not been borne out. Medication algorithms do not limit clinicians_ ability to use their clinical judgment; rather they provide a framework for decision making. The medication algorithms include simplified initial treatment interventions for straightforward cases with an emphasis placed on initial monotherapy trials. Algorithms do not reduce the need for experienced clinicians to use their clinical acumen in diagnosis, assessment of severity, ongoing monitoring, decision making, and ability to establish a therapeutic relationship. In fact, a consistent clinicianYpatient relationship is mandatory in the treatment of a disorder frequently requiring long-term treatment and adherence to medication regimens and close monitoring of initial and emerging side effects. Both parents and adolescents continue to be supportive of the use of the algorithms. They want accurate and understandable information about mental disorders and their appropriate treatment. Family advocates participated in the development of a psychoedu-

682

cational program as a critical component of the algorithm disease management program during the CMAP feasibility trial. They strongly endorsed the need for adequate parent/child education and the need to be involved in treatment decisions (Lopez et al., 2005). The major limitation to developing MDD algorithms for children and adolescents continues to be the lack of empirical double-blind, placebo-controlled studies of various medications that demonstrate efficacy and safety (only five published in the past 6 years). Similarly, almost no research regarding antidepressant combinations or augmentation is available in children or for treating individuals who are treatment resistant. A strength of the Texas algorithmic approach is that it synthesizes available empirical data with expert opinion, clinical experience, and consumer perspectives. It is designed to adjust and incorporate new information on an ongoing basis. A CMAP Web site link (Texas Department of State Health Services, 2005) provides a source for all of the CMAP- and Texas Medication Algorithm ProjectY related documents including downloadable psychoeducational material that has been developed to support the algorithm approach. The members of the Texas Consensus Conference Panel on Medication Treatment of Childhood Major Depressive Disorder are as follows: Graham J. Emslie, M.D. (chair), Department of Psychiatry, University of Texas Southwestern Medical Center, Dallas; M. Lynn Crismon, Pharm.D. (co-chair), College of Pharmacy, University of Texas at Austin; Joan Barcelona, Ft. Worth; Boris Birmaher, M.D., Western Psychiatric Institute and Clinic, Pittsburgh; John Curry, Ph.D., Duke University, Durham, NC; Cindy Hopkins, family member, Texas Department of State Health Services, Austin; Carroll W. Hughes, Ph.D., Department of Psychiatry, University of Texas Southwestern Medical Center, Dallas; Peter S. Jensen, M.D., Center for the Advancement of Children_s Mental Health, Department of Psychiatry, Columbia University and Office of Mental Health, New York; Molly Lopez, Ph.D., Texas Department of State Health Services, Austin; Robin Mallett, M.D., Gulf Coast MHMR Center, League City, TX; Sylvia Musquiz, M.D., Mental Health and Mental Retardation Authority of Harris County, Houston; Steven R. Pliszka, M.D., Department of Psychiatry, University of Texas Health Science Center at San Antonio; Kelly Posner, Ph.D., Columbia University, New York; Valerie Robinson, M.D., Department of Psychiatry, Texas Tech University Health Sciences Center, Lubbock; Susan Rogers, family member, Dallas; Neal Ryan, M.D., Western Psychiatric Institute and Clinic, Pittsburgh; Steve Shon, M.D., Texas Department of State Health Services, Austin; Madhukar Trivedi, M.D., Department of Psychiatry, University of Texas Southwestern Medical Center, Dallas; Sylvia Turner, M.D., Child and Adolescent Unit,

J. AM. ACAD. CHILD ADOLESC. PSYCH IATRY, 46:6, JUNE 2007

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

CHILDHOOD MDD MEDICATION ALGORITHM

Terrell State Hospital, Terrell, TX; Benedetto Vitiello, M.D., National Institute of Mental Health, Bethesda, MD.

Disclosure: Dr. Hughes has received recent research support from GlaxoSmithKline (GSK) and is a consultant to Biobehavioral Diagnostics, Inc. Dr. Emslie receives research support from Eli Lilly, Organon, Inc., and Forest Laboratories; is a consultant to Biobehavioral Diagnostics Inc., Eli Lilly, GSK, Forest, Wyeth-Ayerst, and Pfizer; and is on the speakers’ bureau of McNeil Consumer and Specialty Pharmaceuticals. During the past 3 years Dr. Crismon has received research support from, served as a consultant to, or has been on the speakers’ bureaus of the following: Bristol-Myers Squibb (BMS), Corcept Therapeutics, Cyberonics, Inc., Eli Lilly, Jackson Evidence Based Partnership, Janssen Pharmaceutica, McNeil Specialty and Consumer Products, Pfizer, Seton Health Network, and Shire Pharmaceuticals. As part of an effort to help execute the FDA suicidality classification mandates, Dr. Posner has received research support from GSK, Forest, Eisai, AZ Pharmaceuticals, Johnson & Johnson, Abbott Laboratories, Wyeth-Ayerst Research, Organon USA, BMS, Sanofi-Aventis, Cephalon, Novartis, Shire Pharmaceuticals, and UCB Pharma. Dr. Ryan has served as a consultant to Pfizer, GSK, BMS, Lilly, Wyeth-Ayerst, Abbott, and Janssen and has received grant funding from GSK and WyethAyerst. Dr. Jensen receives investigator-initiated grants from McNeil and unrestricted educational grants from Pfizer, Eli Lilly, and McNeil; participates in the speakers’ bureaus of CMED, UCB, PsychCME, CME Outfitters, and the Neuroscience Education Institute; and consults with Best Practice, Inc., Janssen, Novartis, and UCB. Dr. Pliszka serves as a consultant to and a member of the speakers’ bureaus of Shire and McNeil Specialty and Consumer Products; he has received research support from Eli Lilly, Cephalon, and AZ. Dr. Trivedi has received research support from, served as a consultant to, or been on the speakers’ bureaus of Abdi Brahim, Abbott, Akzo (Organon Pharmaceuticals), AstraZeneca, Bayer, BMS, Cephalon, Corcept Therapeutics, Cyberonics, Eli Lilly, Forest, GSK, Janssen, Johnson & Johnson, Meade Johnson, Merck, Neuronetics, Novartis, Parke-Davis Pharmaceuticals, Pfizer, Pharmacia & Upjohn, Predix Pharmaceuticals, Sepracor, Solvay Pharmaceuticals, and Wyeth-Ayerst. The other authors have no financial relationships to disclose.

REFERENCES AACAP (1998), Summary of the practice parameters for the assessment and treatment of children and adolescents with posttraumatic stress disorder. J Am Acad Child Adolesc Psychiatry 37:997Y1001 AACAP (2004), Practice parameter for use of electroconvulsive therapy with adolescents. J Am Acad Child Adolesc Psychiatry 43:1521Y1539 Abrams R (1992), Electroconvulsive Therapy, 2nd ed. New York: Oxford University Press Akiskal HS, Maser JD, Zeller PJ et al. (1995), Switching from Bunipolar[ to bipolar II: an 11-year prospective study of clinical and temperamental predictors in 559 patients. Arch Gen Psychiatry 52:114Y123 Axelson DA, Perel JM, Birmaher B et al. (2002), Sertraline pharmacokinetics and dynamics in adolescents. J Am Acad Child Adolesc Psychiatry 41:1037Y1044 Beautrais AL (2000), Risk factors for suicide and attempted suicide among young people. Aust N Z J Psychiatry 34:420Y436 Berard R, Fong R, Carpenter DJ, Thomason C, Wilkinson C (2006), An international, multicenter, placebo-controlled trial of paroxetine in adolescents with major depressive disorder. J Child Adolesc Psychopharmacol 16:59Y75 Bertagnolli MW, Borchart CM (1990), A Review of ECT for Children and Adolescents. Cambridge, UK: Cambridge University Press Biederman J, Baldessarini RJ, Wright V, Knee D, Harmatz JS (1989), A

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:6, JUNE 2007

double-blind placebo controlled study of desipramine in the treatment of ADD: I. Efficacy. J Am Acad Child Adolesc Psychiatry 28:777Y784 Birmaher B, Arbelaez C, Brent D (2002), Course and outcome of child and adolescent major depressive disorder. Child Adolesc Psychiatry Clin N Am 11:619Y637 Birmaher B, Axelson DA, Monk K et al. (2003), Fluoxetine for the treatment of childhood anxiety disorder. J Am Acad Child Adolesc Psychiatry 42:415Y423 Birmaher B, Brent DA, Kolko DJ, Baugher M, Bridge J, Iyengar S, Ulloa RE (2000), Clinical outcome after short-term psychotherapy for adolescents with major depressive disorder. Arch Gen Psychiatry 57:29Y36 Brent DA (1993), Depression and suicide in children and adolescents. Pediatr Rev 14:380Y388 Brent DA, Holder D, Kolko DJ et al. (1997), A clinical psychotherapy trial for adolescent depression comparing cognitive, family, and supportive therapy. Arch Gen Psychiatry 54:877Y885 Brent DA, Kolko D, Birmaher B et al. (1998), Predictors of treatment efficacy in a clinical trial of three psychosocial treatments for adolescent depression. J Am Acad Child Adolesc Psychiatry 37:906Y914 Brent DA, Oquendo M, Birmaher B et al. (2002), Familial pathways to early-onset suicide attempt: risk for suicidal behavior in offspring of mood-disordered suicide attempters. Arch Gen Psychiatry 59:801Y807 Buchan H, Johnstone E, McPherson K, Palmer RL, Crow TJ, Grandon S (1992), Who benefits from electroconvulsive therapy? Combined results from the Leicester and Northwick Park trials. BMJ 160:355Y359 Carpenter LL, Yasmin S, Price LH (2002), A double-blind, placebocontrolled study of antidepressant augmentation with mirtazapine. Biol Psychiatry 51:183Y188 Cartwright-Hatton S, Roberts C, Chitsabesan P, Fothergill C, Harrington R (2004), Systematic review of the efficacy of cognitive behavioral therapies for childhood and adolescent anxiety disorders. Br J Clin Psychol 43:421Y436 Cassano P, Fava M (2004), Tolerability issues during long-term treatment with antidepressants. Ann Clin Psychiatry 16:15Y25 Centers for Disease Control and Prevention (2006). Web-based Injury Statistics Query and Reporting System. Available at: http://www.cdc.gov/ ncipc/wisqars. Accessed January 30, 2007 Cheung A, Zuckerbrot R, Jensen PS, the GLAD-PC Cooperative Group (in press), Guidelines for adolescent depression in primary care. Pediatrics Cheung AH, Emslie GJ, Mayes TL (2005), Review of the efficacy and safety of antidepressants in youth depression. J Child Psychol Psychiatry 46:735Y754 Clarke GN, Rohde P, Lewinsohn PM, Hops H, Seeley JR (1999), Cognitive-behavioral treatment of adolescent depression: efficacy of acute group treatment and booster sessions. J Am Acad Child Adolesc Psychiatry 38:272Y279 Cohen JA, Mannarino AP, Knudsen K (2005), Treating sexually abused children: 1 year follow-up of a randomized controlled trial. Child Abuse Negl 29:135Y145 Compton SN, March JS, Brent DA, Albano AM, Weersing VR, Curry JF (2004), Cognitive-behavioral psychotherapy for anxiety and depressive disorders in children and adolescents: an evidence-based medicine review. J Am Acad Child Adolesc Psychiatry 43:930Y959 Conners CK, Casat C, Gualtieri CT et al. (1996), Bupropion hydrochloride in attention deficit disorder with hyperactivity. J Am Acad Child Adolesc Psychiatry 35:1314Y1321 Crismon ML, Buckley PF (2005), Schizophrenia. In: Pharmacotherapy: A Pathophysiologic Approach, 6th ed, DiPirto JT, Talbert RL, Yee GC, Matzke GR, Wells BG, Plsey LM, eds. New York: McGraw-Hill, pp 1209Y1233 Crismon ML, Trivedi MH, Pigott TA et al. (1999), The Texas Medication Algorithm Project: report of the Texas Consensus Conference Panel on Medication of Major Depressive Disorder. J Clin Psychiatry 60: 142Y156 Curry JF (2001), Specific psychotherapies for childhood and adolescent depression. Biol Psychiatry 49:1091Y1100 Curry JF, Rohde P, Simons A et al. (2006), Predictors and moderators of acute outcome in the Treatment for Adolescents With Depression Study (TADS). J Am Acad Child Adolesc Psychiatry 45:1427Y1439

683

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

HUGHES ET AL. Daviss WB, Perel JM, Rudolph G et al. (2005), Steady-state pharmacokinetics of bupropion SR in juvenile patients. J Am Acad Child Adolesc Psychiatry 44:349Y357 Daviss WB, Bentivoglio P, Racusin R, Brown KM, Bostic JQ, Wiley L (2001), Bupropion sustained release in adolescents with comorbid attention-deficit/hyperactivity disorder and depression. J Am Acad Child Adolesc Psychiatry 40:307Y314 Depression Guideline Panel (1993), Clinical Practice Guideline. Depression in Primary Care: Volume 1 and 2. Detection and Diagnosis. Publication No. 93-0550. Washington, DC: U.S. Department of Health and Human Services, Agency for Health Care Policy Research DeVane CL, Nemeroff CB (2002), 2002 guide to psychotropic drug interactions. Prim Psychiatry 9:28Y57 Emslie GJ, Findling RL, Yeung PP, Kunz NR, Li Y (2007), Venlafaxine ER for the treatment of pediatric subjects with depression: results of two placebo-controlled trials. J Am Acad Child Adolesc Psychiatry 46: 479Y488 Emslie GJ, Heiligenstein JH, Hoog SL et al. (2004a), Fluoxetine treatment for prevention of relapse of depression in children and adolescents: a double-blind, placebo-controlled study. J Am Acad Child Adolesc Psychiatry 43:1397Y1405 Emslie GJ, Heiligenstein JH, Wagner KD et al. (2002), Fluoxetine for acute treatment of depression in children and adolescents: a placebocontrolled, randomized clinical trial. J Am Acad Child Adolesc Psychiatry 41:1205Y1215 Emslie GJ, Hughes CW, Crismon ML et al. (2004b), A feasibility study of the childhood depression medication algorithm: the Texas Children_s Medication Algorithm Project (CMAP). J Am Acad Child Adolesc Psychiatry 43:1Y9 Emslie GJ, Kratochvil CJ, Vitiello B et al. (2006a), Treatment for Adolescents With Depression Study (TADS): safety results. J Am Acad Child Adolesc Psychiatry 45:1440Y1455 Emslie GJ, Rush AJ, Weinberg WA (1997a), A double-blind, randomized, placebo-controlled trial of fluoxetine in children and adolescents with depression. Arch Gen Psychiatry 54:1031Y1037 Emslie GJ, Rush AJ, Weinberg WA, Gullion CM, Rintelmann J, Hughes CW (1997b), Recurrence of major depressive disorder in hospitalized children and adolescents. J Am Acad Child Adolesc Psychiatry 36: 785Y792 Emslie GJ, Wagner KD, Kutcher S et al. (2006b), Paroxetine treatment in children and adolescents with major depressive disorder: a randomized, multicenter, double-blind, placebo-controlled trial. J Am Acad Child Adolesc Psychiatry 45:709Y719 Fava M, Rosenbaum JF (1995), Pharmacotherapy and Somatic Therapies. New York: Guilford Fava M, Rush AJ, Trivedi MH et al. (2003), Background and rationale for the sequenced treatment alternatives to relieve depression (STAR*D) study. Psychiatr Clin North Am 26:457Y494 Fergusson DM, Woodward LJ, Horwood LJ (2000), Risk factors and life processes associated with the onset of suicidal behavior during adolescence and early adulthood. Psychol Med 30:23Y39 Fink M, Foley CA (1999), Pediatric ECT: an update. Psychiatric Times 16:63Y65 Food and Drug Administration (2006), Clinical review: relationship between antidepressant drugs and suicidality in adults. Available at: http://www.fda.gov/ohrms/dockets/ac/06/briefing/2006-4272b1-01-FDA. pdf. Accessed March 7, 2007 Gibbons RD, Hur K, Bhaumik DK, Mann JJ (2005), The relationship between antidepressant medication use and rate of suicide. Arch Gen Psychiatry 62:165Y172 Gibbons RD, Hur K, Bhaumik DK, Mann JJ (2006), The relationship between antidepressant prescription rates and rate of early adolescent suicide. Am J Psychiatry 163:1898Y1904 Gilbert DA, Altshuler KZ, Rago WV (1997), Texas Medication Algorithm Project: definitions, rationale and methods to develop medication algorithms. J Clin Psychiatry 59:345Y351 Gould MS, Fisher P, Parides M, Flory M, Shaffer D (1996), Psychosocial risk factors of child and adolescent completed suicide. Arch Gen Psychiatry 53:115Y1162

684

Gould MS, Greenberg T, Velting DM, Shaffer D (2003), Youth suicide risk and preventive interventions: a review of the past 10 years. J Am Acad Child Adolesc Psychiatry 42:386Y405 Gould MS, Marrocco FA, Kleinman M et al. (2005), Evaluating iatrogenic risk of youth suicide screening programs: a randomized controlled trial. JAMA 293:1635Y1643 Gray D, Moskos M, Keller T (2003), Utah Youth Suicide Study: new findings. Paper presented at the Annual Meeting of the American Association of Suicidology, April 22Y26, Santa Fe, NM Hammad TA, Laughren T, Racoosin J (2006), Suicidality in pediatric patients treated with antidepressant drugs. Arch Gen Psychiatry 63:332Y339 Hammen C (1999), The context of depression in clinic-referred youth: neglected areas in treatment. J Am Acad Child Adolesc Psychiatry 38: 64Y71 Harrington R, Fudge H, Rutter M, Pickles A, Hill J (1990), Adult outcomes of childhood and adolescent depression: I. Psychiatric status. Arch Gen Psychiatry 47:465Y473 Hazell P, O_Connell D, Heathcote D, Henry D (2002), Tricyclic drugs for depression in children and adolescents. Cochrane Database Syst Rev 2:CD002317 Hughes CW, Emslie GJ, Crismon ML et al. (1999), The Texas Children_s Medication Algorithm Project: report of the Texas Consensus Conference Panel on Medication Treatment of Childhood Major Depressive Disorder. J Am Acad Child Adolesc Psychiatry 38:1442Y1454 Hughes CW, Emslie GJ (1998), Treatment of Anxiety Disorders in Children and Adolescents. Philadelphia: Williams & Wilkins Isacsson G, Holmgren P, Ahlner J (2005), Selective serotonin reuptake inhibitor antidepressants and the risk of suicide: a controlled forensic database study of 14,857 suicides. Acta Psychiatr Scand 111:286Y290 Isacsson G, Rich CL (2005), Antidepressant drug use and suicide prevention. Int Rev Psychiatry 17:153Y162 Jensen PS, Youngstrom Y Steiner H et al. (2007), Consensus report on impulsive aggression as a symptom across diagnostic categories in child psychiatry: implications for medication studies. J Am Acad Child Adolesc Psychiatry 46:309Y322 Keller MB, Kocsis JH, Thase ME et al. (1998), Maintenance phase efficacy of sertraline for chronic depression: a randomized controlled trial. JAMA 280:1665Y1672 Keller MB, Ryan ND, Strober M et al. (2001), Efficacy of paroxetine in the treatment of adolescent major depression: a randomized, controlled trial. J Am Acad Child Adolesc Psychiatry 40:762Y772 Kowatch RA, Fristad M, Birmaher B, Wagner KD, Findling RL, Hellander JD (2005), Treatment guidelines for children and adolescent with bipolar disorder: Child Psychiatric Workgroup on Bipolar Disorder. J Am Acad Child Adolesc Psychiatry 44:213Y235 Kratochvil CJ, Vitiello B, Walkup J et al. (2006), SSRIs in pediatric depression: is the balance between benefits and risks favorable? J Child Adolesc Psychopharmacol 16:11Y24 Kutcher S, Robertson HA (1995), Electroconvulsive therapy in treatment resistant youth. J Child Adolesc Psychopharmacol 5:167Y176 Leon AC, Marzuk PM, Tardiff K, Teres JJ (2004), Paroxetine, other antidepressants, and youth suicide in New York City: 1993Y1998. J Clin Psychiatry 65:915Y918 Lloyd A, Horan W, Borgaro SR, Stokes JM, Pogge DL, Harvery PD (1998), Predictors of medication compliance after hospital discharge in adolescent psychiatric patients. J Child Adolesc Psychopharmacol 8:133Y141 Lopez MA, Toprac MG, Crismon ML, Boemer C, Baumgartner J (2005), A psychoeducational program for children with ADHD or depression and their families: results from the CMAP feasibility study. Commun Ment Health J 41:51Y66 Ludwig J, Marcotte DE (2005), Anti-depressants, suicide, and drug regulation. J Policy Anal Manage 24:249Y272 Mallinckrodt CH, Prakash A, Andorn AC, Watkin JG, Wohlreich MM (2006), Duloxetine for the treatment of major depressive disorder: a closer look at efficacy and safety data across the approved dose range. J Psychiatr Res 40:337Y348 Manassis K, Monga S (2001), A therapeutic approach to children and adolescents with anxiety disorders and associated comorbid conditions. J Am Acad Child Adolesc Psychiatry 40:115Y117

J. AM. ACAD. CHILD ADOLESC. PSYCH IATRY, 46:6, JUNE 2007

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

CHILDHOOD MDD MEDICATION ALGORITHM Maneeton N, Srisurapanont M (2000), Tricyclic antidepressants for depressive disorders in children and adolescents: a meta-analysis of randomized-controlled trials. JAMA Thailand 83:1367Y1374 Mann JJ, Apter A, Bertolote J et al. (2005), Suicide prevention strategies: a systematic review. JAMA 294:2064Y2074 Mann JJ, Emslie G, Baldessarini RJ et al. (2006), ACNP Task Force report on SSRIs and suicidal behavior in youth. Neuropsychopharmacology 31:473Y492 March JS, Biederman J, Wolkow R et al. (1998), Sertraline in children and adolescents with obsessive-compulsive disorder: a multicenter randomized controlled trial. JAMA 280:1752Y1756 Masand PS, Gupta S (2002), Long-term side effects of newer-generation antidepressants: SSRIS, venlafaxine, nefazodone, bupropion, and mirtazapine. Ann Clin Psychiatry 14:175Y182 Mayes TL, Rao R, Rintelmann JW, et al. (2007), Do children and adolescents have differential response rates in placebo-controlled trials of fluoxetine? CNS Spectr 12:147Y154 Molnar BE, Berkman LF, Buka SL (2001), Psychopathology, childhood sexual abuse and other childhood adversities: relative links to subsequent social behavior in the U.S. Psychol Med 31:965Y977 Mufson L, Dorta KP, Wickramaratne P, Nomura Y, Oflson M, Weissman MM (2004), A randomized effectiveness trial of interpersonal psychotherapy for depressed adolescents. Arch Gen Psychiatry 61:577Y584 Nemeroff CB, Kalali A, Keller MB (2007), Impact of publicity concerning pediatric suicidality data on physician practice patterns in the United States. Arch Gen Psychiatry Olfson M, Gameroff MJ, Marcus SC, Greenburg T, Shaffer D (2005), National trends in hospitalization of youth with intentional self-inflicted injuries. Am J Psychiatry 162:1328Y1335 Olfson M, Marcus SC, Shaffer D (2006), Antidepressant drug therapy and suicide in severely depressed children and adults. Arch Gen Psychiatry 63:865Y872 Olfson M, Shaffer D, Marcus SC (2003), Relationship between antidepressant medication treatment and suicide in adolescents. Arch Gen Psychiatry 60:978Y982 Patel NC, Crismon ML, Hoagwood K, Jensen PS (2005), Unanswered questions regarding antipsychotic use in aggressive children and adolescents. J Child Adolesc Psychopharmacol 15:270Y284 Pappadopulos E, MacIntyre JC, Crismon ML et al. (2003), Treatment recommendations for the use of antipsychotics for aggressive youth (TRAAY): II. J Am Acad Child Adolesc Psychiatry 42:145Y161 Pine DS (2002), Treating children and adolescents with selective serotonin reuptake inhibitors: how long is appropriate? J Child Adolesc Psychopharmacol 12:189Y203 Pliszka SR, Crismon ML, Hughes CW et al. (2006), The Texas Children_s Medication Algorithm Project: a revision of the algorithm for the pharmacotherapy of childhood attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry 45:642Y647 Posner K, Melvin GA, Stanley B, Oquendo MA, Gould M (2007a), Factors in the assessment of suicidality in youth. CNS Spectr 12:156Y162 Posner K, Oquendo MA, Stanley B, Davies M, Gould M (2007b), Columbia Classification Algorithm of Suicide Assessment (C-CASA): classification of suicidal behavior and related events in the FDA_s pediatric suicidal risk analysis of antidepressants. Am J Psychiatry in press Prudic J, Hackett RF, Muslin B et al. (1996), Resistance to antidepressant medication and short-term response to ECT. Am J Psychiatry 153: 985Y992 Quitkin FM, McGrath PJ, Stewart JW et al. (2005), Remission rates with 3 consecutive antidepressant trials: effectiveness for depressed outpatients. J Clin Psychiatry 66:670Y676 Rao U, Neal RD, Birmaher B et al. (1995), Unipolar depression in adolescents: clinical outcome in adulthood. J Am Acad Child Adolesc Psychiatry 34:566Y578 Rey-Sanchez F, Gutierrez-Casares JR (1997), Paroxetine in children with major depressive disorder: an open trial. J Am Acad Child Adolesc Psychiatry 36:1443Y1447 Riddle MA, Reeve EA, Yaryura-Tobia JA et al. (2001), Fluvoxamine for children and adolescents with obsessive-compulsive disorder: a

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:6, JUNE 2007

randomized, controlled, multicenter trial. J Am Acad Child Adolesc Psychiatry 40:222Y229 Riddle MA, Scahill L, King RA et al. (1992), Double-blind, crossover trial of fluoxetine and placebo in children and adolescents with obsessivecompulsive disorder. J Am Acad Child Adolesc Psychiatry 31:1062Y1069 Rosack J (2005), New data show declines in antidepressant prescribing. Psychiatr News 40:1 Rush AJ, Crismon ML, Toprac MG et al. (1998), Consensus guidelines in the treatment of major depressive disorder. J Clin Psychiatry 59: 73Y84 Rush AJ, Kupfer DJ (1995), Strategies and Tactics in the Treatment of Depression, 2nd ed., Vol. 1. Washington, DC: American Psychiatric Press Rush AJ, Rago WV, Crismon ML et al. (1999), Medication treatment for the severely and persistently mentally ill: the Texas Medication Algorithm Project. J Clin Psychiatry 60:284Y291 Rush AJ, Trivedi MH, Wisniewski SR et al. (2006), Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression. N Engl J Med 354:1231Y1242 Ryan ND, Meyer V, Dachille S, Mazzie D, Puig-Antich J (1988), Lithium antidepressant augmentation in TCA-refractory depression in adolescents. J Am Acad Child Adolesc Psychiatry 27:371Y376 Rynn MA, Siqueland L, Rickels K (2001), Placebo-controlled trial of sertraline in the treatment of children with generalized anxiety disorder. Am J Psychiatry 158:2008Y2014 Scheekloth TD, Rummans TA, Logan KM (1993), Electroconvulsive therapy in adolescents. Convulsive Ther 9:158Y166 Schur SB, Sikich L, Findling RL et al. (2003), Treatment recommendations for the use of antipsychotics for aggressive youth (TRAAY): I. A review. J Am Acad Child Adolesc Psychiatry 42:132Y144 Shaffer D, Gould MS, Fisher P et al. (1996), Psychiatric diagnosis in child and adolescent suicide. Arch Gen Psychiatry 53:339Y348 Simon GE, Savarion J, Operskalski, Wang PS (2006), Suicide risk during antidepressant treatment. Am J Psychiatry 163:41Y47 Strober M, Carlson G (1982), Bipolar illness in adolescents with major depression: clinical, genetic, and psychopharmacologic predictors in a three- to four-year prospective follow-up investigation. Arch Gen Psychiatry 39:549Y555 Strober M, Freeman R, Rigali J, Schmidt-Lackner S, Diamond R (1992), The pharmacotherapy of depressive illness in adolescence: II. Effects of lithium augmentation in nonresponders to imipramine. J Am Acad Child Adolesc Psychiatry 31:16Y20 Strober M, Rao U, DeAntonio M et al. (1988), Effects of electroconvulsive therapy in adolescents with severe endogenous depression resistant to pharmacotherapy. Biol Psychiatry 43:335Y338 Stone MB (2006), Antidepressant drugs and suicidality. Food and Drug Administration Center for Drug Evaluation and Research. Psychopharmacologic Drugs Advisory Committee Meeting. December 13, 2006. Available at: http://www.fda.gov/ohrms/dockets/ac/06/slides/20064272s1-03-FDA_files/frame.htm. Accessed March 30, 2007. Texas Department of State Health Services (2005), Children_s Medication Algorithm Project. Available at: http://www.dshs.state.tx.us/mhprograms/ CMAP.shtm. Accessed January 31, 2007 Thomas JM, Benham AL, Gean M et al. (1997), Practice parameters for the psychiatric assessment of infants and toddlers (0Y36 months). J Am Acad Child Adolesc Psychiatry 36:21SY36S Treatment for Adolescents With Depression Study (TADS) Team (2004), Fluoxetine, cognitive-behavioral therapy, and their combination for adolescents with depression: Treatment for Adolescents With Depression Study (TADS) randomized control trial. JAMA 292:807Y820 Trivedi MH, Fava M, Wisniewski SR et al. STAR*D Study Team (2006), Medication augmentation after the failure of SSRIs for depression. N Engl J Med 354:1243Y1252 Trivedi MH, Rush AJ, Crismon ML et al. (2004), Clinical results for patients with major depressive disorder in the Texas medication algorithm project. Arch Gen Psychiatry 61:669Y680 Valuck RJ (2006), SSRI antidepressant use and rates of suicide attempt among adults with MDD and an evaluation of the FDA warnings on antidepressants and suicidality on patterns of care for MDD. FDA Psychopharmacologic Drugs Advisory Committee Meeting. December

685

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

HUGHES ET AL. 13, 2006. Available at: http://www.fda.gov/ohrms/dockets/ac/06/slides/ 2006-4272OPH1-13-valuck.ppt. Accessed March 30 2007 Valuck RJ, Libby AM, Sills MR, Giese AA, Allen RR (2004), Antidepressant treatment risk of suicide attempt by adolescents with major depressive disorder: a propensity-adjusted retrospective study. CNS Drugs 18: 1119Y1132 Vitiello B (2006), An update on publicly funded multisite trials in pediatric psychopharmacology. Child Adolesc Psychiatr Clin N Am 15:1Y12 Vitiello B, Riddle MA, Greenhill LL et al. (2003), How can we improve the assessment of safety in child and adolescent psychopharmacology? J Am Acad Child Adolesc Psychiatry 42:634Y641 von Knorring A, Olsson GI, Thomsen PH, Lemming OM, Hulten A (2006), A randomized, double-blind, placebo-controlled study of citalopram in adolescents with major depressive disorder. J Clin Psychopharmacol 26:311Y315 Wagner KD, Ambrosini PJ, Rynn M et al. (2003), Efficacy of sertraline in the treatment of children and adolescents with major depressive disorder. JAMA 290:1033Y1041 Wagner KD, Berard R, Sterin MB et al. (2004a), A multicenter, randomized, double-blind,placebo-controlledtrialofparoxetineinchildren

686

and adolescents with social anxiety disorder. Arch Gen Psychiatry 61:1153Y1162 Wagner KD, Jonas J, Findling RL, Ventura D, Saikali K (2006), A doubleblind, randomized, placebo-controlled trial of escitalopram in the treatment of pediatric depression. J Am Acad Child Adolesc Psychiatry 45:280Y288 Wagner KD, Robb AS, Findling RL, Jin J, Gutierrez MM, Heydorn WE (2004b), A randomized, placebo-controlled trial of citalopram for the treatment of major depression in children and adolescents. Am J Psychiatry 161:1079Y1083 Walkup J, Labellarte M, Riddle MA et al. Research Units on Pediatric Psychopharmacology Anxiety Study Group (2002), Treatment of pediatric anxiety disorders: an open-label extension of the research units on pediatric psychopharmacology anxiety study. Adolesc Psychopharmacol 12:175Y188 Yates WR, Michell J, Rush AJ et al. (2004), Clinical features of depressed outpatients with and without co-occurring general medical conditions in STAR*D. Gen Hosp Psychiatry 26:421Y429 Zuckerbrot RA, Jensen PS (2006), Improving recognition of adolescent depression in primary care. Arch Pediatr Adolesc Med 160:694Y704

J. AM. ACAD. CHILD ADOLESC. PSYCH IATRY, 46:6, JUNE 2007

Copyright @ 2007 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.