Application for updating fluoxetine age restriction for the treatment of depression from > 8 years to >12 years

Application for updating fluoxetine age restriction for the treatment of depression from > 8 years to >12 years Submitted by Mental Health: Evidence and Research World Health Organization, Geneva

1. Summary statement of the proposal for inclusion, change or deletion Currently fluoxetine is available both on the Essential Medicine List of children and in Table 1 with an age restriction of a>8 years to be used as a complementary medicine. Depression is uncommon in early childhood and the main increase in prevalence happens in adolescence or later. Social and familial factors play a major role in depression of children and there are effective psychological and social interventions for the prevention and treatment of depression in children. Taking into consideration the lower incidence and prevalence of depression in younger children, the evidence reviews on the effectiveness or adverse effects of the medicine for younger age group, the realities of child mental health services in many low resource settings which might favour over-prescription, and the overall risk-benefit aspects; we suggest to increase of the age limit from >8 years to >12 years. 2. Name of the focal point in WHO submitting or supporting the application (where relevant) Dr M.Taghi Yasamy, Medical Officer, Department of Mental Health and Substance Abuse 3. Name of the organization(s) consulted and/or supporting the application --- 4. International Nonproprietary Name (INN, generic name) of the medicine Fluoxetine 5. Formulation proposed for inclusion; including adult and paediatric (if appropriate) Not applicable 6. International availability - sources, if possible manufacturers and trade names Not applicable 7. Whether listing is requested as an individual medicine or as an example of a therapeutic group Application is requested as an individual medicine

8. Information supporting the public health relevance (epidemiological information on disease burden, assessment of current use, target population) Prevalence of depression is lower in childhood and increases over time. It appears to be rare in children younger than six years and of lower incidence before adolescence (1). In a metaanalysis of twenty-six studies including 60,000 observations on children, overall prevalence estimates for children under 13 years old was 2.8% (SE=0.5%) and for those between 13 18 years old it was 5.6% (SE=0.3%) which is close to adult figures (2). A sharp increase in incidence after childhood has been described(3). Mental disorders including depression remain under-treated especially in low and middle income countries. In a WHO study in 42 countries, the median estimated one-year treated prevalence of all mental disorders for children and adolescents is 159 per 100,000 population compared to a treated prevalence of 664 per 100,000 for the adult population (4). Treatment gap for depression across all ages is overall about 50% in high income and 70% in low and middle income countries which in some countries reaches 90% (5,6). The quality of treatment of depression in children is extremely important. The presentation of depression is often more complex in children as compared with adults. Many children and adolescents with depression present with irritability, bodily complaints or problems of behaviour or conduct (7). Because of their limited language abilities, the source of information is usually a parent, a teacher, another child etc, and there is not sufficient data on direct screening of depression in children (8). Depression in children is more often than not related to life adversities and maltreatment than to biological factors (9,10) and requires psychosocial and parental interventions (11). A recent meta-analysis reconfirmed that depression consequent to child maltreatment can take a different course and is more difficult to treat (12). Children who are still in development - are also more susceptible to adverse effects of medicines and many second generation antidepressants have been shown to have an unfavourable risk benefit profile for children and adolescents than for adults. This includes increased suicidal behavior and ideas (13). Antidepressants have been largely overused in children especially after the advent of SSRIs (14). There is a concerning history of people receiving anti-depressants without indication (15, 16). 9. Treatment details (dosage regimen, duration; reference to existing WHO and other clinical guidelines; need for special diagnostics, treatment or monitoring facilities and skills) WHO s mental health Gap Action Programme Intervention Guide (mhgap-ig) was published in 2010 based on WHO guidelines, which were based on review of a series of evidence reviews conducted in 2009 (see http://www.who.int/entity/mental_health/mhgap/evidence/child/q10/en/index.html). The Guideline Development Group came up with the following strong recommendation:

Antidepressants (Tricyclic antidepressants (TCA), Selective serotonin reuptake inhibitors (SSRI)) should not be used for the treatment of children 6-12 years of age with depressive episode/disorder in non-specialist settings. (17,18). Then it adds a standard (weak) recommendation : Fluoxetine, but not Tricyclic antidepressants (TCA) or other Selective serotonin reuptake inhibitors (SSRI), may be considered as one possible treatment of adolescents with depressive episode in non-specialist settings. Adolescents on fluoxetine should be monitored closely for suicide ideas/behaviour. For all adolescents on fluoxetine, support and supervision from a mental health specialist should be obtained, if available. 10. Summary of comparative effectiveness in a variety of clinical settings: The mhgap guideline development group concluded the following (see Appendix 1 for more details, also available on WHO website) (15): In children 6-12 years of age with depressive episode/disorder in terms of proportion of children showing an improvement in depressive symptoms, the evidence for tricyclic antidepressants (TCAs) versus placebo is inconclusive and so it is not possible to determine if there is a clinically important difference (RR 0.86, 0.25 to 1.89). Similarly, in terms of score on a depression measure the evidence for tricyclic antidepressants (TCAs) versus placebo is inconclusive (SMD 0.15, 0.34 to 0.64). For most SSRIs, the evidence is sparse and so it is not possible to determine if there is a clinically important difference between individual SSRIs and placebo. For fluoxetine, in terms of responders and in terms of score on a depression measure, there is limited evidence suggesting a significant beneficial effect. In terms of treatment acceptability and adverse effects, the evidence is sparse and inconclusive. In terms of suicide ideas/behaviour, for the group of selective serotonin reuptake inhibitors, including children and adolescents together, there is evidence of a significant increased risk (RR 1.73, 1.13 to 2.67, absolute risk difference 2.5%). For the other critical outcomes, no evidence is available. The rest of the evidence profile elaborated on safety concerns, difficulty to differentiate clinical depressive disorder from transient reaction and the risk of medicalization of a social problem. Also the facts that children are still in development, cultural issues and possibility of incorrect diagnosis were highlighted in the evidence profile. For the above reasons, the GDG came up with a STRONG recommendation to set the limit of 12 years as mentioned above. FDA

Fluoxetine is the only medication approved by the FDA for treating depression in children 8 years and older. In December 2006, the FDA s Psychopharmacologic Drugs Advisory Committee asked for labeling changes and to inform health care professionals about the increased risk of suicidality in younger adults using antidepressants (17). The black box warning also notes that children and adolescents under SSRI medications should be closely monitored especially during the first four weeks of treatment. It also mentioned that SSRI medications usually have few side effects in children and adolescents, but for unknown reasons, they may trigger agitation and abnormal behavior in certain individuals.... FDA included fluoxetine among other SSRIs to be included in the labeling changes for the medications despite agreement to approve it for use in children and adolescents (17). NICE guidelines NICE has adopted a cautious approach in 2005 and did not recommend medicines for the treatment of moderate to severe depression in young people unless the following conditions are met. It recommends that after intensively reviewing the case, they should offer a specific psychological therapy for at least 3 months. Then if depression still unresponsive to treatment after 4-6 sessions, to carry out a multidisciplinary review. Then they suggest to consider two options: alternative or perhaps additional psychological therapy for the parent and other family members and the patient and then comes considering medication. They suggest that for young people 12-18 years to offer fluoxetine in addition to psychological therapy and for children aged 5-11 years to cautiously consider the addition of fluoxetine. They mention that evidence for its effectiveness in this age group is not established (20). Systematic reviews since mhgap evidence review of 2009: - Williams SB et al 2009 (21). They reviewed all clinical trials with fair and good quality on screening, psychotherapy and SSRIs for children age (7-18) in primary care. Unfortunately the age range was wide and only one trial focused on age 9-12 which was on group psychotherapy and not applicable to our purpose. - Hetrick SE et al, 2012 published their updated Cochrane Review on Newer generation antidepressants for depressive disorders in children and adolescents (22). We had reviewed their previous systematic review in 2009 at times of the mhgap guidelines development. In 2012, Hetrick and colleagues added a Spanish trial of 2005 ( data only) not included before (23). In the systematic review, they considered two age ranges: (1) adolescents with an age range of 12 or 13 to 17 or 18 and (2) a mixed age group for children and adolescents with a lower age range of six to eight years. In their new update, Hetrick et al did not change their conclusions and stated that with these limitations, it is difficult to answer questions about the effectiveness and safety of antidepressants for treating depression in children and adolescents. They also highlight that health providers should provide accurate information to children and adolescents, and their families, about the uncertainties regarding the benefits and risks of newer generation

antidepressant medication as a treatment option for depression. If a decision to use medication is agreed then fluoxetine might be the medication of first choice given guideline recommendations and, if used, the risk of suicide should be assessed and monitored particularly closely. - Cox et al, 2012a (24) conducted a recent systematic review focusing on relapse prevention as an outcome and found only 9 trials. They report that the blinding and allocation concealment has been unclear. The authors communicate that they were unable to make a conclusion and called for more trials. - Cox et al, 2012b (25). They included ten studies to compare psychological therapies alone and in combination for depression versus antidepressants, involving 1235 participants. Their participants were children and adolescents aged eight to 17 years in one trial and adolescents over the age of 11 in nine trials. They were unable to conclude about the effectiveness of individual treatments and preference of any treatment over the other and asked for additional trials. Overall, it seems that few trials have involved young children in their studies (26). According to our reviews, no additional evidence of a different direction has emerged since our previous evidence review on the effectiveness of SSRIs including fluoxetine for children under 12. 11. Summary of comparative evidence on safety: 11.1. Estimate of total patient exposure to date Fluoxetine is the first new generation antidepressant. Millions of people with depression and some other mental disorders have received it since 1974. Only in UK nearly 6 million prescriptions included fluoxetine in 2011(27). Over time children with depression were also started on this medication. We do not have statistics on the number of children and adolescents who have used it but it should be considerably high too. Even since 2004, despite a general decrease in the use of antidepressants in younger people in US (28), it seems that it has been the the only medicine certified for use in this age group in many countries (3,20). 11.2. Description of adverse effects/reactions From WHO Model formulary for children, 2010(26) Clinical worsening of depression or suicidal ideation may occur. Adverse effects: Common:

Nausea, agitation, insomnia, drowsiness, tremor, dry mouth, diarrhoea, dizziness, headache, sweating, weakness, anxiety, weight gain or loss, sexual dysfunction, rhinitis, myalgia, rash, chills, euphoria, yawning. Uncommon: Extrapyramidal reactions (including tardive dyskinesia and dystonia), sedation, confusion, palpitations, tachycardia, hypotension, hyponatraemia (as part of syndrome of inappropriate antidiuretic hormone secretion), abnormal platelet aggregation/haemorrhagic complications (e.g. bruising, epistaxis, gastrointestinal and vaginal bleeding), alopecia, changes in blood sugar, serotonin syndrome. Rare: Elevated liver enzymes, hepatitis, hepatic failure, galactorrhoea, blood dyscrasias, seizures, akathisia, paraesthesia, taste disturbance, toxic epidermal necrolysis and neuroleptic malignant syndrome. 11.3. Identification of variation in safety due to health systems and patient factors 11.4. Summary of comparative safety against comparators In mhgap review of 2009 the data on the relative risk of suicidal behavior attributed to fluoxetine was judged to be difficult to interpret because of wide confidence intervals, although the related adverse events and suicide behaviour/ideation was higher in fluoxetine than placeborelated patients. However, fluoxetine was shown to be less likely than placebo to lead to discontinuation of treatment for any reason. Hyperkinesias, headache and skin rash were more common in fluoxetine than with placebo. - Williams SB et al 2009(21). See the limitations mentioned under 10. - Hetrick SE et al, 2012, see under 10. They reported an increased risk (64%) of suiciderelated outcomes for those on antidepressants compared with those given placebo. Rates of adverse events was higher for those receiving antidepressant. There was no evidence that one particular type of newer generation antidepressant had a larger effect than the others when compared to placebo. (22). The additional trial that they added in their recent update (23) did not address adverse events. - Cox et al, 2012 a, see 10.(24) The majority of trials that involved antidepressant medication reported adverse events including suicide-related behaviours. However, there were not enough data to show which treatment approach results in the most favourable adverse event profile.

- Cox et al, 2012 b (25), see 10. They focused on suicide related adverse events but were not able to combine the different studies as this outcome was not reported in the same way in different trials. Suicidal ideation was found measurable in one study and was significantly higher in the antidepressant medication group compared with the psychological therapy Group. Overall, systematic reviews since mhgap review of 2009 have not offered a different safety profile for fluoxetine for children below age of 12. This has also been highlighted in the past by some other groups of reviewers indicating that to conclude for recommending fluoxetine for younger children we still need additional studies (29,30). Though psychotherapy is at least in the short term effective for youth with depression in general, again it has been shown to be more effective in adolescents than children (11). After the March 2004 Food and Drug Administration (FDA) warning on antidepressant suicidality indicating the possible risk of increased suicidal behaviour and ideas in children and adolescents which excluded fluoxetine, an overall decrease in antidepressant treatment for youth was reported and psychotherapy without medication increased (28). In developing countries mental health care to children is predominantly delivered through primary care or by general psychiatrists (not specialized in children) and pediatricians (31). According to a WHO survey in EMRO, in 67% of the countries the majority of service is provided by general psychiatrists, paediatricians and primary care physicians or non-physician primary health care workers. A considerable proportion of them may not have the skills and the time to communicate with the child, parents and school or to consider alternative treatments before reverting to medicines (32). Currently fluoxetine is a complementary medicine for children >8 years in WHO essential list of medicines for children. It seems that moving this age to > 12 years would be a more prudent risk-benefit decision. 12. Summary of available data on comparative cost** and cost-effectiveness within the pharmacological class or therapeutic group: Psychotherapy is somewhat more costly than pharmacotherapy for UK inpatients care (33). However, the situation is different in low and middle income countries, where staff costs are much lower. Structured psychological interventions delivered through community and primary care providers have been shown to be effective and low cost (34,35,36,37) and have been recommended in mhgap. 13. Summary of regulatory status of the medicine (in country of origin, and preferably in other countries as well) Not applicable 14. Availability of pharmacopoeial standards (British Pharmacopoeia, International Pharmacopoeia, United States Pharmacopoeia)

Not applicable 15. Proposed (new/adapted) text for the WHO Model Formulary No adaptation needs to be made to the adult formulary if the decision is taken to move it to the adult list, except the following: Clinical worsening of depression or suicidal ideation may occur especially in adolescents and young adults up to 24. Depression and certain other mental disorders are themselves associated with increased risk of suicide. Special Notes: WHO age restriction: > 12 years. Acknowledgement: I appreciate comments from Drs M.Van Ommeren, C. Servili and S. Saxena. References: 1) Sharp LK, Lipsky MS. Screening for depression across the lifespan: a review of measures for use in primary care settings. Am Fam Physician. 2002 Sep 15;66(6):1001 8. 2) Jane Costello E, Erkanli A, Angold A. Is there an epidemic of child or adolescent depression? J Child Psychol Psychiatry. 2006 Dec;47(12):1263 71. 3) FDA approved Medication Guide, Highlights of prescribing information, PROZAC (fluoxetine) www.accessdata.fda.gov/drugsatfda_docs/label/2009/018936s075s077lbl.pdf accessed 12 Dec 2012 4) Morris J, Belfer M, Daniels A, Flisher A, Villé L, Lora A, et al. Treated prevalence of and mental health services received by children and adolescents in 42 low-and-middle-income countries. Journal of Child Psychology and Psychiatry. 2011;52(12):1239 46. 5) Kohn R, Saxena S, et al. The treatment gap in mental health care. Bull World Health Organ 2004 Nov;82(11):858-66. Epub 2004 Dec 14. 6) World Health Organization, Depression Fact sheet N 369,October 2012 http://www.who.int/mediacentre/factsheets/fs369/en/index.html Accessed 12 December 2012 7) IACAPAP Textbook of Child and Adolescent Mental Health, 2012. iacapap.org/wpcontent/uploads/e.1-depression-072012.pdf, accessed 10 December 2012.

8) Williams SB, O Connor E, Eder M, Whitlock E. Screening for Child and Adolescent Depression In Primary Care Settings [Internet]. 2009 Available from: http://www.ncbi.nlm.nih.gov/books/nbk35127/ Accessed 12 December 2012. 9) Liu RT, Alloy LB. Stress generation in depression: A systematic review of the empirical literature and recommendations for future study. Clinical psychology review. 2010;30(5):582 93. 10) Levinson DF. The Genetics of Depression: A Review. Biological Psychiatry. 2006 Jul 15;60(2):84 92. 11) Watanabe N, Hunot V, Omori IM, Churchill R, Furukawa TA. Psychotherapy for depression among children and adolescents: a systematic review. Acta Psychiatrica Scandinavica. 2007;116(2):84 95. 12) Nanni V, Uher R, Danese A. Childhood maltreatment predicts unfavorable course of illness and treatment outcome in depression: a meta-analysis. Am J Psychiatry. 2012 Feb;169(2):141 51. 13) Whittington CJ, Kendall T, Fonagy P, Cottrell D, Cotgrove A, Boddington E. Selective serotonin reuptake inhibitors in childhood depression: systematic review of published versus unpublished data. The Lancet. 2004 Apr 24;363(9418):1341 5. 14) Jureidini J, Tonkin A. Overuse of antidepressant drugs for the treatment of depression. CNS Drugs. 2006;20(8):623 32. 15) Esposito E, Wang JL, Adair CE, Williams JV, Dobson K, Schopflocher D, et al. Frequency and adequacy of depression treatment in a Canadian population sample. Can J Psychiatry 2007; 52: 780 9. 16) Demyttenaere K, Bonnewyn A, Bruffaerts R, De Girolamo G, Gasquet I, Kovess V, Haro JM, et al. Clinical factors influencing the prescription of antidepressants and benzodiazepines: results from the European study of the epidemiology of mental disorders (ESEMeD). J Affect Disord 2008; 110: 84 93. 17) WHO mhgap Evidence Resource Centre www.who.int/mental_health/mhgap/evidence/en/ accessed 24 November 2012 18) mhgap Intervention Guide for mental, neurological and substance use disorders in non-specialized health settings http://www.who.int/mental_health/publications/mhgap_intervention_guide/en/index.html accessed 24 November 2012 19) FDA approved Medication Guide, PROZAC (PRO-zac),(fluoxetine hydrochloride)http://www.accessdata.fda.gov/scripts/cder/drugsatfda/ accessed 12 Dec 2012 20) National Institute for Health and Clinical Excellence (NICE), Depression in children and young people, Clinical Guideline 28, September 2005. 21) Williams SB, O Connor EA, Eder M, Whitlock EP. Screening for Child and Adolescent Depression in Primary Care Settings: A Systematic Evidence Review for the US Preventive Services Task Force. Pediatrics. 2009 Apr 1;123(4):e716 e735. 22) Hetrick SE, McKenzie JE, Cox GR, Simmons MB, Merry SN. Newer generation antidepressants for depressive disorders in children and adolescents. In: The Cochrane Collaboration, Hetrick SE, editors. Cochrane Database of Systematic Reviews [Internet]. Chichester, UK: John Wiley & Sons, Ltd; 2012 [cited 2012 Dec 4]. Available from: http://summaries.cochrane.org/cd004851/newerantidepressants-for-depression-in-children-and-adolescents 23) Friederichsen A, Montes LGA. Tratamiento del trastorno depresivo mayor con fluoxetina en niños y adolescentes. Estudio doble ciego, controlado con placebo. Psiquiatría biológica: Publicación oficial de la Sociedad Española de Psiquiatría Biológica. 2005;12(5):198 205 (abstract)

24) Cox GR, Fisher CA, De Silva S, Phelan M, Akinwale OP, Simmons MB, et al. Interventions for preventing relapse and recurrence of a depressive disorder in children and adolescents. In: The Cochrane Collaboration, Cox GR, editors. Cochrane Database of Systematic Reviews [Internet]. Chichester, UK: John Wiley & Sons, Ltd; 2012 [cited 2012 Dec 4]. Available from: http://summaries.cochrane.org/cd007504/treatments-for-preventing-the-recurrence-of-depressionin-children-and-adolescents 25) Cox GR, Callahan P, Churchill R, Hunot V, Merry SN, Parker AG, et al. Psychological therapies versus antidepressant medication, alone and in combination for depression in children and adolescents. In: The Cochrane Collaboration, Cox GR, editors. Cochrane Database of Systematic Reviews [Internet]. Chichester, UK: John Wiley & Sons, Ltd; 2012 [cited 2012, Dec 5]. Available from: http://summaries.cochrane.org/cd008324/psychological-therapies-versus-antidepressantmedication-alone-and-in-combination-for-depression-in-children-and-adolescents 26) Sakolsky D, Birmaher B. Developmentally informed pharmacotherapy for child and adolescent depressive disorders. Child Adolesc Psychiatr Clin N Am. 2012;21(2):313 325, viii. 27) BBC - Health: Prozac [Internet]. Available from: http://www.bbc.co.uk/health/emotional_health/addictions/prozac.shtml accessed 14 October 2012 28) Valluri S, Zito JM, Safer DJ, Zuckerman IH, Mullins CD, Korelitz JJ. Impact of the 2004 Food and Drug Administration pediatric suicidality warning on antidepressant and psychotherapy treatment for new-onset depression. Med Care. 2010;48(11):947 54. 29) Tsapakis EM, Soldani F, Tondo L, Baldessarini RJ. Efficacy of antidepressants in juvenile depression: meta-analysis. BJP. 2008;193(1):10 7. 30) Brendt D. Effective treatments for suicidal youth, pharmacological and psychosocial approaches. In D. Wasserman and C. Wasserman (Editors), Suicidology and suicide prevention, Oxford University Press Inc, New York, 2009: 671-9 31) World Health Organization. ATLAS, child and adolescent mental health resources, WHO, Geneva, 2005. http://www.who.int/entity/mental_health/resources/child_ado_atlas.pdf accessed 1 Dec, 2012. 32) World Health Organization, Eastern Mediterranean Region, 2011. Atlas: child, adolescent, and maternal mental health resources in the Eastern Mediterranean Region. http://www.emro.who.int/mental-health/publications/atlas-and-reports.html accessed 1 Dec 2012. 33) Knapp M, Ilson S. Economic aspects of depression and its treatment : Current Opinion in Psychiatry [Internet]. Available from: http://journals.lww.com/copsychiatry/fulltext/2002/01000/economic_aspects_of_depression_and_its_treatment.12.as px. Accessed 14 Dec 2012. 34) Patel V, Simon G, Chowdhary N, Kaaya S, Araya R. Packages of Care for Depression in Low- and Middle-Income Countries. PLoS Med. 2009 Oct 6;6(10):e1000159. 35) Rahman A, Malik A, Sikander S, Roberts C, Creed F. Cognitive behaviour therapy-based intervention by community health workers for mothers with depression and their infants in rural Pakistan: a cluster-randomised controlled trial. Lancet. 2008 Sep 13;372(9642):902 9. 36) Rahman A. Challenges and opportunities in developing a psychological intervention for perinatal depression in rural Pakistan--a multi-method study. Arch Womens Ment Health. 2007;10(5):211 9.

37) Bolton P BJ. Group interpersonal psychotherapy for depression in rural uganda: A randomized controlled trial. JAMA. 2003 Jun 18;289(23):3117 24.

Q10: Are antidepressants (Tricyclic antidepressants (TCA), Selective serotonin reuptake inhibitors (SSRIs)) effective and safe in children 6 12 years of age with depressive episode/disorder? Background Compared with adult depression, depression in children (6 12 years) may have a more insidious onset, may be characterized more by irritability than sadness, and occurs more often in association with other conditions such as anxiety, conduct disorder, hyperkinesis, and learning problems. The prevalence of major depression is estimated to be approximately 2% in children. In children with depression, pharmacological treatment is based on antidepressant drugs, initially tricyclic antidepressants and more recently selective serotonin reuptake inhibitors (SSRIs). Although the population to be covered in this scoping question includes children only, it is anticipated that some randomized controlled trials, and some meta analyses included in systematic reviews, were carried out in the group of children and adolescents considered together. The body of evidence is therefore presented for children only whenever possible, and for children and adolescents considered together if no data are available for adolescents only. Population/Intervention(s)/Comparator/Outcome(s) (PICO) Population: Interventions: Comparator: Outcomes: children (6 12 years) with depressive episode/disorder antidepressant drugs: Tricyclics and related, selective serotonin reuptake inhibitors placebo symptom reduction overall performance at school 1

family functioning adverse effects of treatment improvement in physical health user and family satisfaction reduction in risk behaviour List of the systematic reviews identified by the search process INCLUDED IN GRADE TABLES Hazell P et al (2002). Tricyclic drugs for depression in children and adolescents. Cochrane Database of Systematic Reviews, (2):CD002317. Last assessed as upto date: 11 February 2008. Hetrick SE et al (2007). Selective serotonin reuptake inhibitors (SSRIs) for depressive disorders in children and adolescents. Cochrane Database of Systematic Reviews, (3):CD004851. EXCLUDED FROM GRADE TABLES Hazell P (2009). Depression in children and adolescents. BMJ Clinical Evidence, 01:1008. NICE (2005). Depression in Children and Young People. Identification and management in primary, community and secondary care. National Clinical Practice Guideline Number 28. London: The British Psychological Society & The Royal College of Psychiatrists. Tsapakis EM et al (2008). Efficacy of antidepressants in juvenile depression: meta analysis. British Journal of Psychiatry, 193:10 17. Whittington CJ et al (2004). Selective serotonin reuptake inhibitors in childhood depression: systematic review of published versus unpublished data. Lancet, 363:1341 5. 2

Usala T et al (2008). Randomized controlled trials of selective serotonin reuptake inhibitors in treating depression in children and adolescents: a systematic review and meta analysis. European Neuropsychopharmacology, 18:62 73. Dubicka B, Hadley S, Robertrs C (2006). Suicidal behaviour in youths with depression treated with new generation antidepressants. Meta analysis. British Journal of Psychiatry, 189:393 8. Hammad TA, Laughren T, Racoosin J 2006). Suicidality in paediatric patients treated with antidepressant drugs. Archives of General Psychiatry, 63:332 9. PICO table Serial no. Intervention/Comparison Outcomes Systematic reviews used for GRADE 1 Tricyclics and related vs. Symptoms Hazell et al, 2002 (update 2008) placebo Overall performance at school No data Explanation Hazell et al, 2002 (update 2008) is a recently updated Cochrane Review Family functioning No data Adverse effects of treatment (acceptability profile) No data Improvement in physical health No data User and family satisfaction No data 3 Individual SSRI vs. placebo Reduction in risk behaviour Symptoms Overall performance at school No data Hetrick et al (2007) No data Hetrick et al (2007) is a recently published Cochrane Review No data 3

Family functioning Adverse effects of treatment (acceptability profile) Improvement in physical health User and family satisfaction Hetrick et al (2007) No data No data Reduction in risk behaviour Narrative description of the studies that went into the analysis According to Hazell et al (2002) Cochrane Review on tricyclic antidepressants, thirteen studies reported data in a manner that could be extracted and pooled. Seven trials were directed to adolescents (aged 12 years and over), four trials were directed to children (aged 12 years and under) and two trials involved subjects spanning childhood and adolescence. Participants were outpatients in seven trials, inpatients of child or adolescent psychiatric units in five trials, and a mixture of inpatients and outpatients in one trial. Five trials involved imipramine, four trials involved amitriptyline, two trials involved desipramine and two trials involved nortriptyline. The control treatment in all cases was inactive placebo. Follow up intervals ranged from 4 weeks to 10 weeks. The shorter follow up intervals are arguably insufficient to adequately determine treatment responsiveness. According to Hetrick Cochrane Review, of 12 trials comparing one of the SSRIs with placebo were eligible for inclusion, and 10 of these included data that could be extracted and pooled in one or more meta analyses. There were three trials of paroxetine, four trials of fluoxetine, two trials of citalopram, one of escitalopram and two trials of sertraline. There were five trials in adolescents with an age range of 12 or 13 to 17 or 18, and seven in children and adolescents with a lower age limit of between 6 8 years. The treatment period of the included trials was between 7 and 12 weeks. All trials were of major depressive disorder and all except one stated that diagnoses were based on a structured clinical interview such as the K SADS P & L. Authors of all reports, except one describe depressive disorder symptom severity at baseline for the treatment and placebo groups. Mean severity scores at baseline from the individual trials range from 54.5 to 65.5 on the CDRS R (range 17 113) and from 25.9 to 32.5 on the K SADS 9 item depression score (range 9 56). GRADE tables Table 1 4

Author(s): Corrado Barbui, Taghi Yasamy Date: 2009 05 25 Question: Should tricyclic antidepressants vs. placebo be used for children with depression? Settings: Bibliography: Hazell P et al (2002). Tricyclic drugs for depression in children and adolescents. Cochrane Database of Systematic Reviews, (2):CD002317. Last assessed as up to date: 11 February 2008. No of studies Quality assessment Design Limitations Inconsistency Indirectness Imprecision Other considerations No of patients tricyclic antidepressants placebo Summary of findings Relative (95% CI) Effect Absolute Quality Importance lack of response 2 1 randomized trials serious 2 no serious inconsistency no serious indirectness very serious 3 none 25/39 (64.1%) 27/38 (71.1%) OR 0.69 (0.25 to 1.89) 82 fewer per 1000 (from 330 fewer to 112 more) VERY LOW depressive symptoms (Better indicated by lower values) 3 4 randomized trials serious 5 no serious inconsistency no serious indirectness serious 3 none 34 31 SMD 0.15 higher (0.34 lower to 0.64 higher) LOW overall functioning (Better indicated by lower values) 0 no evidence available none 0 0 MD 0 higher (0 to 0 higher) adverse effects 0 no evidence available none 0/0 (0%) 0/0 (0%) Not estimable 0 fewer per 1000 (from 0 fewer to 0 fewer) user and family satisfaction (Better indicated by lower values) 0 no evidence available none 0 0 MD 0 higher (0 to 0 higher) IMPORTANT 5

1 From Analysis 1.4 of Hazell et al (2002) Cochrane Review. 2 Unclear dropout in one study, unclear randomization and allocation concealment. 3 Less than 100 patients are included in the analysis, and the estimate ranges from appreciable benefit to appreciable harm. 4 From Analysis 1.6 of Hazell Cochrane Review. 5 Unclear dropouts in two studies, unclear randomization and allocation concealment. Table 2 Author(s): Corrado Barbui, Taghi Yasamy Date: 2009 05 25 Question: Should citalopram vs. placebo be used for children with depression? Settings: Bibliography: Hetrick SE et al (2007). Selective serotonin reuptake inhibitors (SSRIs) for depressive disorders in children and adolescents. Cochrane Database of Systematic Reviews, (3):CD004851. No of studies Quality assessment Design Limitations Inconsistency Indirectness Imprecision Other considerations No of patients citalopram placebo Relative (95% CI) Summary of findings Effect Absolute Quality Importance Response 0 no evidence available none 0/0 (0%) 0/0 (0%) Not estimable 0 fewer per 1000 (from 0 fewer to 0 fewer) depressive symptoms (Better indicated by lower values) 1 1 randomized trials serious 2 no serious inconsistency serious 3 serious 4 none 0 5 0 5 MD 0.05 lower (6.02 lower to 5.92 higher) VERY LOW overall functioning (Better indicated by lower values) 1 6 randomized trials serious 2 no serious inconsistency serious 3 serious 4 none 0 5 0 5 MD 1.8 lower (6.89 lower to 3.29 higher) VERY LOW 6

adverse effects 0 no evidence available none 0/0 (0%) 0/0 (0%) Not estimable 0 fewer per 1000 (from 0 fewer to 0 fewer) user and family satisfaction (Better indicated by lower values) 0 no evidence available none 0 0 MD 0 higher (0 to 0 higher) IMPORTANT 1 From Analysis 5.2 of Hetrick et al (2007) Cochrane Review. 2 Dropout rate exceeds 30%. 3 Only one study contributed to the analysis. 4 Estimate ranges from appreciable benefit to appreciable harm. 5 Not reported. 6 From Analysis 5.2 of Hetrick et al (2007) Cochrane Review. Table 3 Author(s): Corrado Barbui, Taghi Yasamy Date: 2009 05 25 Question: Should fluoxetine vs. placebo be used for children with depression? Settings: Bibliography: Hetrick SE et al (2007). Selective serotonin reuptake inhibitors (SSRIs) for depressive disorders in children and adolescents. Cochrane Database of Systematic Reviews, (3):CD004851. No of studies Quality assessment Design Limitations Inconsistency Indirectness Imprecision Other considerations No of patients fluoxetine placebo Summary of findings Relative (95% CI) Effect Absolute Quality Importance Response 1 1 randomized trials no serious limitations no serious inconsistency very serious 2 serious 3 None 27/61 (44.3%) 10/55 (18.2%) RR 2.43 (1.3 to 4.56) 260 more per 1000 (from 55 more to 647 more) VERY LOW 7

depressive symptoms (Better indicated by lower values) 2 4 randomized trials very serious 5 no serious inconsistency no serious indirectness no serious imprecision None 0 6 0 6 MD 6.72 lower (10.55 to 2.88 lower) LOW overall functioning (Better indicated by lower values) 1 7 randomized trials very serious 8 no serious inconsistency serious 2 serious 9 None 0 6 0 6 MD 3.76 higher (3.19 lower to 10.71 higher) VERY LOW adverse effects 0 no evidence available None 0/0 (0%) 0/0 (0%) RR 0 (0 to 0) 0 fewer per 1000 (from 0 fewer to 0 fewer) user and family satisfaction (Better indicated by lower values) 0% 0 fewer per 1000 (from 0 fewer to 0 fewer) 0 no evidence available None 0 0 MD 0 higher (0 to 0 higher) IMPORTANT 1 From Analysis 3.1 of Hetrick et al (2007) Cochrane Review. 2 Only one study contributed to the analysis. 3 Less than 100 patients are included in the analysis. 4 From Analysis 3.2 of Hetrick et al (2007) Cochrane Review. 5 One study has a dropout rate exceeding 30%. 6 Not reported. 7 From Analysis 3.3 of Hetrick et al (2007) Cochrane Review. 8 Dropout rate exceeds 30%. 9 Estimate ranges from appreciable benefit to appreciable harm. Table 4 Author(s): Corrado Barbui, Taghi Yasamy Date: 2009 05 25 Question: Should paroxetine vs. placebo be used for children with depression? Settings: Bibliography: Hetrick SE et al (2007). Selective serotonin reuptake inhibitors (SSRIs) for depressive disorders in children and adolescents. Cochrane Database of Systematic Reviews, (3):CD004851. 8

No of studies Quality assessment Design Limitations Inconsistency Indirectness Imprecision Other considerations No of patients paroxetine placebo Summary of findings Relative (95% CI) Effect Absolute Quality Importance Response 1 1 randomized trials no serious limitations no serious inconsistency serious 2 serious 3 none 22/49 (44.9%) 22/47 (46.8%) RR 0.96 (0.62 to 1.48) 19 fewer per 1000 (from 178 fewer to 225 more) LOW depressive symptoms (Better indicated by lower values) 1 4 randomized trials no serious limitations no serious inconsistency serious 2 serious 5 none 0 6 0 6 MD 5.27 higher (0.00 to 10.54 higher) LOW overall functioning (Better indicated by lower values) 0 no evidence available none 0 0 MD 0 higher (0 to 0 higher) adverse effects 1 randomized trials 7 no serious limitations no serious inconsistency serious 2 serious 3 none 34/49 (69.4%) 30/47 (63.8%) RR 1.09 (0.82 to 1.44) 57 more per 1000 (from 115 fewer to 281 more) LOW user and family satisfaction (Better indicated by lower values) 0 no evidence available none 0 0 MD 0 higher (0 to 0 higher) IMPORTANT 1 From Analysis 1.1 of Hetrick et al (2007) Cochrane Review. 2 Only one study contributed to the analysis. 3 Estimate ranges from appreciable benefit to appreciable harm. 4 From Analysis 1.2 of Hetrick et al (2007) Cochrane Review. 5 Estimate rangesa from appreciable harm to no difference. 6 Not reported. 7 From Analysis 1.8 of Hetrick et al (2007) Cochrane Review. 9

Table 5 Author(s): Corrado Barbui, Taghi Yasamy Date: 2009 05 25 Question: Should sertraline vs. placebo be used for children with depression? Settings: Bibliography: Hetrick SE et al (2007). Selective serotonin reuptake inhibitors (SSRIs) for depressive disorders in children and adolescents. Cochrane Database of Systematic Reviews, (3):CD004851. No of studies Quality assessment Design Limitations Inconsistency Indirectness Imprecision Other considerations No of patients sertraline placebo Relative (95% CI) Summary of findings Effect Absolute Quality Importance Response 0 no evidence available None 0/0 (0%) 0/0 (0%) Not estimable 0 fewer per 1000 (from 0 fewer to 0 fewer) depressive symptoms (Better indicated by lower values) 2 1 randomized trials no serious limitations no serious inconsistency serious 2 serious 3 None 0 4 0 4 MD 2.34 lower (7.01 lower to 2.33 higher) LOW overall functioning (Better indicated by lower values) 0 no evidence available None 0 0 MD 0 higher (0 to 0 higher) adverse effects 0 no evidence available None 0/0 (0%) 0/0 (0%) Not estimable 0 fewer per 1000 (from 0 fewer to 0 fewer) user and family satisfaction (Better indicated by lower values) 0 no evidence available None 0 0 MD 0 higher (0 to 0 higher) IMPORTANT 10

1 From Analysis 4.2 of Hetrick et al (2007) Cochrane Review. 2 Both included studies are from the same research group so directness may be a problem. 3 Estimate ranges from appreciable benefit to appreciable harm. 4 Not reported. Table 6 Author(s): Corrado Barbui, Taghi Yasamy Date: 2009 06 16 Question: Should selective serotonin reuptake inhibitors vs. placebo be used for children with depression? Settings: Bibliography: Hetrick SE et al (2007). Selective serotonin reuptake inhibitors (SSRIs) for depressive disorders in children and adolescents. Cochrane Database of Systematic Reviews, (3):CD004851. No of studies Quality assessment Design Limitations Inconsistency Indirectness Imprecision Other considerations No of patients selective serotonin reuptake inhibitors placebo Summary of findings Relative (95% CI) Effect Absolute Quality Importance Suicide ideas / behaviour (children and adolescents, 6 18 years) 9 1 randomized trials very serious 2 no serious inconsistency serious 3 no serious imprecision none 57/978 (5.8%) 30/886 (3.4%) RR 1.73 (1.13 to 2.67) 25 more per 1000 (from 4 more to 57 more) VERY LOW 1 From Analysis 6.2 of Hetrick et al (2007) Cochrane Review. 2 Four studies reported a dropout rate higher than 30%. 3 Children and adolescents are considered together. Additional information that was not GRADEd (safety and tolerability issues, cost, resource use, and other feasibility issues, if appropriate) From NICE (2005): Tricyclic antidepressants In children and young people, it is unlikely that tricyclic antidepressants have clinically important benefits over placebo for remission, response to treatment (50% reduction in symptoms) or reduction in symptoms. At least in young people, there is limited evidence that Tricyclics produce more side effects than 11

placebo and are more likely to lead to discontinuation of treatment. It is also known that tricyclic antidepressants (except lofepramine) are highly toxic in overdose. Fluoxetine (SSRI) Fluoxetine (up to 40 mg/day for 7 to 12 weeks) showed efficacy across a range of outcomes in 7 18 year olds. When compared with placebo, fluoxetine produced clinically important improvement in depressive symptoms (when measured with a clinician completed rating scale) and improved the likelihood of both remission and response to treatment, and had a positive impact regarding general clinical improvement and the severity of depression. Evidence is inconclusive regarding the impact on functional status. The relative risk of serious adverse events and suicidal behaviour is difficult to interpret because of wide confidence intervals, although the rate of harm related adverse events and suicidal behaviour/ideation was higher in fluoxetine than placebo treated patients. However, there is evidence that fluoxetine is less likely than placebo to lead to discontinuation of treatment for any reason. Treatment emergent adverse events were generally similar between fluoxetine and placebo with the exception of hyperkinesias, headache and skin rash, where there is evidence suggesting increased risk for fluoxetine. Paroxetine (SSRI) In one study, paroxetine (up to 40 mg/day for 8 to 12 weeks) improved the likelihood of remission in 12 18 year olds. However, further evidence suggested paroxetine had little impact on response to treatment, symptom levels, functional status, or clinical improvement. There is evidence suggesting that paroxetine is more likely than placebo to bring about serious adverse effects, and limited evidence of increased risk of suicidal behaviour/ideation and early discontinuation from treatment because of adverse events or any reason. Paroxetine is more likely than placebo to cause the following treatment emergent adverse events: dizziness, hostility, insomnia, somnolence and tremor. Sertraline (SSRI) Sertraline (up to 200 mg/day for 10 weeks) when compared with placebo produced a small improvement in depressive symptoms in 6 17 year olds. However, the evidence regarding remission, response to treatment, and clinical improvement is inconclusive. Evidence suggests no impact on functional status. There is evidence suggesting that children (6 11 years) treated with sertraline are more likely to discontinue treatment because of adverse events, and for children/young people there is limited evidence of increased risk of suicidal behaviour/ ideation. Evidence is inconclusive regarding serious adverse events. There is limited evidence for an increased risk of discontinuation of treatment for any reason. In children (6 11 years), sertraline is more likely than placebo to cause the following treatment emergent adverse events: nausea, diarrhoea, and anorexia; and may increase the risk of vomiting, agitation, urinary incontinence, and purpura. In young people (12 17 years), sertraline is more likely than placebo to cause vomiting and diarrhoea. Citalopram (SSRI) 12

There was limited evidence that citalopram (up to 40 mg/day for 8 to 12 weeks), when compared with placebo, improved the chance of remission and response to treatment, and improved depressive symptoms in 7 18 year olds. There was limited evidence that citalopram increases the risk of treatment emergent adverse events, suicidal behaviour/ideation, early discontinuation because of suicide attempts, and early discontinuation because of adverse events. Citalopram is more likely than placebo to cause the following treatment emergent adverse events: rhinitis, nausea, flu like symptoms, fatigue, diarrhoea, and pharyngitis. NICE (2005) conclusion: Fluoxetine is the only SSRI/atypical antidepressant where there is evidence of clinical effectiveness across a range of outcome measures. The evidence suggests that tricyclic antidepressants should not be used. There is limited evidence that all SSRIs/atypical antidepressants (including fluoxetine) may increase the risk of suicidal ideation and/or behaviour and increase the risk of discontinuation of treatment because of adverse events. In the WHO Model List of Essential Medicines for Children (WHO 2007) fluoxetine is included for the pharmacological treatment of adolescents (>8 years) References Dubicka B, Hadley S, Robertrs C (2006). Suicidal behaviour in youths with depression treated with new generation antidepressants. Meta analysis. British Journal of Psychiatry, 189:393 8. Hammad TA, Laughren T, Racoosin J 2006). Suicidality in paediatric patients treated with antidepressant drugs. Archives of General Psychiatry, 63:332 9. Hazell P (2009). Depression in children and adolescents. BMJ Clinical Evidence, 01:1008. Hazell P et al (2002). Tricyclic drugs for depression in children and adolescents. Cochrane Database of Systematic Reviews, (2):CD002317. Hetrick SE et al (2007). Selective serotonin reuptake inhibitors (SSRIs) for depressive disorders in children and adolescents. Cochrane Database of Systematic Reviews, (3):CD004851. NICE (2005). Depression in Children and Young People. Identification and management in primary, community and secondary care. National Clinical Practice Guideline Number 28. The British Psychological Society & The Royal College of Psychiatrists. 13

Tsapakis EM et al (2008). Efficacy of antidepressants in juvenile depression: meta analysis. British Journal of Psychiatry, 193:10 17. Usala T et al (2008). Randomized controlled trials of selective serotonin reuptake inhibitors in treating depression in children and adolescents: a systematic review and meta analysis. European Neuropsychopharmacology, 18:62 73. Whittington CJ et al (2004). Selective serotonin reuptake inhibitors in childhood depression: systematic review of published versus unpublished data. Lancet, 363:1341 5. WHO (2007). WHO Model List of Essential Medicines for Children. Geneva, World Health Organization. From evidence to recommendations Factor Explanation Narrative summary of the evidence base In terms of proportion of children showing an improvement in depressive symptoms, the evidence for tricyclic antidepressants (TCAs) versus placebo is inconclusive and so it is not possible to determine if there is a clinically important difference (RR 0.86, 0.25 to 1.89). Similarly, in terms of score on a depression measure the evidence for tricyclic antidepressants (TCAs) versus placebo is inconclusive (SMD 0.15, 0.34 to 0.64). For most SSRIs, the evidence is sparse and so it is not possible to determine if there is a clinically important difference between individual SSRIs and placebo. For fluoxetine, in terms of responders and in terms of score on a depression measure, there is limited evidence suggesting a significant beneficial effect. In terms of treatment acceptability and adverse effects, the evidence is sparse and inconclusive. In terms of suicide ideas/behaviour, for the group of selective serotonin reuptake inhibitors, including children and adolescents together, there is 14