The National Institute for Health and Clinical Excellence (NICE) published a guideline on depression in children and young people in late 2005 (National Institute for Health and Clinical Excellence, 2005). It is one of only two guidelines specifically addressing child and adolescent mental health problems, thus recognising the specific developmental and treatment needs of young people. It is clear, concise and comprehensive, and there is much to commend in it. Nevertheless we would like to draw attention to flaws, which include some poor matches between the existing evidence and important treatment recommendations.

Poor match between evidence and recommendations

First, the guideline conveys some confusing uncertainty about the diagnosis of depression in adolescents. At the same time it recommends that the accuracy of diagnosing depression in child and adolescent mental health services should be improved by the use and adaptation of psychiatric research interviews. Although we fully concur with the advice that specialist child and adolescent mental health staff should be trained in the recognition of depression in adolescents, and interviews using operationally defined criteria may in the future be demonstrated to have a useful role, this should not substitute for adequate clinical training in the psychiatric diagnosis, differential diagnosis and clinical formulation of adolescent depression.

Second, the guideline emphasises screening in primary care for depression in children or young people exposed to single recent life events such as bereavement, divorce or severely disappointing events, with a view to ‘ providing support and the opportunity to talk over the event’. There is, however, little evidence that this is beneficial and one might argue that in the absence of evidence of benefit, diverting scant resources in this way might cause harm. What has been published in the UK cautions against this approach, as most bereaved young people do not wish to speak to professionals but would rather speak to family and friends (Reference Harrison and HarringtonHarrison & Harrington, 2001). Young people with depression and other mental health problems do not usually present to primary healthcare services with stress reactions to distressing events (Reference Kramer and GarraldaKramer & Garralda, 1998). Of more direct relevance and proven benefit would be focused interventions by primary care staff in which the identification of depression is combined with a brief intervention tailored to the primary care setting and based on principles of cognitive—behavioural therapy (CBT; Reference Gledhill, Kramer and GarraldaGledhill et al, 2003). A randomised controlled trial (RCT) has found benefit from multi-modal treatment interventions in which usual care in the primary care setting is enhanced by support from mental health workers and treatment choice, including manualised CBT and treatment with selective serotonin reuptake inhibitors (SSRIs; Reference Asarnow, Jaycox and LabordeAsarnow et al, 2005).

Third, recommendations about the management of moderate-to-severe depression are most concerning. The guideline states that the first-line treatment should be a specific psychological therapy (individual CBT, interpersonal therapy or shorter-term family therapy) for at least 3 months. It is further stated that if there is no response another psychological therapy should be tried, and the possibility of using drug therapy (fluoxetine) considered. This is, however, not strictly in line with existing evidence. Most RCTs of psychological therapies in adolescents with depression that have shown beneficial effects have involved young people with mild-to-moderate (as opposed to moderate-to-severe) depression and used CBT — and to a lesser extent interpersonal therapy (National Institute for Health and Clinical Excellence, 2005). The results cannot therefore automatically be extrapolated to the more severe cases of depression seen in secondary care.

Moreover, in recommending other psychological treatments if CBT is not efficacious in mild-to-moderate depression, the guidelines do not appear to make it sufficiently clear that the existing evidence favours CBT over other therapies such as family or supportive work (Reference Brent, Holder and KolkoBrent et al, 1997). Family therapy may be recommended for the relationships problems or poor social functioning that act as maintaining factors for the disorder in some cases (Reference Kolko, Brent and BaugherKolko et al, 2000) rather than for depression itself, as implied in the guideline.

There is little evidence for efficacy of individual psychodynamic psychotherapy, and none for the claim that problems persisting after 3 months call for 30 sessions of individual psychotherapy. This assertion was based on only one study comparing psychodynamic psychotherapy and family therapy, which in fact failed to document differences between these two treatment arms (Reference Trowell, Joffe and CampbellTrowell et al, 2007).

The guideline recommendations do not sufficiently acknowledge that the best evidence for efficacy in moderate-to-severe depression in adolescents is for the use of a SSRI (fluoxetine). The Treatment for Adolescents with Depression Study (TADS), an RCT of treatments for moderate-to-severe depression in adolescents, found that fluoxetine was significantly more effective than placebo, CBT was similar to placebo, and CBT in combination with fluoxetine resulted in marginal gain (Reference March, Silva and PetryckiMarch et al, 2004; Reference Apter, Kronenberg and BrentApter et al, 2005). Furthermore, those receiving fluoxetine (with or without CBT) achieved more rapid response than those receiving placebo, and the combination of fluoxetine and CBT was associated with a more rapid response than CBT alone (Reference Kratochvil, Emslie and SilvaKratochvil et al,2006). In spite of the lengthy discussion regarding the dangers of drug treatments in the guidelines, the TADS Team (Reference March, Silva and PetryckiMarch et al, 2004) found that fluoxetine was associated with significant reduction of suicidal ideation and that there were too few suicidal attempts for statistical analysis.

Despite the critique of the study, which was met with vigorous rebuttal by the lead investigators (Reference Jureidini, Tonkin and MansfieldJureidini et al, 2004; Reference MarchMarch, 2005), the findings were not a surprise. Earlier RCTs had demonstrated the efficacy of fluoxetine over placebo (Reference Emslie, Heiligenstein and WagnerEmslie et al,2002) and it had previously been shown that CBT was associated with a lesser response for young people with more severe depression (Reference Jayson, Wood and KrollJayson et al, 1998; Reference Brent, Kolko and BirmaherBrent et al, 2002). Evidence from these studies goes directly against the guideline recommendation that 3 months should elapse before the drug is started, during which time 4–6 sessions of psychological treatment should be offered.

Conclusions

We are not advocating the use of antidepressant medication indiscriminately for all young people with moderate-to-severe depression because there might be a slightly increased risk of suicidal symptoms (Reference Dubicka, Hadley and RobertsDubicka et al, 2006; Reference Hammad and LaughrenHammad et al, 2006). A recent meta-analysis concluded that the clinical benefits of antidepressants appear to be considerably greater than the risks from suicide ideation or suicide attempts (Reference Bridge, Iyengar and SalaryBridge et al, 2007), but it still seems prudent for SSRIs to be introduced by specialists and to be monitored carefully. We are simply pointing out that adherence to the recommendations of the guideline could inhibit the appropriate use of efficacious treatment in specialist practice.

The guideline highlights the importance of identifying and treating depression in young people and covers many important aspects helpfully. Nevertheless the limitations call for caution in the way the recommendations are implemented.

Declaration of interest

None.