What is the most effective treatment to offer to children and young people who have depression? Which psychotherapies will children and adolescents engage with? These are the questions that a new systematic review and network meta-analysis seeks to answer (Zhou et al, 2015).

These questions are important and the answers really matter. As Zhou and colleagues highlight, depression is all too common during adolescence. The immediate impact of depression is debilitating. Being depressed interferes with normal development and activities and, even more concerning, having had an episode of depression as a teenager puts you at risk of depression later in life, and increases the risk of many other mental and physical disorders and adverse life circumstances.

Existing reviews and meta-analyses suggest that available treatments for adolescent depression are only modestly effective (Weisz et al, 2013), something of grave concern given the severity of depression. The bottom line of this review is the same.

Methods

Nine different treatments, from 52 different studies were included:

1. Cognitive behaviour therapy (CBT)
2. Interpersonal therapy (IPT)
3. Supportive therapy
4. Cognitive therapy
5. Family therapy
6. Play therapy
7. Behavioural therapy
8. Problem solving therapy
9. Psychodynamic therapy

These were compared to a range of control conditions including psychological placebo, treatment as usual, and waiting list.

The efficacy of each treatment was assessed in a conventional way, by comparing depression symptoms at the beginning and end of treatment, (and where possible also at follow up).

Results

Efficacy of treatment

• Only two forms of psychological treatments (IPT and CBT) were significantly better than the control conditions.

Acceptability of treatment

Unusually, this review also examined the ‘acceptability’ of each psychological therapy. Acceptability was measured by the proportion of participants who discontinued treatment.

• Patients receiving IPT and problem solving therapy were significantly less likely to discontinue treatment than patients who were receiving CBT and cognitive therapy
• The authors speculate that treatments that focus on cognitions may be inherently less easily accessible to children and young people and this explanation is very plausible
• In fact, one concerning feature of the 9 therapies identified in this review is that the majority were not developed with children and adolescents in mind, they were adapted from treatments for adults
• Only family therapy and play therapy can reasonably be said to have emerged from a primary interest in depression as a developmental problem.

Strengths and limitations

But is discontinuation of treatment a valid measure of ‘acceptability’? I am not convinced. The basis for this index is presumably the assumption that children and young people will ‘vote with their feet’ and if they do not find a specific therapy acceptable that they will stop coming. This is certainly a crude measure of acceptability, but can we trust it?

Discontinuing treatment might happen for many reasons; for example patients may improve before the end of therapy and therefore decide not to attend any remaining sessions. This would indicate a good outcome and would certainly not indicate that a treatment was unacceptable. In addition, discontinuing treatment may be associated with specific patient characteristics including the severity of depression, co-morbid problems, practical difficulties (e.g. travelling to sessions) and demographic variables. Thus discontinuation can only be a good index of acceptability if these other characteristics are independent of type of treatment; this is possible but untested.

This is a careful review with many strengths. An impressive feature is the extensive and thorough search for relevant literature. No restrictions were put on the language in which the studies were published and studies that included children and/or adolescents were included. Studies were assessed for potential bias and ratings were assessed for reliability.

Network meta-analysis

An additional novel feature of this review is the use of a new technique ‘network meta-analysis’. The usual form of meta-analysis combines data from different studies that compare the same two (or three) interventions (e.g. all studies that compare CBT to other therapies can be combined, or all studies that compare family therapy to other therapies can be combined).

‘Network meta-analysis’ allows further comparison of interventions that have not been directly compared. Thus the amount of data available for analysis is significantly increased and the effectiveness of all forms of psychotherapy that have been studied can be compared. A potential advantage of this method is that it can be more comprehensive and, because it can include many more studies, have greater power.

Unfortunately even the most advanced analytic technique cannot compensate for the very small number of studies focusing on most of the 9 forms of psychological therapy. Compared with the evaluation of CBT on over 1,000 children and adolescents, IPT which was the next largest category included only 344 participants; in the smallest categories, 44 children or young people received problem solving therapy and 35 received psychodynamic therapy.

Conclusions

It is a safe bet that the findings relating to efficacy are valid and the key headline of this review is not surprising:

• Most psychological therapies for child and adolescent depression are not better than control conditions
• Even those that are superior to control are not very impressive

This study highlights the absence of developmentally sensitive, theoretically based interventions that target key processes in child and adolescent depression. We have a lot to do.

Links

Primary paper

Zhou X, Hetrick SE, Cuijpers P, Qin B, Barth J, Whittington CJ, Cohen D, Del Giovane C, Liu Y, Michael KD, Zhang Y, Weisz JR, Xie P. (2015) Comparative efficacy and acceptability of psychotherapies for depression in children and adolescents: A systematic review and network meta-analysis. World Psychiatry, 14: 207–222. doi: 10.1002/wps.20217

Other references

Weisz JR, Kuppens S, Eckshtain D, et al. (2013) Performance of evidence-based youth psychotherapies compared with usual clinical care: a multilevel meta-analysis. JAMA Psychiatry. 2013;70:750–61.

Martin Good / Shutterstock.com