Refusing to attend school is a common problem amongst children and adolescents and can cause significant disruption for both the young person, including problems with social adjustment and academic achievement, and their family.

This recent systematic review provides a clearer picture as to whether psychosocial interventions (most commonly, CBT) help young people and their families overcome problems with school refusal (Maynard et al, 2015).   What it tells us is that CBT does produce a significant positive effect on school attendance but not on a child’s anxiety levels.

What is school refusal?

School refusal refers to a young person’s difficulty in attending school. It is often linked to severe emotional distress, most commonly anxiety, but also depression. 50% of school refusers meet the criteria for an anxiety disorder (Walter et al, 2010).

School refusal is understood differently to truancy; truancy is associated with antisocial behaviour rather than emotional distress and truants tend to be absent from school without their parents’ knowledge.

1-2% of young people in the general population experience school refusal, it is equally prevalent in boys and girls, but it is more common in secondary school compared to primary school children (Heyne and King, 2004; Hersov, 1985).

What is CBT and how might it help?

Cognitive behavioural therapy (CBT) is a psychological therapy that has been used for the past 25-30 years to treat a range of childhood difficulties including anxiety and depression.

A key component of CBT is graded exposure; this involves gradually encouraging the young person to face their fears by entering into the situation/s that they find anxiety-provoking.

Another important aspect of CBT is the identification of anxious/negative cognitions and the implementation of a number of strategies to shift or weaken these cognitions or thoughts, which include Socratic questioning and the use of behavioural experiments.

Further strategies that are included under the umbrella of CBT are problem solving, relaxation and contingency management (e.g. use of positive and negative reinforcement).

CBT has a strong evidence base for the treatment of childhood anxiety disorders (e.g. Cartwright-Hatton et al, 2004) and there is some evidence for its effectiveness to treat depression (e.g. March et al, 2008). However, there have been relatively few studies exploring the efficacy of CBT for school refusal; previous reviews have been limited to published data and have not examined the quality of the data provided.

This new systematic review (Maynard et al, 2015) includes the first meta-analysis of psychosocial interventions for school refusal.

Methods

The questions that the authors wanted to address in this review were:

1. Do psychosocial interventions targeting school refusal reduce anxiety?
2. Do psychosocial interventions targeting school refusal increase attendance?

Maynard et al (2015) conducted a comprehensive literature review and included published and unpublished studies from 1980 to 2013. Inclusion criteria included the use of statistical controls and baseline data on outcomes.

Eight studies were included in the meta-analysis, highlighting the paucity of research into school refusal interventions:

• 6 studies were randomised controlled trials
• 6 studies examined the efficacy of psychosocial interventions alone
• 2 studies explored the impact of a psychosocial intervention with medication
• 7 of the 8 studies examined the efficacy of CBT or a variant of CBT

Most studies were conducted in a clinic setting whilst the others were conducted in school or at home. 435 participants were included across the 8 studies, whilst the average age of participants was 11.9 years.

Results

1. Overall, psychosocial interventions did not produce a significant effect on the young person’s anxiety levels
2. Overall, psychosocial interventions produced a significant positive effect on school attendance
3. Similar results were found for studies that used a combination of CBT and medication

Discussion

Although the interventions did result in improved school attendance, it is puzzling that young people’s anxiety levels did not reduce. This is particularly confusing given that anxiety is seen as a significant factor in the development and maintenance of school refusal and that we know that there is a good evidence base for the treatment of anxiety difficulties using CBT.

The authors propose that the increased exposure to school (a key component of CBT) will inevitably result in increased attendance but may actually also lead to increased anxiety in the short-term, as the young person is facing their fears but their anxiety has not yet reduced following habituation to the anxiety provoking situation. This may well be true as young people who are not attending school (due to school refusal problems) often report only limited anxiety, as they are not actually being exposed to their fears. The authors rightly say that in order to see a drop in anxiety levels, longer term follow-up data is required.

Limitations

It is also important to consider any limitations of the studies reviewed and the implications of these for the findings.

• Firstly, almost all studies (except two) compared two interventions and the authors found that there was improvement on at least one of the two outcome measures across both groups. There is thus no firm evidence currently that CBT is better than another intervention. However, we must not lose sight of the fact that CBT does seem to have a positive effect on school attendance.
• A further limitation related to the varying forms of CBT delivered. Although there were similarities, there were also significant differences, in terms of length of treatment (4-12 sessions), and who was involved in treatment; some studies included only individual child interventions, others also included parent and teacher training. More manualised CBT treatment trials should be conducted to address this issue.
• Finally the authors point out the presence of performance and detection bias issues, and non-random allocation of participants in some trials, highlighting the need for better controlled studies and large sample sizes.

So should we use CBT to treat school refusal?

This review provides tentative evidence for the use of CBT for school refusal. However, we cannot claim that CBT works better than other interventions at this stage; more studies examining alternative interventions and/or comparing them to CBT are needed before we can recommend one type of intervention over another.

We do not yet know if CBT interventions for school refusal reduce anxiety levels but there is some evidence that they improve school attendance.

Given the absence of any evidence for alternative treatments, it would seem sensible to use CBT for school refusal until further evidence comes to light either to strengthen the support for CBT or provide robust evidence for something else.

Links

Primary paper

Maynard BR. et al Psychosocial Interventions for School Refusal with Primary and Secondary School Students: A Systematic Review. Campbell Systematic Reviews, 2015 May 11 (12).

Other references

Cartwright-Hatton S. et al (2004). Systematic review of cognitive behaviour therapies for childhood and adolescent anxiety disorders. British Journal of Clinical Psychology, 43, 421-436. [DARE summary]

Hersov L. (1985).  School refusal In M. Rutter and L. Hersov (Eds.), Child and adolescent psychiatry: Modern approaches (2^nd ed.) (pp.382-399). Oxford: Blackwell.

Heyne, D., and King, N.J. (2004).  Treatment of school refusal.  In P.M. Barrett and T.H. Ollendick (Eds.), Handbook of interventions that work with children and adolescents: Prevention and treatment (pp. 243-272). Chichester: John Wiley and Sons

March  JS. et al (2008). The treatment for adolescents with depression study (TADS): long-term effectiveness and safety outcomes. Archives of General Psychiatry 64(10), 1132-43.

Walter D. et al (2010). Short term effects of inpatient cognitive behavioral treatment of adolescents with anxious-depressed school absenteeism: An observational study. European Child and Adolescent Psychiatry, 19, 835-844 [PubMed abstract]