A Crash Course in School Refusal

School refusal behavior is one of the most stressful things that can happen to a family. It is woefully understudied. Only recently has the term begun to be more commonly acknowledged as providers give more attention to this issue. This topic is also one I am quite personally interested in treating as comprehensive treatment often involves individual, family, and other environmental intervention. It is an extremely challenging issue to treat, so I figured I would review a recent summary article on school refusal behavior and anxiety to provide an overview.

Maynard, B. R., Brendel, K. E., Bulanda, J. J., Heyne, D., Thompson, A., & Pigott, T. D. (2015). Psychosocial interventions for school refusal with primary and secondary students: A systematic review. Campbell Systematic Reviews, 12.

First, what on earth is school refusal behavior? 

This is not actually just defined as kids who don’t go to school for several days. For example, kids who are ill for a week aren’t really refusing school! School refusal behavior typically entails difficulty attending school, and often long absences from classes and/or entire school days. There seems to be no consensus of what amount of missed school constitutes school refusal behavior. However, sometimes studies define success by achieving a 90% attendance rate. In addition, there is some controversy about whether truancy should be considered school refusal behavior. The authors in this study believe that truancy is different because it doesn’t involve distress, often occurs without parental knowledge, and is highly associated with conduct. I disagree, as I believe others in the literature do (e.g. Kearney, 2008). While truancy behavior is most certainly different, it is simply a different function of school refusal behavior – That is, it occurs for a different reason. That’s all, it’s just a different reason for the child who refuses school.

School refusal behavior is a huge problem. It is related to a whole host of problems such as suicide attempts, drug use, antisocial behavior, and teenage pregnancy. While these are long-term consequences, the short-term consequences are equally concerning. Obviously, school performance can be severely impacted, but think about what else happens that is also terribly detrimental: They start to lose contact with their friends. This social and academic difficulty can be the start of many of the long-term problems. Oh, and as an aside, it is also EXTREMELY stressful for families.

School refusal behavior occurs for around 1-2% of kids. For children receiving psychological treatment this number is typically higher (upwards of 5%), which makes sense given that it’s associated with various mental health disorders. It’s as common for males as females and is seen more with adolescents. In addition, the possible reasons seem to vary across ages. For example, younger children tend to experience more separation anxiety, whereas older children may be more likely to experience social or general anxiety. However, about 1/3rd of children do not meet criteria for any diagnosis whatsoever.

According to one prominent researcher in this field (Kearney), there are 4 primary reasons why a child engages in school refusal behavior. These are called the 4 functions of the behavior and include:

  1. Negative affect – Avoiding school is associated with relief from school-related negative feelings
  2. Escape evaluation – Avoiding school is associated with relief from anxiety in social/performance settings
  3. Reinforcement – Avoiding school is associated with positive things (e.g. playing video games)
  4. Attention – Avoiding school is associated with increased attention, such as from parents

According to Kearney, understanding the function of school refusal behavior for any one child will help guide treatment. For example, knowing that the main reason for refusal is due to escape evaluation suggests that the treatment should focus on helping the child to face anxious situations. Knowing that refusal behavior increases attention from parents should suggest that parent training is part of treatment in order to shift the balance so that the child gets much more attention for attending school, and that refusal behavior is ignored.

And therein lies part of the difficulty of treating school refusal behavior: The children are a “heterogeneous group”. This just means they are all very different. Although each child’s school refusal behavior problem may be very different, here is a little of what we know:

  • Anxiety is related in around 50% of cases. The exact disorder is often developmentally related, with separation anxiety more common in younger children. Very often, kids complain about somatic symptoms (e.g. stomach hurts), which is often used as a reason to not attend school. This is why some treatment focuses on helping to treat and manage anxiety.
  • Oppositional behavior is related in around 30% of cases. This is why some treatment focuses on helping parents to manage these behaviors.
  • Depression definitely occurs, but it is not as often a cause of school refusal.

The Review Study

Because each individual is so different, having one intervention to treat everyone is difficult. However, there are typically two outcomes that we clinically want to see when we treat school refusal: 1) That school attendance increases, and 2) That any psychological problem contributing to this problem (e.g. anxiety, depression, oppositional behavior) is managed. This article helps the literature because it addresses whether these things actually happen. Specifically, the purpose of the review article is to understand whether:

  • School refusal interventions affect anxiety
  • School refusal interventions affect school attendance.

The article is a meta-analysis, which I explain a little in this article. To do this they have to first review all the research in the area. I won’t go into the methods too much, but the good news is they only reviewed research that included a control condition-this is a sign of higher quality research. One really interesting thing also stood out to me. In their review of articles they decided to include both published and unpublished data. This is both a strength and a weakness, and it’s important you understand why.

Why is it a strength?

The review covers more information! Just because something isn’t published doesn’t mean it isn’t good research. Many dissertations that take years of work go unpublished, for example, because the authors don’t pursue academic fields. In addition, due to the publication bias, studies that don’t find anything are often not published, yet the fact they don’t find anything can be just as important.

Why is it a weakness?

Unpublished research may be unpublished because it is low quality. Published research is mostly peer-reviewed meaning other researchers think that the study is of high enough quality. If a study is unpublished, for example because it had poor research design, then including it in the review could lead to false conclusions.

In their review they found eight studies that were done between 1980-2013 that assessed effects of interventions upon attendance and anxiety for youth with school refusal behavior. This equates to 435 youths studied,  which is not a huge number. Two of the eight studies were called “quasi-experimental” which means that they didn’t include a control group. Most of the studies assessed CBT, some included parent and teacher interventions. Most of the interventions were 4-12 sessions total, which in real world practice is not very much. Follow-up assessments varied across studies, with the longest being 1 year.

What Did They Find?

Anxiety Outcomes

Four studies assessed this, three of which compared interventions to “alternate interventions”. They found that the effect size was .06, which basically means the interventions didn’t affect anxiety. But let’s break this down. First, the only study that found that CBT treatment was worse than an alternative treatment was an unpublished dissertation, and this dragged down the effect size. In this study, the “CBT” study was really cognitive therapy plus problem solving – two things we know are not the first line in treating anxiety for kids. It’s not a surprise their intervention didn’t work. Two of the studies essentially found that CBT treatment was equal to an alternate treatment. The study that showed treatment to be effective was the only one which used a waitlist control rather than an alternative treatment control, which makes sense why the effects would be larger in this one study.

Beyond this, two studies assessed CBT plus medications versus CBT alone. The effect size was -.05, again showing no real effect.

As such, it seems that CBT is not great at reducing anxiety for children exhibiting school refusal behavior. But hold your horses…We will discuss this more in a moment.

Attendance Outcomes

Six studies with control groups, and one study without a control group assessed this. They found the effect size was .54, which is medium (and pretty good).

Two studies assessed CBT plus medication versus CBT alone. They found an effect size of .61. One of the studies clearly favored CBT plus medication as the best treatment, whereas the other study still favored the combination, but a lot less.

As such, it seems that CBT likely improves attendance outcomes, and that CBT plus medication may be even better, but is probably just as good.

The Authors’ Conclusions

First, they rightly conclude that there is nowhere near enough quality research on school refusal behavior. This is a big problem, because it means our interventions are not super well guided.

They also note that a “problem” of the research is that many studies compare CBT to another form of treatment (though actually this is a good thing). The reason this is a problem is because the differences between CBT vs other treatments are likely small, whereas for CBT vs no treatment the differences are large. This is the same as me comparing a running program to either a weight lifting program or sitting on the sofa all day for fitness – the running program will look much better when compared to sitting on the sofa! Although this does limit our findings, it is generally a good thing because it means the interventions are being better tested.

The burning question from this study is why anxiety and attendance don’t both seem to improve….Only attendance improves. The authors’ theory is that the improved attendance resulting from treatment actually increases anxiety, and that continued attendance should result in a future decrease. I don’t think they discuss this in enough depth, and they present no evidence for this claim, but it seems reasonable. It seems like the authors challenge the traditional proposed mechanism of action:

Proposed Mechanism
Traditional Proposed Mechanism of Treatment

With this one….

Screen Shot 2017-10-22 at 4.11.51 PM
Authors’ New Theoretical Mechanism of Treatment

I think they make an error in assuming anxiety and attendance are related, especially since we know that plenty of kids with school refusal behavior do not meet criteria for anxiety disorders. However, let’s just assume we are exclusively talking about youth who have difficulty attending school due to anxiety. In addition to their conclusions, there are other reasons why CBT could improve school attendance without changing anxiety. I have four thoughts:

  1. Many kids with anxiety don’t refuse to attend school to avoid anxiety (i.e. the function of school refusal is not all about anxiety)- There are many other reasons for avoiding school.
  2. Some forms of anxiety just have much worse outcomes (e.g. generalized anxiety disorder). It could be that the types of anxiety associated with school refusal behavior are more resistant to intervention.
  3. The CBT treatment given was not great (e.g. heavy cognitive focus, not enough exposure therapy), or simply an insufficient dose for change to occur. This is similar to how an antibiotic drug may work, but only taking one of them won’t have much effect.
    • The CBT applied in some of these studies is not highly appropriate for youths with an average age of 12 years. Far too much focus is spent on cognitive restructuring, problem solving, and relaxation training – when we have an overwhelming anxiety literature showing that exposure therapy for anxiety is the key ingredient. In addition, treatment length often seems too brief considering how complex this issue is.
    • The authors also noted that CBT interventions vary a lot, some including parent training, some including longer durations, more sessions etc.
  4. With school refusal behavior, anxiety often builds around missing work that accrues, in addition to whatever was causing anxiety. The can take a while to remove, even if the kid goes back to school, meaning that anxiety could remain high for a while.

In addition to these thoughts, you might be asking one major question: If CBT doesn’t reduce anxiety, and the authors propose that anxiety is reduced through school attendance, shouldn’t the treatment focus be more about parent behavior training and contingency management? This would facilitate faster return to school, and then allow anxiety treatment to progress through natural exposure (being in school) and continued exposure to fear within treatment. Because this is getting confusing, I have created a diagram on potential mechanisms of treatment:

Screen Shot 2017-10-22 at 4.52.41 PM
Potential Mechanisms of Treatment

That’s a lot of possibilities!!! What do I think? Based on their research and conclusions, other literature, and clinical work, the most well-suited treatment involves several components:

  • A good functional understanding of the reasons for school refusal behavior
  • Parent behavior training and contingency management
  • Individual CBT focused on exposure therapy for anxiety
  • Getting the child back into school before waiting for anxiety to decrease


It’s very clear that school refusal behavior is difficult to treat! It’s extremely stressful for everyone involved. This study really adds to the literature for kids who demonstrate school refusal behavior while also displaying anxiety. The main finding is that we are not bad at improving school attendance outcomes, at least in the short-term (remember, most studies did not have a long follow-up), but we don’t seem to be able to reduce anxiety. This is over a 4-12 week period. There are many reasons I have outlined for this, all of which are totally theory. I think the starting point is to truly understand the function of school refusal behavior. This will help understand whether anxiety and school refusal behavior are related. Once assessment is complete, focusing on getting back to school is critical, and effective. Both individual and parent interventions seem to achieve this, though in my experience it is parent involvement and contingency management that are particularly helpful. While anxiety may not decrease through this process, it seems far too soon to write off CBT interventions, especially since the literature supporting the role of exposure therapy is so effective in treating anxiety. It is more likely that simply further research is needed. Specifically, we need longer studies, ones that target anxiety with a heavier exposure emphasis, and ones that involve more consistent treatment protocols.

I think the overarching message is one of hope. If you or someone you know has a child demonstrating school refusal behavior, psychological intervention can help even if the protocols are not yet perfected. Take action quickly before the behavior becomes established. I’ll follow up soon with an article on what to look for to know that you are getting evidence-based treatment!

Thank you for reading, let me know what you think below 🙂

Follow me on Facebook to receive updates on future articles.

Leave a Reply