Skip to main content
Free AccessReview Article

Developmental Issues Associated with Adolescent School Refusal and Cognitive-Behavioral Therapy Manuals

A Practitioner Review

Published Online:https://doi.org/10.1024/1422-4917/a000881

Abstract

Abstract. It is important to consider developmental issues when addressing school refusal (SR) in adolescence. Referral rates appear to be higher among adolescents relative to children, and treatment appears to be less effective among adolescents. This paper provides an in-depth review of developmental issues associated with adolescent SR and treatment via cognitive-behavioral therapy (CBT). It begins by considering the reasons for the higher referral and poorer treatment outcomes, including the higher level of absenteeism in adolescence, higher rates of concurrent social anxiety disorder and depressive disorder, and the developmental challenges inherent to adolescence. Such challenges include increased academic and social demands in the secondary-school environment, and increasing autonomy which may contribute to family conflict. These developmental issues may potentiate and exacerbate an adolescent’s difficulty attending school, make it difficult for families to cope, and complicate practitioners’ efforts to provide effective treatment for SR. Further, the review describes CBT manuals for SR and the extent to which they are developmentally sensitive. There are five CBT manuals, which vary in their sensitivity to developmental issues. Various multimodal treatments employ interventions in addition to CBT, such as medication or inpatient treatment, to address the complexity of SR in adolescence. Nevertheless, nonresponse to treatment for adolescent SR ranges from one-third to two-thirds of youths. Attention thus needs to be given to ways of improving treatment outcomes.

Entwicklungsfragen im Zusammenhang mit jugendlicher Schulverweigerung und kognitiv-behaviorale Therapieleitfäden

Zusammenfassung. Das Jugendalter ist eine wichtige Entwicklungsphase bezüglich Schulverweigerung (SV). Die Inzidenz scheint höher zu sein als bei jüngeren Kindern, und die Behandlung scheint bei Jugendlichen weniger wirksam zu sein. Dieser Artikel gibt einen ausführlichen Überblick über die Entwicklungsprobleme im Zusammenhang mit SV bei Jugendlichen sowie die Behandlung durch kognitive Verhaltenstherapie (KVT). Er beginnt mit einer Betrachtung der Gründe für die höhere Inzidenz und die schlechteren Behandlungsergebnisse, einschließlich der höheren Fehlzeiten in der Adoleszenz, der höheren Raten sozialer Angststörungen und depressiver Störungen sowie der mit der Adoleszenz verbundenen Entwicklungsprobleme. Zu diesen Herausforderungen gehören die erhöhten akademischen und sozialen Anforderungen in der Sekundarstufe und die zunehmende Autonomie, die zu familiären Konflikten beitragen kann. Diese Entwicklungsprobleme können die Abneigung eines Jugendlichen gegen die Schule verstärken und verschlimmern, was es den Familien erschwert, damit umzugehen. Außerdem werden die Bemühungen der Therapeuten erschwert, eine wirksame Behandlung für SV anzubieten. Darüber hinaus beschreibt der Bericht KVT-Manuale für SV und das Ausmaß von deren Verständnis für entwicklungsbedingte Probleme. Es gibt fünf KVT-Manuale, die sich in ihrer Sensibilität für Entwicklungsthemen unterscheiden. Verschiedene multimodale Behandlungen setzen zusätzlich zur KVT Interventionen ein, wie z. B. Medikamente oder stationäre Behandlung, um die Komplexität von SV im Jugendalter zu bewältigen. Dennoch sprechen ein Drittel bis zwei Drittel der Jugendlichen nicht auf eine Behandlung für jugendliche SV an. Es muss daher nach Wegen gesucht werden, die Behandlungsergebnisse zu verbessern.

Introduction

School Refusal

School refusal (SR) is the type of school attendance problem characterized by the emotional distress a young person1 experiences regarding going to school and/or being at school, often resulting in absenteeism. Working definitions have varied across the decades, but a recent review of definitions suggests that SR occurs when:

(1) A young person is reluctant or refuses to attend school, in conjunction with emotional distress that is temporal and indicative of aversion to attendance (e. g., excessive fearfulness, temper tantrums, unhappiness, unexplained physical symptoms) or emotional distress that is chronic and hindering attendance (e. g., depressive affect; sleep problems), usually but not necessarily manifest in absence (e. g., late arrivals; missing whole school days; missing consecutive weeks, months, or years); and (2) the young person does not try to hide associated absence from their parents (e. g., they are at home and the parents are aware of this), and if they previously hid absence then they stopped doing so once the absence was discovered; and (3) the young person does not display severe antisocial behaviour, beyond resistance to parental attempts to get them to school; and (4) the parents have made reasonable efforts, currently or at an earlier stage in the history of the problem, to secure attendance at school, and/or the parents express their intention for their child to attend school full-time. (Heyne et al., 2019, p. 22).

The second and third criteria help to differentiate SR from truancy, and the fourth criterion helps to differentiate it from school withdrawal.

Absence from school because of SR poses risks to the young person’s development, including delays in academic achievement (Aucejo & Romano, 2016; Carroll, 2010, 2020; Gershenson et al., 2017) and early school dropout (Rumberger, 1995; Schoeneberger, 2012). Socioemotional development may also be impaired (Carroll, 2011; Gottfried, 2014; Malcolm et al., 2003). Further, SR may contribute to frustration and helplessness among parents (Dannow et al., 2020) and cause family stress and conflict (Christogiorgos & Giannakopoulos, 2014; McAnanly, 1986). School attendance problems also impact teachers (Balu & Ehrlich, 2018; Wilson et al., 2008) and practitioners (Finning et al., 2018).

Focus on Adolescence

It was long assumed that age is associated with the causation and presentation of SR as well as with the response to treatment (e. g., Berg, 1970). Rubenstein and Hastings (1980) argued that it is of crucial clinical importance to divide SR youths into distinct age groups. In line with this, it was argued that the developmental level of youths displaying SR must be accounted for during assessment (Ollendick & King, 1998) and treatment (King et al., 1995). Literature on the prevalence of SR and response to treatment, described below, confirms that adolescence is an important developmental period concerning SR.

SR occurs among 1–7 % of youths in the general population and 5–16 % of youths referred to clinical settings (Heyne & Brouwer-Borghuis, 2022). Authors contend that it is more prevalent among preadolescent and adolescent youths than among youths in early or middle childhood (Hersov, 1985; Last, 1992). Age-related patterns in samples of referred youths lend support to this contention (e. g., Heyne et al., 2002; King et al., 1998; Last et al., 1987; Last & Strauss, 1990; McShane et al., 2001; Morgan, 1959; Wu et al., 2013). For example, among youths aged 6–14 years and displaying SR, referrals peaked between 11–13 years (Morgan, 1959). This was attributed to the rapid intellectual, physical, and sexual development taking place, combined with social demands. In addition, Waite and Creswell (2014) reported that difficulty attending school was more common among adolescents with an anxiety disorder relative to children with anxiety disorders.

It is unclear whether higher referral for adolescents stems from a greater likelihood of onset in adolescence or a greater propensity for adolescents to be referred than children. Egger et al.’s study (2003) of nonreferred youths aged 9–16 years and displaying SR showed a mean onset age of 10.9 years and a significantly higher prevalence among younger youths. This suggests that the higher referral rate for adolescents does not stem from a higher prevalence in adolescence; rather, it may be explained by the severity and complexity of the problem in adolescence, as presented below (Understanding Higher Referral and Poorer Treatment Response Among Adolescents).

Cognitive-behavioral therapy (CBT) has the most scientific support in the treatment of SR (Elliott & Place, 2019; Heyne et al., 2020; Maynard et al., 2018). The outcomes from evaluations of CBT manuals for SR are presented below (Cognitive-Behavioural Therapy for School Refusal). It is worth noting here that two research groups examined outcomes according to age, using samples comprising children and adolescents. Last and colleagues (1998) compared CBT with educational support therapy (EST) among 56 youths aged 6–17 years. Younger participants in CBT and EST were more likely to achieve at least 95 % attendance by posttreatment (results were not reported for the treatment conditions separately). Heyne et al. (2002) evaluated youth-focused CBT, parent/teacher-focused CBT, and a combination thereof in 61 youths aged 7–14 years. The rates of response and nonresponse (i. e., above or below 90 % attendance) were reported by age (Heyne, 1999). Most youths aged 7–11 years were classified as responders. Similar rates of response and nonresponse were observed among youths aged 12–13 years, and most youths aged 14 years were classified as nonresponders. Only 22 % of the 14-year-olds attended school at least 90 % of the time at follow-up, compared with 66 % of those aged 13 and younger. The age-related outcomes reported by these two research groups are consistent with the inferior outcomes observed among older youths in studies evaluating interventions other than CBT (Goh, 1989; Prabhuswamy et al., 2007; Rodriguez et al., 1959; Valles & Oddy, 1984).

Age-related outcomes were also reported in three studies of CBT plus other interventions for SR (i. e., multimodal treatment). Layne and colleagues (2003) found no relationship between age and posttreatment school attendance for 41 adolescents (12–18 years) who had received CBT + medication or CBT + placebo. Walter et al. (2013) reported on 147 adolescents (12–18 years) who had received inpatient treatment comprising CBT and other interventions. Age was unrelated to school attendance and mental-health problems at posttreatment and 2-month follow-up. The one study of multimodal treatment that yielded an age-related effect employed a slightly lower age for inclusion. Melvin et al. (2017) reported on 62 adolescents and preadolescent children (11–16.5 years) who had received either CBT, CBT + fluoxetine, or CBT + placebo. For the whole group combined, older age was associated with worse attendance at the 12-month follow-up.

In sum, inferior outcomes among older youths were reported in CBT studies comprising children (or preadolescents) and adolescents, and no age-related effects were reported in CBT studies focused exclusively on adolescents. A tentative conclusion is that CBT for SR is more effective for children. This conclusion, tempered by the small sample sizes (41 to 147 youths), contrasts with the lack of a significant age effect in a meta-analysis of CBT for anxiety-disordered children and adolescents (Bennett et al., 2013). It also contrasts with Kendall and Peterman’s (2015) conclusion that adolescents do not consistently show poorer outcomes than children following CBT for anxiety disorders. The fact that CBT treatments for SR appear to be more effective for children supports Berg and colleagues’ (1993) contention that SR is a problem worthy of consideration in its own right.

Aim

This practitioner review focuses on CBT for SR in adolescence. The aim is to support practitioners by providing (1) a narrative synthesis of the literature that helps to explain higher referral and poorer treatment response among adolescents and (2) a description of CBT manuals for SR, including their attention to developmental issues and their effectiveness. Literature was included in this review if SR was defined at least by the first criterion in the SR definition presented in the section on “School Refusal” above, unless otherwise stated. Using age as a proxy for adolescence, adolescence refers to youths between 12 and 18 years.

Understanding Higher Referral and Poorer Treatment Response Among Adolescents

Heyne et al. (2014) introduced four factors likely associated with poorer treatment response among adolescents, some of which may also explain higher referrals in adolescence. The factors include the severity of SR in adolescence, the complexity of its clinical presentation among adolescents, developmental challenges inherent to adolescence, and the extent of developmental sensitivity in treatment planning and delivery.

The Severity of School Refusal in Adolescence

The severity of SR may be measured by the amount of absence and by the level of emotional distress associated with school attendance.

The amount of absence among youths referred for SR is significantly higher for older versus younger youths (Hansen et al., 1998; Heyne, 1999; Prabhuswamy et al., 2007). Hansen et al. also found that age was the strongest of 15 predictors of absenteeism. One might expect the high prevalence of absence among older youths could be explained by the chronicity of SR, inasmuch as older youths have had more time to experience SR, and earlier absences because of SR may exacerbate later absenteeism. However, Hansen et al. (1998) found no relationship between the level of absenteeism and the length of SR or the occurrence of a prior episode. It seems that higher absence among older youths is explained by factors other than chronicity. According to Hansen and colleagues, higher absence among older youths may be influenced by the complexity of the adolescent developmental period, addressed in the next section.

Little attention has been paid to age-related patterns in the level of fear or anxiety among youths referred for SR treatment. The only study (Hansen et al., 1998) identified reported lower levels of fear among youths with higher absenteeism, which was explained in terms of decreased exposure to school as the feared stimulus. It is worth noting that the measure of fear was broad (the Fear Survey Schedule for Children – Revised; Ollendick, 1983) and not a measure of fear of school per se. Studies among community samples have examined levels of fear or phobia of school. Burnham et al. (2006) reported that U.S. youths aged 12–19 years were twice as likely to demonstrate a profile of school-related fears (e. g., fear of getting bad grades or looking foolish) than youths aged 7–11 years. In an Indian sample of youths aged 11–19 years, Nair et al. (2013) found a moderate positive correlation between age and the level of SR2.

SR severity appears to be associated with treatment outcome. Hella and Bernstein (2012) attributed the poor outcomes observed in the study by Bernstein and colleagues (2000) of treatment for adolescent SR to the severe symptoms observed among the adolescents. Moreover, higher absenteeism at the start of treatment for SR was associated with a poorer response to treatment (Last et al., 1998; Layne et al., 2003).

The Complexity of the Clinical Presentation in Adolescence

Early accounts suggested that SR beginning in early adolescence has a very different clinical presentation from SR beginning during primary school (Atkinson et al., 1985). Authors described older youths as more disturbed than younger youths (e. g., Coolidge et al., 1957; Waldfogel et al., 1957), whereby “one encounters chronic, deeply imbedded problems that yield slowly to treatment, and in which the prognosis is anything but bright” (Coolidge et al., 1960, p. 599). It was also argued that older youths are more likely to display an insidious onset, whereas acute onset is more typical of younger youths (Atkinson et al., 1985; Hersov, 1985).

Studies attest to the complexity of SR in adolescence. Diagnostic comorbidity is common (McShane et al., 2001), depressive symptoms are more likely (Baker & Wills, 1978), and adolescents displaying school refusal behavior3 are more likely than children to meet the diagnostic criteria for depression (Kearney, 1993). In fact, between one-quarter and two-thirds of adolescents displaying SR and attending a treatment facility were reported to have experienced a depressive disorder (Carpentieri et al., 2022; McShane et al., 2001; Walter et al., 2010). Whereas separation anxiety is frequently associated with younger youths displaying SR, social anxiety is more typical of older youths displaying SR (Last & Strauss, 1990). Supporting this, Kearney and Albano (2004) found that youths seeking to escape aversive social and/or evaluative situations at school were older than youths seeking to avoid school for other reasons such as attention. In one study (Bernstein et al., 1999), more than two-thirds of adolescents displaying SR (12–18 years) were diagnosed with social anxiety disorder. Another study (Nair et al., 2013) revealed an 8-fold increase in social anxiety disorder among adolescents (11–19 years) displaying SR relative to those not displaying SR, even when controlling for comorbid depressive disorders. This corresponds to Ingul and Nordahl’s (2013) finding that anxious adolescents (16–21 years) not attending school were more socially anxious than anxious adolescents attending school. They also found that the former group had fewer friends. Carpentieri et al. (2022) similarly reported that help-seeking adolescents (14–18 years) displaying SR had a higher level of social impairment than those not displaying SR. In short, adolescence is a more difficult and distressing stage of life because of the increase in social anxiety (Kearney & Albano, 2018) and problematic social functioning seems to be a distinguishing feature of SR in adolescence (Carpentieri et al., 2022).

The complex clinical presentation associated with adolescent SR, often involving depression and/or social anxiety, likely contributes to poorer treatment outcomes. Bernstein and colleagues (2001) argued that depression among adolescents displaying SR may have interfered with their ability to participate in CBT. They cited the example of the difficulty youths have getting out of bed in the morning. Tolin and colleagues (2009) reported that depression may have made it difficult for one of the adolescents participating in CBT for SR to identify meaningful rewards. Social anxiety disorder is associated with a poorer response to treatment for SR (Layne et al., 2003), even long after the treatment has ended (McShane et al., 2004). Social anxiety disorder among adolescents displaying SR is also treatment-resistant (Bernstein et al., 2001; Heyne et al., 2011). The co-occurrence of social anxiety and SR presents a “double dilemma” for practitioners: School nonattendance reduces opportunities to build social confidence, and low social confidence reduces the inclination toward school attendance (Heyne & Maynard, 2016). Evans (2000) suggested that older SR youths who perceive social danger at school tend to have higher levels of anxiety in their younger years and perhaps traumatic experiences, so when they are confronted with challenging situations in later years, their anxiety becomes highly exacerbated. Being bullied may be one such traumatic experience, and it is commonly associated with absenteeism (Gren-Landell et al., 2015) and SR (Havik et al., 2014, 2015; Heyne et al., 2022; Place et al., 2002). For some adolescents, the experience of being bullied at primary school may prime them to be more sensitive to difficult social experiences at secondary school, which could complicate the process of returning to school and jeopardize treatment outcomes.

The clinical presentation of SR is not limited to diagnostic profiles; it includes interactions between youths and parents regarding school attendance. Hansen and colleagues (1998) suggested that older youths are more likely to resist – and in some cases more physically capable of resisting – their parents’ and teachers’ efforts to get them to school. Evans (2000) suggested that older youths may use more elaborate avoidance strategies, such as inducing vomiting or claiming racial problems at school. The more elaborate the avoidance strategy, the more complex the practitioner’s work may be.

Developmental Challenges Inherent to Adolescence

Developmental challenges inherent to adolescence likely contribute to the development and complexity of SR in adolescence, higher rates of referral, and poorer treatment outcome. As noted by Rubenstein and Hastings (1980), difficulties negotiating developmental tasks can present as SR in adolescence, including difficulties with individuation from the family, defining a sense of self, and responding to sociocultural pressure from peers. The challenges associated with the secondary school context and the family context are described next.

Secondary school context

It has long been argued that, as children grow older, school attendance problems are increasingly under the influence of school-based factors (Galloway, 1985). One likely explanation is the changing demands associated with school attendance. Nishida et al. (2004) argued that the first step in school attendance “is to leave mothers whom children depend on or [to leave] home where children feel comfortable. The second step is to attend groups in schools, which they must join” (p. 246). Accordingly, they argued that primary school youths (6–12 years) stumble at the first step and junior high school youths (12–15 years) stumble at the second step. Indeed, as noted above, separation anxiety disorder is more common among children displaying SR than among adolescents, supporting the notion that primary school youths stumble at the first step. Furthermore, youths who displayed SR and had a primary diagnosis of a separation anxiety disorder (i. e., the younger youths) showed greater improvement in attendance than youths who had other anxiety disorders (Last & Hansen, 2001, cited in Layne et al., 2003). This suggests that the developmental challenge associated with the second step – fitting in socially – may contribute to the inferior outcomes for adolescents displaying SR when compared to children displaying SR. In the words of Carpentieri et al. (2022, p. 25), “feelings of enjoyment and belongingness within a social group represent a fundamental developmental goal of adolescence” and failure to achieve this goal may lead to negative outcomes including SR.

With the transition from primary school to secondary school, youths are confronted with a larger and more complex social environment (e. g., more teachers and more students), and greater demands are put on their independence (e. g., moving between classes, engaging in more self-directed learning and time management) (Coffey, 2013; Hedges et al., 2014). These aspects of secondary school are likely to lead some vulnerable adolescents to become overwhelmed and escape to the security of the home environment, reflective of SR. Indeed, Hansen and colleagues (1998) argued that the higher rate of absence seen among older youths displaying SR occurs because teenagers have increasing difficulty coping with fears related to school. As noted above, adolescents are twice as likely as children to report school-related fears (Burnham et al., 2006). Evans (2000) also argued that, as school-leaving age approaches, adolescents may claim to be disinterested in continuing their education when in fact they are failing to cope with being at school; for some adolescents, dropping out of school is a way to permanently reduce their anxiety about attending.

Other developmental issues that are interwoven with the school context and potentially increase adolescents’ risk for SR and poorer treatment outcomes are identity formation (Rubenstein & Hastings, 1980), the increasing importance of academic performance in the context of complex curricula and high stakes testing (Elliott & Place, 2019), the increasing fear of social evaluation (Westenberg et al., 2007), the increasing importance of the peer context during adolescence, and the increasing relationship between school engagement and peer support (Li et al., 2011). Perhaps the socioemotional challenges inherent to the secondary school environment help to explain the higher prevalence of social anxiety disorder among SR adolescents than among children, beyond the fact that social anxiety disorder is generally more prevalent among older youths (Costello et al., 2003). Even if the greater demands of the secondary school environment do not explain the higher prevalence of social anxiety disorder among adolescents displaying SR than among children displaying SR, they will impact both parents’ and practitioners’ efforts to help adolescents attend school.

School climate and the related construct of school engagement are also associated with absenteeism and school drop-out (Bacon & Kearney, 2020; Bryant et al., 2013; Miranda-Zapata et al., 2018). No studies have directly examined these constructs concerning SR, though Ogilvie et al.’s (2019) report on youths (mean age 15.4 years) hospitalized with psychiatric illness did note an indirect relationship. These youths generally reported “considerable school avoidance … very low cognitive engagement in school … [and] somewhat low behavioral and emotional engagement with school” (p. 424). According to the authors, the reasons for school avoidance need to be understood to better support school engagement and thus school completion. They speculated that the youths’ emotional problems may negatively impact their ability to engage at school, and, conversely, that low school engagement may aggravate mental health problems and strain the youths’ effort or ability to cope.

Given the complexity of the secondary school setting and the convergence of developmental issues in adolescence, we need to direct our attention to the interplay between SR, on the one hand, and school climate and school engagement, on the other hand. Less school engagement, and thus greater risk for absenteeism, is more typical of adolescents who have experienced depressive symptoms (Dorio et al., 2019; Garvik et al., 2014) and peer victimization (Dorio et al., 2019), both of which are associated with SR in adolescence (Baker & Wills, 1978; Brouwer-Borghuis et al., 2019). Other studies addressing absenteeism suggest that adolescents might be more susceptible to SR, for example, when they feel unheard by school personnel (Bryant et al., 2013), have a negative attitude toward school (Gubbels et al., 2019), and their schoolwork is not career-focused (Plasman et al., 2021). The reader is referred to Van Den Berghe et al. (2022) for a discussion of strategies to promote adolescents’ and young adults’ engagement to prevent school drop-out.

Family Context

Changes within the family accompany the transition to secondary school. As the child moves through adolescence, developing a greater level of autonomy, there is greater potential for family conflict over a broad range of issues, including decisions about school attendance. For example, decisions need to be made about the nature and extent of the parents’ involvement in helping an adolescent attend school (Heyne & Sauter, 2013). A small qualitative study with three adolescents and their parents pointed to a dilemma for parents: “How to strike the right balance between how much the child is capable of and how much the parents should challenge the child” (Dannow et al., 2020, p. 31). Tolin et al. (2009) described a case in which an adolescent’s antagonistic relationship with his mother made it hard for her to implement contingency management procedures recommended during intervention for SR.

Conflict in families has been associated with the onset of SR among adolescents (McShane et al., 2001) and with the maintenance of SR (Kearney & Silverman, 1995). Curiously enough, Fornander and Kearney (2019) found that elevated scores for family conflict were predictive of 5+% absenteeism, while lower conflict scores were predictive of 10+% absenteeism4. It was suggested that some families may become so frustrated that they disengage from efforts to solve the school attendance problem, decreasing conflict. Whether family conflict precedes absenteeism or is a negative effect of absenteeism, it very likely negatively impacts the resolution of absenteeism. Indeed, Valles and Oddy (1984) reported higher levels of conflict among families of unsuccessfully treated youths presenting with SR than among the families of successfully treated youths.

Other forms of problematic relationships are observed in the families of youths displaying school attendance problems, including enmeshed, detached, and isolated relating (Kearney & Silverman, 1995). Two prominent themes in the SR literature are family enmeshment and insufficient independence for youths (e. g., Bernstein et al., 1990; Hansen et al., 1998; Last & Strauss, 1990; Place et al., 2000, 2002, 2005). Hansen et al. (1998) found greater rates of absenteeism among youths displaying SR when the youths came from families with less emphasis on personal development. Bernstein et al. (1999) found increased disengagement among the families of adolescents displaying SR, and the disengagement was interpreted as a possible reaction to earlier enmeshment in these families.

In all, one-half to two-thirds of families of youths displaying SR experience maladaptive family functioning (Bernstein et al., 1999; Kearney & Silverman, 1995). Summarizing three studies, Chockalingam et al. (2022) reported that family functioning was significantly more maladaptive among families of youths displaying SR than among families not affected by SR. The one study that focused on adolescents indicated that family dysfunction – including dysfunction in communication and problem-solving – differentiated between those displaying SR and those who attended school (Carless et al., 2015). Regarding treatment response, a more extreme family type (i. e., marked imbalance in cohesion and adaptability) was a marginally significant predictor of poorer outcomes 1 year after CBT for SR in adolescence (Bernstein et al., 2001).

The Extent of Developmental Sensitivity in Treatment Planning and Delivery

Developmentally sensitive CBT with youths, accounting for cognitive, emotional, and social development, constitutes “quality CBT” (Kendall & Peterman, 2015, p. 526). Practitioners ideally access literature, training, and supervision to support their developmentally sensitive work. Some treatment manuals assume practitioners have an adequate level of developmental sensitivity, while others explicitly address developmental issues. Treatment manuals that give due attention to the developmental issues inherent to adolescent SR may enhance treatment outcomes.

The developmentally sensitive delivery of CBT with adolescents displaying SR requires attention to their engagement in treatment, which is influenced by a broad range of developmental issues. The developing need for autonomy may make it difficult for the adolescent to accept help (Edgette, 1999, 2002) and may lead to resistance, detachment, or disengagement (Rubenstein, 2003; Stallard, 2002). Socioemotional development can have a considerable impact on participation in CBT. Youths with better-developed emotion regulation may be in a better position to acquire and use adaptive coping strategies learned in CBT (Bailey, 2001; Kingery et al., 2006; Suveg et al., 2009). It can be easier for youths with a greater capacity for social perspective-taking to participate in CBT because clients are often asked to consider and anticipate the effects of their behavior on others (Kinney, 1991). Adolescents with less capacity for self-reflection and insight into thoughts, feelings, and behaviors, in turn, may benefit less from CBT (Grave & Blissett, 2004), especially if treatment does not account for inadequacy or delay in these capacities. Kendall and Peterman (2015) added that adolescents’ busy schedules may impede CBT outcomes. Social engagements, extracurricular activities, and a high academic workload may negatively impact willingness to attend therapy and carry out CBT home-tasks. For a review of developmental issues relevant to planning and delivering treatment for anxious adolescents, see Sauter et al. (2009).

Developmental sensitivity in treatment planning includes consideration of the role of parents in CBT for SR, because their role is influenced by a range of developmental issues (Heyne & Sauter, 2013). For example, adolescents are more likely to want to decide for themselves about when and how they return to school, because of the drive toward independence, which may fuel opposition toward external control (Rubenstein & Hastings, 1980). Older youths are more likely to resist the efforts of parents and teachers to get them to school (Hansen et al., 1998). When parents endeavor to get their child to school, family stress and conflict likely arise, especially if the adolescent’s ideas about returning to school diverge from the parents’ ideas. Treatment manuals that encourage practitioners to provide support to families experiencing stress and conflict as they work toward school return may sustain the motivation of parents and adolescents for achieving school return. Further, the parents of a socially anxious adolescent may be less willing or able to get their child to attend school than the parents of a socially anxious child (Blöte et al., 2015), so treatment for SR needs to account for the higher co-occurrence of social anxiety among adolescents displaying SR. The higher rate of depression among SR adolescents also influences the planning and delivery of treatment for adolescents and parents.

Cognitive-Behavioral Therapy for School Refusal

The following is an overview of five CBT manuals for SR, their developmental sensitivity, and effectiveness. Developmental sensitivity was appraised via careful review of the manuals, except in one case where no manual was available, so descriptions of the treatment were based on two reports of its effectiveness5. Effectiveness was appraised via results identified in published studies, most of which were efficacy studies rather than effectiveness studies (see Johnsen et al., 2021, for a discussion of the difference). The overview of CBT manuals for SR is followed by an overview of four multimodal treatments including CBT for SR.

Cognitive-Behavioral Therapy Manuals for School Refusal

Five English-language CBT manuals have been developed to address SR (Heyne & Rollings, 2002; Heyne et al., 2008; Kearney & Albano, 2007, 2018; Last, 1993; Tolin et al., 2009)6. Table 1 presents the components of these manuals and summarizes studies in which they were evaluated.

Table 1 Scope and components of cognitive-behavioral therapy manuals for the treatment of school refusal

Elements common to all five manuals include delivery per case (as opposed to group treatment), psychoeducation, cognitive therapy interventions as optional or standard, graded exposure to school attendance, emphasis on the completion of home-tasks, and consultation with school personnel. Regarding graded exposure, Kearney and Albano’s (2007, 2018) treatments for positively reinforced school refusal behavior give less attention to this than the treatments for negatively reinforced school refusal behavior. All but one manual (i. e., Last, 1993) include problem-solving training with youths as optional or standard, relaxation training as optional or standard, and social skills training as optional. The manuals make implicit reference to preparing the young person for increased school attendance (e. g., first developing an exposure hierarchy), but Heyne and colleagues (2008) explicitly distinguish between a preparation phase and an implementation phase.

The manuals differ concerning the number of treatment sessions and the intensity of treatment. For example, Kearney and Albano’s (2007) interventions comprise approximately 8 sessions across 4 to 8 weeks, while the approach of Tolin and colleagues (2009) comprises 15 sessions across 3 weeks. Even though all manuals include standard or optional cognitive therapy, interventions are sometimes limited to training in coping self-statements to facilitate engagement in exposure tasks (e. g., Last, 1993). In other manuals, cognitive therapy involves the broader application of cognitive restructuring (e. g., Heyne et al., 2008; Kearney & Albano, 2007, 2018). Only three manuals incorporate family work on communication and problem-solving when indicated (Heyne et al., 2008; Kearney & Albano, 2007, 2018; Tolin et al., 2009). The earliest manual was standardized, with all cases receiving the same interventions. Later manuals advocate personalized treatment based on the main function(s) served by the young person’s behavior (Kearney & Albano, 2007, 2018) and/or a broader case formulation (Heyne & Rollings, 2002; Heyne et al., 2008; Tolin et al., 2009).

The five manuals also differ in the extent to which they help practitioners account for developmental issues when planning and delivering treatment, as described next.

Last (1993)

Last’s (1993) manual contains one explicit reference to the developmental tailoring of treatment. Additional support should be provided for very young children when they rate their fear and avoidance levels for items in the exposure hierarchy. Practitioners are encouraged to remind parents that the young person determines the rate of progress through the hierarchy, but this applied equally to the parents of younger and older youths.

The CBT treatment was compared to EST in a randomized controlled trial with 56 youths aged 6–17 years (Last et al., 1998). At posttreatment, 65 % of youths participating in CBT achieved at least 95 % school attendance. Both treatments yielded improved attendance and reduced emotional distress, and there was no posttreatment difference concerning the remission of the primary anxiety disorder. At the 1-month follow-up, 86 % of youths participating in CBT showed improvement in attendance, as opposed to 60 % of youths participating in EST.

Heyne and Rollings (2002)

Heyne and Rollings’ (2002) manual includes an introductory paragraph drawing the practitioner’s attention to the need to modify treatment according to the developmental level of the young person. Modifications include the use of more concrete cognitive strategies with younger children, considering the impact of academic pressure on secondary school students, utilizing more joint discussion between adolescents and parents, and the possibility that more time might be needed to engage adolescents in therapy. The description of treatment with parents includes no explicit reference to the youth’s developmental level.

The treatment was evaluated in a component analysis study in which 61 youths (7–14 years) displaying SR were randomly allocated to youth-focused CBT, parent/teacher-focused CBT, or a combination thereof (Heyne et al., 2002). At the 4.5-month follow-up, all conditions were associated with increased school attendance and reduced emotional distress, with no between-group differences. In all, 60 % of youths attended school at least 90 % of the time.

Kearney and Albano (2007)

Kearney and Albano’s (2007) manual for the treatment of school refusal behavior – which includes SR and truancy – comprises four treatments selected based on the predominant function(s) of the youth’s refusal to attend school: (1) to avoid school-related stimuli that provoke negative affectivity; (2) to escape aversive social and/or evaluative situations; (3) for attention; and (3) for tangible rewards outside of school. Treatments associated with the first three functions may be suitable for SR because these functions are often associated with emotional distress (Heyne et al., 2017), whereas the fourth function is held to be relevant to truancy (Gonzálvez et al., 2016; Kearney, 2008; Lyon & Cotler, 2009).

The therapist manual contains various developmentally relevant suggestions. For example, the suggestions for working with younger children are to use a slower pace, seldom use flooding, keep concepts and language simple, use more visual aids, and invite parents to help with home-tasks that are too difficult for the young person to complete independently. With adolescents, the development of a working alliance is emphasized. The therapist can decide on the level of parents’ involvement in exposure exercises based on the child’s developmental level. Forced school attendance by parents is only for youths under 11 years. The selection and dosing of treatment components is not explicitly guided by youths’ developmental level.

Kearney and Albano’s (2018) updated manual includes qualifications about the use of Socratic questioning, that it be used if developmentally relevant, and that a more directive approach to cognitive restructuring may be needed when working with younger children. Attention is also given to developmental milestones and tasks. For each of the four functions, there are suggestions for continuing treatment as needed, paying attention to the youth’s development. This is to help youths achieve social, academic, and other developmental competencies, such as handling negative feedback, dating, handling relationship breakups, making appointments for internship interviews, self-identity development, taking initiative to look for knowledge and opportunities, and living independently. Parents may also be helped to face their own developmental challenges such as learning to foster a child’s autonomy.

The authors regard the four interventions as being more relevant for Tier 2 cases (i. e., emerging, mild, or moderate SR) than Tier 3 cases (i. e., complicated, severe, and chronic SR) (Kearney & Albano, 2018). Positive outcomes have been documented in case studies (e. g., Chorpita et al., 1996; Kearney, 2002; Kearney et al., 2001; Kearney & Silverman, 1990) and a small, controlled study with 8 youths (6–16 years; Kearney & Silverman, 1999) but large-scale studies have not been conducted.

Heyne et al. (2008)

The @school program (Heyne & Sauter, 2013; Heyne et al., 2008) is an adaptation of Heyne and Rollings’ (2002) original manual. It was designed to increase the effectiveness of CBT for SR in adolescence by addressing a range of developmental issues. Three modules not included in the original manual were inspired by developmental issues associated with adolescent SR. The module “Thinking About the Teenage Years” was included to help the adolescent, parents, and practitioner consider the impact of developmental transitions and tasks upon the adolescent’s school attendance and well-being as well as upon the parents’ well-being and their relationship with their adolescent child. Information arising during the implementation of the module can influence the case formulation and treatment plan (e. g., a greater focus on the adolescent’s cognitions related to self; increased attention to overprotective parenting). The youth module “Dealing with Depression” was included because of the high rate of depression-related symptoms and disorders among adolescents displaying SR and the impact depression can have on adolescents’ participation in CBT for SR. The module “Solving Family Problems,” conducted with the adolescent and parents together, was included because of the impact that increasing adolescent autonomy and family conflict can have upon adolescent SR and its treatment (Heyne & Sauter, 2013; McShane et al., 2001).

Other modules were based on treatment elements in the original manual but modified in various ways. For example, the youth module “Dealing with Social Situations” gives greater attention to peer involvement (e. g., traveling to and from school with peers; peer support in keeping up with school-work) because of the increased importance and influence of peers. To better account for the developmental capacities of adolescents, the youth module “Dealing with Cognition” encourages practitioners to conduct a formal and informal assessment of adolescents’ CBT-relevant cognitive capacities. Cognitive therapeutic techniques associated with detecting and discrediting unhelpful cognition as well as discovering helpful cognition are selected according to how cognitively demanding they are (e. g., the less demanding procedure of self-instructional training to encourage the use of helpful cognition vis-à-vis the more demanding procedure of discrediting unhelpful cognition via Socratic questioning). Throughout treatment, practitioners are encouraged to use developmentally sensitive language and materials and to use additional procedures to enhance motivation as required (e. g., judicious use of contact via digital media; collaborative session planning; Sauter et al., 2009).

Parent involvement differs somewhat from the recommendations in the original manual. In adolescent SR, there is a need to help parents confidently respond to the interplay between their adolescent child’s SR, normative transitions and tasks (e. g., social and academic demands, wanting more autonomy), and the impact of these upon family functioning. As mentioned, parents participate in the modules “Thinking About the Teenage Years” and “Solving Family Problems.” The parent module “Facilitating School Attendance” encourages practitioners to consider how parents might best facilitate an adolescent’s school attendance, differentiating between an autonomy-granting (supportive) role and an authoritative (steering) role. The module “Bolstering a Parent’s Confidence” is drawn from the original manual but extended to promote identification and modification of unhelpful parent cognition, including cognition related to the management of an adolescent’s SR.

School consultation is especially important for adolescent SR (Heyne et al., 2008). By “lowering the hurdles” for school attendance (e. g., temporarily reducing academic demands; ensuring social support at school), the young person may feel more comfortable and confident to re-engage with schooling. Sometimes it is difficult to gain the cooperation of school personnel because they have limited time to address the needs of youths displaying SR, and this may be compounded in secondary schools because adolescents have contact with more teachers. When the practitioner visits the school during the assessment phase, a working alliance is fostered by taking time to learn about school personnel’s experience of and perspectives on the SR. School consultation also occurs before and after the planned school return. The content of the school-based modules is drawn from the original manual, but module descriptions prompt developmental sensitivity during implementation (e. g., fostering the use of developmentally sensitive reinforcement for attendance).

A nonrandomized trial of this developmentally sensitive CBT was conducted with 20 adolescents (11–17 years) whose current episode of SR averaged 6.5 months (Heyne et al., 2011). Significant improvements were observed for primary outcomes (school attendance, school-related fear, anxiety) and secondary outcomes (depression, overall functioning, adolescent and parent self-efficacy). Improvements were maintained at a 2-month follow-up, and effect sizes were medium to large. However, only 50 % of adolescents were free of anxiety disorder, and only 45 % were attending school regularly (i. e., at least 80 % of the time). The authors noted considerable room for improvement in treatment for adolescent SR, a topic addressed in another paper (Heyne, 2022).

Tolin et al. (2009)

The intensive (daily) CBT of Tolin and colleagues (2009) was originally trialed in a case series of four adolescents displaying SR7. The authors did not discuss their treatment with respect to developmental issues associated with the cases. Moreover, depression-related interventions were employed in just one case, although two adolescents were diagnosed with depressive disorder and another had elevated depression scores. Three of the four adolescents showed improvement in school attendance at posttreatment. There were mixed reports of psychosocial functioning at posttreatment; indications of some improvement were accompanied by indications of continued impairment and the need for ongoing treatment. At the 3-year follow-up, three youths who could be located were all engaged in alternative educational programs; none were attending their regular school.

Hannan et al. (2019) reported a small open trial of the intensive CBT for SR by Tolin and colleagues (2009), conducted with 25 youths (9–18 years) and their parents. There was no mention of tailoring treatment according to a youth’s developmental level8, except that problems with executive functioning indicated the need to help youths with planning and organizational skills. Across the sample, there was a significant pretreatment to posttreatment increase in school attendance, but this was based on construed categories of attendance rather than actual percent of attendance. 60 % of youths attended school more than 90 % of the time. There was a significant decrease in youth depression but not anxiety. Because 80 % of these youths were prescribed medication before entering the study, the unique influence of intensive CBT on posttreatment results is unclear.

Multimodal Treatments Using Cognitive-Behavioral Therapy Manuals for School Refusal

Four treatments include interventions for SR that are additional to manualized CBT (see Table 2). They are reviewed here because they signal efforts to address the challenging field of adolescent SR. Other multimodal treatments were not included in this review because they did not necessarily include CBT (McCune & Hynes, 2005; McShane et al., 2004, 2007), or because their focus was not on SR per se but included truancy (Reissner et al., 2015).

Table 2 CBT interventions and adjunctive interventions employed in multimodal treatments for school refusal

Bernstein et al. (2000) argued that multimodal treatment is needed for SR in adolescence “because of the significant psychopathology of these teenagers” (p. 276). They combined CBT based on Last’s (1993) manual with pharmacotherapy (imipramine) for adolescents who displayed SR and had been diagnosed with an anxiety disorder and major depressive disorder. No account was given of the need to be developmentally sensitive when planning or delivering CBT, beyond the one reference in Last’s manual to developmental tailoring (see above in Last, 1993). A comparison of CBT + imipramine and CBT + placebo among 63 adolescents (12–18 years) revealed reduced anxiety and depression for both groups, but only the CBT + imipramine group showed improvement in school attendance (Bernstein et al., 2000). However, just 54 % of adolescents receiving CBT + imipramine attended school at least 75 % of the time at posttreatment. High rates of disorder were found in a 1-year naturalistic follow-up study; almost two-thirds of the sample (i. e., youths receiving CBT + imipramine and youths receiving CBT + placebo) continued to meet the criteria for an anxiety disorder, and one-third continued to meet criteria for a depressive disorder (Bernstein et al., 2001). No attendance data was reported at 1-year follow-up. Based on the follow-up results, the authors concluded that aggressive treatments are important when SR is associated with comorbid anxiety and depressive disorders.

Melvin et al. (2017) reported the second study of CBT + medication specifically for adolescents displaying SR. CBT was based on Heyne and Rollings’ (2002) manual but with more sessions to permit greater attention to social skills training and depressive symptoms. Melvin et al. (2017) noted that these adaptations paralleled the published recommendations for work with adolescents displaying SR (Heyne & Sauter, 2013; Heyne et al., 2014). 62 youths (11–16.5 years) with an anxiety disorder and often with comorbid depressive disorder received CBT, CBT + fluoxetine, or CBT + placebo. Increased attendance and decreased anxiety and depression were observed for all groups, with no group differences. There was a medium effect for attendance, increasing from an average of 15 % attendance at baseline to 54 % at 1-year follow-up. Only 35 % of youths attended school at least 80 % of the time at follow-up, and 40 % still had an anxiety disorder. Based on the results, the authors emphasized the seriousness and persistence of SR and the great need for innovation in its treatment.

Walter et al. (2010) conducted an observational study of an inpatient program provided to 147 adolescents (12–18 years) with chronic school absenteeism and an anxiety disorder (40 %), depressive episode (27 %), or mixed disorder of conduct and emotions (33 %). Thus, all youths had some form of emotional distress, fulfilling a key criterion for SR (Berg, 2002)9. Adolescents received self-management therapy and CBT and had access to other interventions such as a special school and pharmacotherapy. The self-management therapy was adapted for adolescents to address self-esteem problems and problems in achievement and relationships. This is meritorious because the intervention attends to key issues in adolescent development. Specific CBT interventions were based on the interventions proposed by Kearney and Albano (2007). Results at the 2-month follow-up were positive concerning increased attendance and reduced anxiety, depression, and disruptive behavior. 74 % of the youths attended school continuously at 2-month follow-up, but only 63 % of all youths attended school in the regular school system; others had dropped out of school or still needed to attend a special school setting. Walter et al. (2014) subsequently reported a within-subject controlled study of the same inpatient treatment, showing positive results for school attendance. At 9-month follow-up, 64 % of adolescents were attending school regularly (some within a special school setting) or were employed. There was no effect for anxiety or depression. An explanation offered by the authors is that treatment was focused on increasing attendance and reintegrating adolescents into the regular school system, and not on mental health problems.

Wu et al. (2013) provided CBT + fluoxetine or CBT alone for 75 youths aged 8–18 years. CBT was based on Heyne and Rollings’ (2002) manual but conducted across 12 weeks instead of 4. Both treatment groups displayed significant reductions in anxiety and depression. A month after treatment there was no group difference concerning the percentage of youths achieving at least 80 % school attendance (CBT = 72 %; CBT + fluoxetine = 82 %). Results were not discussed concerning age.

Autism and Cognitive-Behavioral Therapy for School Refusal

Autistic youths are especially vulnerable to disengaging from school: school-related anxieties can stem from the complexity of autism characteristics and impact school attendance and engagement (Preece & Howley, 2018). It is only recently that studies have addressed school attendance among autistic youths.

Some studies addressed absenteeism generally (e. g., Munkhaugen et al., 2017, 2019; Ochi et al., 2020), while others included a focus on SR. For example, SR was found to be the most common reason for the absence of autistic youths in community samples (Adams, 2021; Totsika et al., 2020). Professionals working with youths referred for SR estimated that around 50 % have an autism spectrum disorder (Heyne et al., 2022; McKay-Brown & Birioukov-Brant, 2021). SR among autistic youths increases with age (Adams, 2021; Totsika et al., 2020), which might be explained in part by the difficulty they have transitioning from primary to secondary school (Mandy et al., 2016), and the fact that autistic youths experience the secondary-school environment as challenging because of inconsistencies in teachers and their expectations (Hedges et al., 2014). Two studies indicated that more than 50 % of autistic youths bullied at school asked their parents if they could stay home from school (Bitsika, Heyne et al., 2021; Bitsika, Sharpley et al., 2021), reflective of emerging SR. This is concerning given that autistic youths in mainstream schools are more likely to be bullied than those in special schools (Humphrey & Hebron, 2015). The co-occurrence of autism and attention-deficit/hyperactivity disorder further increases the risk for SR following bullying (McClemont et al., 2021).

The occurrence of SR among autistic youths underscores the need for interventions that respond to their special needs and competencies. Unfortunately, the lack of research on SR among autistic youths until quite recently mirrors the lack of attention to autism in treatment manuals for SR. Indeed, autistic youths were often excluded from evaluations of CBT for SR (Hannan et al., 2019; Heyne et al., 2011; Tolin et al., 2009) and multimodal treatments incorporating CBT (Walter et al., 2010, 2014; Wu et al., 2013). Just one CBT manual for SR includes considerations for addressing absence among autistic youths. Kearney and Albano (2018) suggested that practitioners work closely with the school to develop individualized education plans, identify problems that might need to be addressed, and focus on the graded increase in attendance.

Practitioners addressing SR among autistic youths will benefit from (1) the literature on supporting autistic youths at school, (2) the literature on adapting CBT for autistic youths, and (3) ad-hoc suggestions about working with autistic youths displaying SR. These areas are addressed next.

There is considerable literature on supporting autistic youths at school, especially concerning social support. For example, Sansosti (2010) outlined tiers of increasing support at school to help autistic youths develop social skills, given their difficulty with social interaction. Dillon et al. (2016) reported that autistic and nonautistic youths in secondary school value group work within the classroom setting, although youths with autism have a much narrower peer group. Hodges et al. (2021) described a school-based intervention in primary school to improve school participation among autistic youths and their typically developing peers. The intervention does not target SR, but school participation is understood to incorporate both school attendance (being there) and involvement (the experience of participation while attending) (Imms et al., 2016). Part of the impetus for developing the intervention was the impact of autistic youths’ social communication skills on their participation at school, including the establishment of friendships, engagement in social interactions, and asking for help (Hodges et al., 2020). Dean and Chang’s (2021) systematic review of school-based social skills interventions for autistic youths in inclusive preschools, primary schools, and secondary schools suggests that these interventions help improve social outcomes.

There is also considerable literature on adapting CBT during intervention with autistic adults (e. g., Spain & Happé, 2020) and youths (e. g., Cooper et al., 2018; Walters et al., 2016), including adaptations for autistic youths with anxiety (e. g., Chalfant et al., 2011). For example, practitioners are encouraged to use a slower pace (e. g., longer sessions), be more concrete (e. g., more visual information), draw on special interests to create engaging analogies that illustrate the principles of intervention, emphasize emotional recognition, use a more structured and less Socratic approach during cognitive interventions (e. g., multiple-choice options when identifying alternative thoughts), increase opportunities for exposure, and use “social stories, acronyms and role-plays to accommodate features of autism spectrum disorder including literal understanding and theory of mind deficits” (Walters et al., 2016, p. 149). A recent meta-analysis indicated that adapted CBT reduces anxiety among autistic youths at posttreatment, and that there is insufficient data to support continued benefits at follow-up (Sharma et al., 2021). A study underway addresses the effectiveness of CBT for SR delivered in a real-world setting, in which a high proportion of youths have an autism spectrum disorder (Johnsen et al., 2021).

There are various ad-hoc suggestions for working with autistic youths in the literature on attendance problems and SR. Russell’s (2022) recent chapter on absenteeism among youths with neurodevelopmental conditions suggests that the interventions used with typically developing youths can be adapted according to autistic youths’ strengths and difficulties, including psychoeducation, problem-solving training, graded exposure, and family work on communication. In particular, Russell recommends personalized expectations and supportive responses when youths arrive late to school, because refusal to attend school might be influenced by anxiety about the reactions of school personnel and by the youths’ rigid ideas that late arrival precludes attending for the rest of the day. Melin et al. (2022) interviewed practitioners about psychological intervention for autistic youths displaying SR. Some raised doubts about working with CBT techniques, while others emphasized the effectiveness of behavioral activation and exposure when these are tailored to autistic youths. Other suggestions for practitioners addressing SR among autistic youths include psychoeducation about autism and its relationship with SR (Melin et al., 2022; Sauter et al., 2010), assistance with academic skills such as planning (Sauter, 2010), consistency and structure at school (Heyne, Brouwer-Borghuis, et al., 2022; Preece & Howley, 2018), reducing bullying and improving school safety (Bitsika, Heyne, et al., 2021), facilitating social contact (Heyne, Brouwer-Borghuis et al., 2022) with a focus on individual social goals (Preece & Howley, 2018), involving parents and providing parental support (Bitsika, Sharpley et al., 2021; Melin et al., 2022), ensuring coordination between professionals from mental health, education, and social services (Melin et al., 2022), and providing more intensive intervention (Sauter, 2010).

Regarding more intensive intervention, the reader is referred to Ma and Travers’ (2022) practical guide about adjusting the intensity of intervention for youths with autism. While the guide primarily focuses on the in-class behavior of youths, it may help with the decision-making about intensifying components of intervention for SR. Alternative educational programs have also been employed with autistic youths displaying SR (Brouwer-Borghuis et al., 2019; Preece & Howley, 2018), discussed in an accompanying paper (Heyne, 2022).

Conclusion

SR in adolescence differs from that in childhood; referral rates are higher and response to treatment appears to be lower. This paper presented possible explanations, including higher absenteeism, a greater likelihood of co-occurring social anxiety disorder and/or depressive disorder, and the developmental challenges inherent to adolescence such as increased academic and social demands in the secondary-school environment and increasing autonomy that may contribute to family conflict. Developmental issues like these may potentiate and exacerbate an adolescent’s difficultly in attending school, also rendering it more difficult for families to cope with SR and complicating the efforts of practitioners to provide effective treatment. Because this was a nonsystematic narrative review, other explanations may be found in the literature not included in this paper. Furthermore, while attention was given to SR among autistic youths, this paper did not consider other challenges for adolescents, such as intellectual disability and learning difficulties.

The extent to which treatment is developmentally sensitive may also impact the outcome of CBT for SR. This paper presented an overview of five CBT manuals for SR and the extent to which they are developmentally sensitive. Some manuals offered minimal guidance for working in a developmentally sensitive way. Perhaps it was assumed that practitioners are aware of the need to work in a developmentally sensitive way and how that can be done. To improve outcomes in adolescence, the @school program explicitly addressed developmental issues in its design and delivery. Other efforts to improve outcomes for adolescents displaying SR have included multimodal treatments that incorporate CBT and other interventions such as medication, inpatient treatment, and temporary participation in an alternative educational program. Despite efforts to account for developmental issues when providing treatment for adolescent SR, treatment outcomes remain modest; around one-third to two-thirds of adolescents are not helped to attend school regularly. To address the persistence of absenteeism, we need further developments in treatments for adolescent SR. Hopefully the signposts offered in an accompanying paper (Heyne, 2022) prove helpful.

Finally, there are few empirical studies of the differences between SR in childhood and adolescence. Longitudinal studies that include wave-based measurement of school engagement, response to school transition, mental health and its problems, autonomy development, socioemotional functioning, family functioning, and school attendance could provide a clearer understanding of the interplay of developmental influences on SR. In treatment research, studies ideally examine similar variables, alongside age, as predictors, moderators, and/or mediators of outcome.

Literature

1 The terms “young person” and “youths” are used to refer to children and adolescents collectively.

2 These authors applied the term “school phobia” when youths had a score of 3 or higher on the School Phobia subscale of the Screen for Child Anxiety Related Emotional Disorders (Birmaher et al., 1999).

4 These authors investigated absenteeism in general, not SR specifically.

5 The treatment employed by Tolin et al. (2009) and later by Hannan et al. (2019) was described in these two publications.

7 The presence of internalizing behavior and the lack of concealed absenteeism and severe antisocial behavior point to the presence of SR.

9 Youths with conduct disorder in the absence of comorbid anxiety or depressive disorder were excluded, likely reducing the number of truanting youths in the sample.

3 “School refusal behavior” is used by some to refer to various school attendance problems, including school refusal and truancy.

6 Two manuals not in English and not specific to SR are Back2School, which also targets truancy (Thastum et al., 2019), and Hemmasittarprogrammet, which targets school absenteeism more generally (Strömbeck et al., 2021).

8 This is based on the description of treatment provided in the published article.