PROTOCOL: Psychosocial Interventions for School Refusal Behavior With Elementary and Secondary School Students

The Problem, Condition or Issue

School refusal behavior is a psychosocial problem for students characterized by severe emotional distress and anxiety at the prospect of going to school, leading to difficulties in attending school and, in some cases, significant absences from school (Kahn, Nursten, & Carroll, 1981). In addition to severe emotional upset, researchers have differentiated school refusal from truancy in terms of two other features: children who exhibit school refusal behavior remain at home with their parents' knowledge, and they do not exhibit characteristics of conduct disorder (Elliot, 1999; Heyne, King, Tonge, & Cooper, 2001).

In the past two decades, the conceptualization and definition of school refusal behavior has evolved. Discussion has ensued in the literature as to whether school refusal behavior should encompass any reason for students being absent from school, as Kearney (2007) suggests, or whether school refusal behavior should be distinguished from truancy as a different type of school attendance problem, as Heyne and colleagues (2001) recommend. Although there is no definitive consensus, there seems to be general agreement among a number of scholars in this area that there are different types of “nonattenders” (Elliott, 1999; Heyne et al., 2001). Scholars have described school refusal behavior fairly consistently, with a few exceptions (e.g., Kearney, 2008), as a subtype of nonattender: students who have attendance difficulties resulting from emotional distress. Although there is a lack of consistency in the literature regarding the operationalization of school refusal, this review will adopt the definition of school refusal as school non-attendance associated with anxiety or distress.

The prevalence of school refusal behavior is difficult to ascertain, due to the discrepancy in how school refusal is defined and lack of any national reporting; however, most researchers estimate that less than 5% of school-age children exhibit school refusal behavior (Burke & Silverman, 1987; Elliott, 1999; Fremont, 2003; King, Ollendick, & Tonge, 1995; Ollendick & Mayer, 1984). The prevalence of school refusal is similar across socioeconomic groups and gender but is more common between the ages of 5 and 8 and 10 and 15, when children are either starting school or experiencing transitions between schools (Fremont, 2003; Heyne et al., 2001; Last & Strauss, 1990). Children who present with school refusal may meet criteria for multiple internalizing and externalizing behavior problems, including anxiety, depression, phobia, separation anxiety, aggression, temper trantrums, and non-compliance (Egger, Costello, & Angold, 2003; Heyne et al., 2001; Kearney, 2001).

School refusal is a complex problem that has been found to have multiple causes and be maintained by factors across the child's ecology (Thambirahah, Grandison, & De-Hayes, 2008). The onset of school refusal can occur gradually, either with no obvious etiology or as a result of a specific trigger. Stressors implicated in the onset of school refusal behavior include illness, problematic family dynamics, traumatic experiences, and school-related factors (Kearney & Bates, 2005). Individual factors associated with school refusal include personality characteristics, such as introversion, temperament, low self-confidence, and behavioral inhibition (Thambirajah et al., 2008). Family factors include increased rates of panic disorder and agoraphobia in parents of children with school refusal behavior, dysfunctional family interactions (e.g., overdependency, conflict, detachment, and isolation of family members), and poor communication (Bernstein & Borchardt, 1996; Martin, Cabrol, Bouvard, Lepine, & Mouren-Simeoni, 1999). Individual, family, and school context appear to be important to understanding the causes as well as the maintenance of school refusal behavior.

Children and parents experience significant adverse consequences from school refusal. A child may miss an excessive number of days of school, leading to poor academic performance and disruptions in social and extracurricular activities (King & Bernstein, 2001). School refusal may also negatively affect family and peer relationships (Berg & Nursten, 1996). Long-term problems in social adjustment may also occur, including psychiatric disturbance (Heyne et al., 2001).

The Intervention

Psychosocial interventions for youth who exhibit school refusal behavior generally fall into one of four categories: behavioral approaches, cognitive-behavioral therapy (CBT), family therapy, and non-behavioral, non-CBT individual therapy. All psychosocial interventions that aim to increase attendance and decrease anxiety in school-age youth who exhibit school refusal behavior will be eligible for inclusion in this review. Because we are interested in psychosocial interventions that can be implemented by school or mental health professionals, we will exclude solely pharmacological and medical interventions from this review; however, we will include studies that use pharmacotherapy as part of a psychosocial intervention.

Psychosocial interventions for reducing school refusal behavior can vary in format, duration, setting, treatment components, and intervention targets. The formats can include individual, group, and/or family interventions. Duration can vary from brief interventions (6–8 weeks) to interventions that span across a school semester or school year. Interventions are most commonly delivered in a school or clinic setting. Treatment components can vary as well. Although most school refusal interventions involve CBT, they employ a number of different strategies that vary from program to program. School refusal interventions are generally either child or parent-focused. Master's or doctoral therapists generally deliver school refusal behavior interventions to children individually or in a group setting and to parents in the form of parental skills training.

How the Intervention Might Work

Behavioral interventions for the treatment of school refusal typically employ relaxation training, exposure-based strategies, contingency management, and/or social skill training techniques to reduce the children's fears and anxiety and help them return to school. Relaxation training involves teaching children to employ strategies to relax or calm themselves when presented with stressful or anxiety provoking situations. In the case of school refusal behavior, relaxation training is employed as a means of reducing feelings of psychological and physiological arousal and somatic symptoms associated with school or separation from a parent (King, Heyne, & Ollendick, 2005). Exposure-based interventions, such as systematic desensitization and flooding, have also been utilized in the treatment of school refusal behavior. Exposure-based interventions are designed to expose children to the stressful event (i.e., returning to school) to help them overcome their avoidance to school, either in the slow, gradual manner utilized with systematic desensitization, or the forced, rapid return to school utilized with the flooding technique. Contingency management, drawing from operant conditioning principles, is another commonly utilized behavioral intervention to treat school refusal behavior (Elliott, 1999). Contingency management involves providing home or school-based rewards to attend school and increase the amount of time children stay in school. Social skills training is a commonly utilized behavioral intervention for children who exhibit school refusal behavior as a result of poor peer relationships or social anxiety (King et al., 1998). Social skills training frequently focuses on teaching assertiveness and other social skills through the use of modelling and rehearsal of strategies; such skills should help children better prepare for and cope with social interactions with peers or teachers and reduce social anxiety. Behavioral interventions are employed directly with students and/or their parents. Family interventions using behavioral strategies typically train the parents in behavioral management and contingency contracting techniques, which parents can then implement in the home. Skills training to enhance parenting and problem-solving skills are often included as well (Elliot, 1999; Fremont, 2003; Kearney & Bates, 2005).

Combining behavioral interventions with cognitive therapy, cognitive-behavioral interventions (CBT) employ the use of cognitive therapy to challenge inappropriate or problematic beliefs that may be contributing to students' anxiety and refusal to attend school. Therapists assist students in identifying, monitoring, and replacing faulty beliefs with more adaptive self-statements to decrease anxiety related to their refusal to attend school (Elliott, 1999). Cognitive-behavioral interventions may be delivered individually or in group settings. In addition, CBT may also be delivered with parents or families. For parents, CBT is generally designed to help parents understand their role and reduce anxiety and faulty cognitions that may be contributing to their child's anxiety and refusal behavior. In addition, CBT for parents and families can involve training parents to use behavior management strategies to reinforce their children's return to school.

Behavioral and cognitive-behavioral approaches have received the most attention in prior reviews; however, a range of additional strategies have been used in the treatment of school refusal behavior. A broad range of theoretical models targeting different mechanisms informs these approaches. Non-CBT and non-behavioral psychosocial school refusal interventions typically target the mechanisms contributing to anxiety as hypothesized by the model being utilized. Psychosocial interventions outside of behavioral or CBT models include educational-support strategies and non-CBT individual and family therapy. The strategies employed by these approaches range from providing information and supportive psychotherapy to assist students in understanding and overcoming their fears and anxieties about school, to family therapy techniques aimed at changing family functioning and dynamics that are hypothesized to contribute to the child's anxiety and difficulty attending school—such as family conflict, enmeshed or detached family relationships, and fear and anxiety of the parents.

Why it is Important to do the Review?

Although there have been systematic reviews and meta-analyses conducted on the effects of interventions for children and adolescents with anxiety disorders (see Brendel, 2011; Ishikawa, Okajima, Hirofumi, & Sakano, 2007), we have not located a systematic review or meta-analysis of interventions focused specifically on school refusal behavior. A number of narrative reviews on the topic of school refusal behavior are available and summarized in Appendix A. Prior reviews on school refusal behavior have focused on what is known about school refusal behavior in terms of etiology, prevalence, assessment, and treatment; however, few prior reviews have focused specifically on intervention outcomes. Reviews that have focused more specifically on outcomes of interventions have not employed systematic review methods or meta-analytic techniques to quantitatively synthesize the results.

King and colleagues (King et al., 2005; King, Tonge, Heyne, & Ollendick, 2000) conducted two reviews of school refusal intervention outcome studies. The 2000 review included eight published studies of CBT interventions using a range of research designs. Their report concluded, “At first glance, our review of research suggests empirical support for cognitive-behavioral therapy in the treatment of school refusal…” (p. 501). “However, since very few controlled studies have been reported at this stage in treatment research, it would be premature to extol the clinical virtues of cognitive-behavior therapy” (p. 506). King et al.'s 2005 review focused on a broader topic of anxiety and phobic disorders and included seven studies examining effects of behavioral or CBT interventions with school refusal behavior. Although the authors used substantially the same studies in both reviews, the two reviews came to different conclusions. In the 2005 study, the authors concluded, “Overall, school refusal has responded to CBT programs as demonstrated in a number of controlled studies, with general maintenance of gains” (p. 249).

Prior reviews of school refusal intervention research have been limited to published research and have primarily employed either qualitative or vote-counting methods for synthesizing study outcomes. Moreover, the existing reviews of school refusal behavior have tended to focus on CBT interventions. Taken together, the past reviews are very important to our understanding of school refusal behavior and can provide some evidence to guide interventions; however, they do not systematically or quantitatively address the question of whether and which interventions are effective for increasing school attendance and decreasing anxiety for children exhibiting school refusal behavior.

This proposed systematic review will expand and improve upon prior work in several ways. First, this review will apply a systematic and transparent process for searching, retrieving, and coding studies, including the search for unpublished studies. Using a systematic method to conduct the review of outcome research limits bias and reduces chance effects, leading to more reliable results (Cooper, 1998). Searching for unpublished studies could produce additional studies that have not been included in prior reviews. Further, explicitly and transparently describing the review process allows for others to replicate and expand the review to include new studies or criteria.

Second, this review will seek to include evaluations of a variety of interventions operating in a broader set of geographical contexts than previous reviews, including programs across the United States and other countries. This broader reach will allow for the possibility of identifying studies that may not have been included in previous reviews.

Lastly, prior reviews have been limited to a narrative approach, presenting a description of programs or using a vote-counting method to categorize outcomes of programs as significantly positive, significantly negative, or of no significance. The vote-counting method, however, disregards sample size, relies on statistical significance, and does not take into account measures of the strength of the study findings, thus leading to erroneous conclusions (Glass, McGaw, & Smith 1981). Meta-analysis, on the other hand, represents key findings in terms of effect size, rather than statistical significance. Thus, meta-analysis provides information about the strength and importance of a relationship, the magnitude of the effects of the interventions, and the characteristics of effective interventions.

Criteria for including studies in the review

Types of studies: To be eligible for inclusion in the review, studies must use an experimental or quasi-experimental design. Studies must include a comparison of treatment and control conditions to which students are randomly assigned or nonrandomly assigned. In addition, studies must either match on pretests, risk factors, and/or other relevant characteristics, use statistical controls, or report baseline data to examine group equivalence. This review will not include single-group pretest-posttest studies or other study designs.

Types of participants: This review will include school-age youth, defined as attending kindergarten through 12^th grade (or equivalent in countries with a different grade structure), who meet criteria for school refusal behavior. Because there is no consensus on what constitutes a “diagnosis” of school refusal behavior, this review will include only studies in which participants have both an attendance problem and anxiety or a similar clinical symptom(s) related to stress, mood, or anxiety that affects their school attendance. We will exclude studies in which participants have dropped out of school.

Types of settings: The review will include interventions conducted in any setting that serves primary or secondary school students. This review will not include studies conducted in residential facilities, as these settings are highly controlled and not typical of regular school settings.

Types of intervention: This review will include all psychosocial intervention types.

Types of outcomes: To be included, a study must assess intervention effects on school attendance and anxiety.

Geographical context: This review will include studies from any geographical context. The authors will make every attempt to translate studies in languages other than English for inclusion in the review. If translating a study is not possible due to a lack of resources, the authors will note the study but not otherwise include it in the review.

Time period: This review will include studies published between 1980 and the present, even though the research itself might have been conducted prior to 1980. Focusing on the past 30 years will lead to a comprehensive and contemporary review of interventions.

Exclusion criteria: As defined above, we will exclude studies that involve only medication, are conducted in a residential setting, or are conducted with youth who have dropped out of school.

Search strategy

We propose to include all studies that meet the inclusion criteria outlined above. We will attempt to identify and retrieve both published and unpublished studies. We will use several sources to identify eligible studies, including the following: 1)

Electronic databases

Databases 1)

Academic Search Premier

Research Registers 1)

Cochrane Collaboration Library

Search Terms and Keywords

Database-specific strategies will be explored for each database (i.e., age limiters) and thesauri will be consulted to utilize more precise search strategies within each database. We will use combinations of the following terms and keywords related to the outcomes of interest, intervention, and population to search the electronic databases: 1)

Outcome: attendance OR absen∗ OR anxiety OR “school refus∗” OR “school phobia” OR “school anxiety”

AND

Grey Literature Search

A search for grey literature will be conducted via the following strategies: 1)

Internet searches: Google and Bing will be searched utilizing the above key words employing Boolean logic and limiting the search to .gov and .edu domains.

Reference Lists

We will review reference lists of prior reviews and related meta-analyses for relevant studies. In addition, we will examine the references of the retrieved primary studies for potential studies relevant to the review.

Description of Methods Used in Primary Research

Studies to be included in this review will employ experimental or quasi-experimental research designs that compare outcomes for an intervention group to those for a control or comparison condition. Most potentially eligible studies include both pretest and posttest measurements. Pretest measurements generally occur at or immediately prior to the beginning of the intervention. Posttest measurements generally occur at the end of the intervention. The posttest measurements comparing the intervention and comparison conditions are the key outcome measurements of interest for the proposed review. Some studies may measure outcomes at time points following the posttest measure which will be considered as follow-up measures.

One study that exemplifies the methods likely to meet the eligibility criteria for the proposed review is a study of a cognitive-behavioral intervention conducted by King et al. (1998). In this study, thirty-four school-refusing children were randomly assigned to a cognitive-behavioral treatment condition or a waiting-list control condition. Treatment consisted of individual child cognitive-behavioral therapy plus parent/teacher training in child behavior management skills. Pretest and posttest measures reported of interest to this review included attendance and scores on the Revised-Children's Manifest Anxiety Scale.

Criteria for Determination of Independent Findings

We are interested in two primary outcomes for this review: attendance and anxiety. All codable effect sizes for these two outcomes will be extracted, and each outcome construct will be analyzed separately. We expect that some studies will provide more than one effect size for attendance and/or anxiety (e.g., report two measures of anxiety). This circumstance creates statistical dependencies that violate the assumptions of standard meta-analysis methods. In the case where the construct is measured in more than one way, we will retain only one of the effect sizes in the analysis. We will select the measure that has better psychometric properties, or, if there is a measure that is commonly used across several studies to measure the same construct, we will retain the effect size for the analysis using the common measure.

In cases where there are multiple points of follow-up for a given outcome measure, we will record all points of follow-up and conduct a separate analysis for effect sizes at similar points of follow-up. If there are an adequate number of studies with longitudinal follow-up, we will examine changes in effect sizes over time.

In cases where we encounter studies with multiple outcomes for dependent or overlapping samples (e.g., multiple treatments compared against one control group), we will code all of the effect sizes but only include one treatment/control comparison in the meta-analysis. We will select the primary, or most relevant, treatment being tested to include in instances where both treatments are within the subgroup pooled.

As some studies may be reported in multiple reports or multiple reports reported in single studies, care will be taken to ensure that the studies are reporting independent findings. If it is unclear whether reports and studies provide independent findings, the authors of the reports will be contacted.

Details of Study Coding Categories

We will code all studies meeting the inclusion criteria by using a coding instrument specifying the information to be extracted from each eligible study (see Appendix B). The coding instrument will include items related to bibliographic information and source descriptors; methods and procedures; context, nature, and implementation of the intervention; sample characteristics; and outcome data needed to calculate effect sizes.

To ensure reliability of coding procedures and decisions, the second and third authors will independently code 100% of the included studies and will compare coding decisions for all studies. These authors will discuss any inter-rater differences to refine coding schemes and resolve any discrepancies. The coders will consult the first author if they cannot resolve coding discrepancies, and the first author will make the final coding decision. We will use Excel to manage data and conduct descriptive analysis. We will use Comprehensive Meta-Analysis 2.2 to conduct meta-analysis.

Statistical Procedures and Conventions

We anticipate using standardized mean difference effect sizes for outcomes on continuous measures and odds ratios for outcomes presented as dichotomous variables. Main effects and moderator analysis will be conducted separately on each outcome construct with the latter done as multivariate (meta-regression) analysis when possible. Random effects statistical models will be used throughout unless a compelling case arises for fixed effect analysis. The main objective of the analyses will be to describe the direction and magnitude of the effects of the different school refusal interventions on the different outcome constructs. Additionally, the analysis will attempt to identify the characteristics of the study methods, interventions, and student samples that are associated with larger and smaller effects on the various outcome constructs. Based on prior theory and research, the following moderators will be examined for their influence on effect sizes: 1) treatment modality; 2) treatment duration; 3) theoretical basis of intervention (e.g., CBT, family systems, etc.); 4) grade level of sample; and 5) race or socioeconomic status of sample. Analysis will be conducted using Comprehensive Meta-Analysis 2.2 (CMA) when possible. Currently, CMA 2.2 only allows for meta-regression with continuous scaled variables. For meta-regression with categorical variables, we will use SPSS or STATA with appropriate macros.

Summary and descriptive statistics of the study-level contextual characteristics, methodological quality characteristics, and participant and intervention characteristics will be used to describe the included studies.

Sensitivity analysis will be conducted to examine the potentially biasing affects of outliers and studies not reporting or controlling for group differences. If necessary, additional sensitivity analysis will be conducted if other issues arise that may impede our confidence in the estimated pooled effect size estimates.

A test of homogeneity (Q-test) will be conducted to compare the observed variance to what would be expected from sampling error. The Q statistic is distributed as a chi-square with k-1 degrees of freedom (k = the number of effect sizes). The Q statistic is calculated by adding the squared deviations of each study's effect size from the mean effect size, weighting their contribution by its inverse variance. A significant Q rejects the null hypothesis, indicating that the variability of effect sizes between studies is greater than what would be expected by sampling error alone. The I² statistic will also be used to describe the percentage of total variation across studies due to the heterogeneity rather than chance.

Publication bias will be assessed using the trim and fill procedure, and the resulting funnel plot produced in CMA will be visually inspected. Egger et al.'s Regression Test (1997) will also be used to assess the possibility of publication bias.

Treatment of Qualitative Research

Qualitative research will not be included in this review.