https://orcid.org/0000-0002-1174-0667
ABSTRACT
We investigated changes in youths’ intentions to quit school after following a group-based cognitive behaviour therapy (CBT) based intervention for depressed adolescents in upper secondary school: the Adolescent Coping with Depression Course (ACDC). Data were collected from 228 youths, 133 of whom received the 14-week ACDC intervention and 95 who received usual care (UC). Structural equation models showed that symptoms of depression at post-test positively predicted intention to quit school at six-month follow-up after the intervention ended (β = .291, p < .001). Furthermore, ACDC, which showed a decrease in depressive symptoms post-test, had an indirect effect on adolescents’ intention to quit school at six-month follow-up (indirect β = −0.117 p = .018). Hence, targeting depression and implementing effective preventive programmes may prevent dropout from upper secondary school. Such interventions should thus be investigated in future studies.
Dropping out of upper secondary school is a problem in the Scandinavian countries (Markussen, Citation2010; Statistisk sentralbyrå, Citation2020). Dropout is associated with preceding intention to quit school (Beck & Ajzen, Citation1991; Vasalampi et al., Citation2018), and they are both influenced by mental health issues like depression (Kearney, Citation2008). The aim of our study was to investigate whether a group-based cognitive behavioural therapy (CBT) intervention for depressed adolescents in upper secondary school (the Adolescent Coping with Depression Course, ACDC), which has been found to be effective in reducing depressive symptoms among adolescents (Garvik et al., Citation2014; Idsoe et al., Citation2019; Keles & Idsoe, Citation2021), could also reduce intention to quit school over time.
Ten years ago, the European Union had already set the objective of a dropout rate of less than 10% for all upper secondary schools in EU countries (Markussen, Citation2010). In contrast to this goal, 2019 statistics show that 78.1% of all students in upper secondary school in Norway complete their study programme within five years and as few as 59% finish with standard study progression (i.e., three years for study-preparational studies and four years for vocational studies). Although addressing dropout in upper secondary schools in Norway has been a long-term policy priority, dropout rates have remained at the same high levels, with some minor variations, for many years (Statistisk sentralbyrå, Citation2020). Completing upper secondary school is a necessity for continuing education and essential for qualifying for a job and developing important life skills (Frostad et al., Citation2015; Markussen, Citation2010).
To understand the process of dropping out of school, the intention of quitting school is examined. Intention is the factor that describes how much effort an individual is willing to expend in performing a specific behaviour. As a rule, the more favourable attitude and subjective norm a person has connected to a specific behaviour, and the greater the perceived behavioural control, the stronger is an individual’s intention to perform the behaviour i (Beck & Ajzen, Citation1991; Vasalampi et al., Citation2018). Intention to quit school as a behaviour can be expressed and understood in different ways. In previous research and literature focusing on school attendance problems or dropout, school refusal behaviour is a term used to describe a heterogeneous set of behaviours that can be expressed by both internal and external symptoms, such as anxiety and depression, and can lead to intention to quit school. The adolescent can experience difficulty with coming to, or staying in, classes for an entire day. School refusal behaviour is common in Norway and other countries and affects 5–28% of pupils (Kearney, Citation2002). There are many different reasons for attendance problems among adolescents, such as various degrees of anxiety, depression, and somatic nuisance, or other challenges like tantrums, defiance, running away from school, or misbehaviour/aggression in school.
Studies suggests that mental health issues are risk factors for educational non-achievement and dropout, and depression is one of the most common mental health issues amongst adolescents, with a peak for first onset between the ages of 15 and 21 years (Fletcher, Citation2008; Hankin et al., Citation1998). Depression is also the leading cause of disability worldwide (World Health Organization, Citation2020), with increased rates in adolescent years (Costello et al., Citation2011). In general, up to 15–20% of Norwegian adolescents report clinical range mental health problems including depression, but only 16–17% of these adolescents consult mental health services to get help with their mental health problems (Helland & Mathiesen, Citation2009; Sund et al., Citation2011).
Depression is a severe mental illness that can be difficult to define and measure. The illness can lead to different intensities of recurring and disabling behaviour, such as lower ability to concentrate, losing interest and pleasure in activities previously enjoyed, feeling low levels of initiative and self-esteem, and social withdrawal (Derdikman-Eiron et al., Citation2013). The various difficulties that follow depression can consequently lead to less cognitive capacity for academic achievements and initiative in learning (Derdikman-Eiron et al., Citation2011). Other features of depression that are related to academic achievements are fatigue or loss of energy. Loss of energy can reduce the ability to sustain work over longer periods of time, which is crucial for pupils and students so that they can reach their academic expectations or goals. Over time, experiences of not responding to their teachers, their parents, or their own expectations of themselves can lead to a lack of belief in their ability to influence events in their own life (Garvik et al., Citation2016).
While depression is used as the designation for the diagnosis, satisfying criteria given in the diagnostic manuals (APA, Citation2013; WHO, Citation2018), depressive symptoms are the designation of having symptoms of depression without necessarily satisfying the diagnostic criteria (i.e., having certain levels and durations of symptoms and so on). When experiencing difficult situations, relations or activities, an adolescent with depressive tendencies can interpret these difficulties as a product of their own abilities (internal reason), which they believe will never change (a stable situation), and see them as negatively influencing other social interactions (global effects), which eventually lead to the development of depressive symptoms and/or depression (Liu et al., Citation2015). Hence, if mental illnesses, like depression, are allowed to develop without treatment, they can have considerable effects on social functioning and relations, academic results, and other areas of life over time (Garvik et al., Citation2014). Because of the prevalence and long-term impact of depression and depressive symptoms, it is essential to identify and implement effective measures.
Different studies have examined the relation between depressive symptoms and school performance (Derdikman-Eiron et al., Citation2011; Garvik et al., Citation2016) and depressive symptoms have been linked not only to poor academic achievement but also to intention to quit school (Garvik et al., Citation2016; Honjo et al., Citation2001) and eventually increased rates of school dropout (Kearney, Citation2008). As is the case with depression, dropping out of school can have pervasive effects on adjustment in social, academic, and other areas of life. Even though international studies show that mental health is a factor for school dropout (Haynes, Citation2002), measures and interventions in Norway have taken this into account to only a limited extent (Idsoe & Keles, Citation2016).
Because intention to quit school is regarded as a significant predictor of dropout (Beck & Ajzen, Citation1991), we examined whether we could reduce intention to quit school via reducing youths’ depressive symptoms. Using data from the ACDC intervention (Idsoe & Keles, Citation2016; Idsoe et al., Citation2019; Keles & Idsoe, Citation2021), two research questions were explored:
Do depressive symptoms measured at post-test after the end of the intervention positively predict intention to quit school at six-month follow-up? That is, will youth with higher depressive symptoms at post-test have higher intention to quit school at six-month follow-up?
Will the ACDC intervention reduce youths’ intention to quit school measured at six-month follow-up through reducing their depressive symptoms in the intervention period?
As the prevalence of depression and depressive symptoms is higher among girls, and as it becomes more apparent in adolescence compared with boys (Lewinsohn et al., Citation1993; Lewis et al., Citation2015), it was important to include gender and age as control variables in the current study.
Method
The data came from the Adolescent Coping with Depression Course (ACDC), two-armed parallel cluster randomized control trial (RCT), with course leaders as the unit of allocation and youth participants as the unit of analysis (Idsoe et al., Citation2019; Keles & Idsoe, Citation2021). Data were collected from youths in Norway, through a self-report questionnaire before and after the ACDC intervention (i.e., pre-, post-, follow-up design). When reporting on this RCT, we followed guidelines in the Consolidated Standards of Reporting Trials (CONSORT) 2010 statement (Schulz et al., Citation2010) (see Appendix).
Sample and procedure
Participants were 228 youths attending upper secondary school in Norway, aged 16–19 years (88% girls; with a mean age of 16.70 years, SD = 1.14), who reported mild to moderate depressive symptoms and attended first or second grade of upper secondary school (Idsoe et al., Citation2019). All participants provided informed consent. The participants were recruited by the course leader and then allocated to the intervention (ACDC) or control (usual care, UC) group. The ACDC/UC leaders were employed in a place with a referral system such as community health clinics, public health services, and schools. Course leaders made contact with possible participants by presenting information about the intervention via different channels (posters, websites, newspapers etc.). The ACDC course leaders, who had a minimum of three-year university/university-college education in child/adolescent mental health, were certified after attendance at a five-day ACDC certification course (Idsoe et al., Citation2019; Keles & Idsoe, Citation2021). In total, 35 course leaders/facilitators participated in the study and were randomized to one of two conditions by research administrative personnel at the Norwegian Center for Child Behavioral Development using sequential numbering (Keles & Idsoe, Citation2021): 18 to the experimental condition (ACDC) and 17 to the control condition (UC).
The ACDC participants were invited to take part in the study after a screening interview by the ACDC/UC leaders, which involved a Beck Depression Inventory (BDI) screening followed by a semi-structured interview. This interview also helped determine whether the potential participants met exclusion criteria such as bipolar disorders, psychosis, substance abuse, ADHD, ADD (or other criteria that are listed in the ACDC manual) (Idsoe & Keles, Citation2016; Keles & Idsoe, Citation2021). After informing the participants about both conditions, without knowing which condition they were going to be assigned to, the participants signed a consent form (Idsoe et al., Citation2019). In total, the 35 course leaders recruited the 228 adolescents who were included in the study (Keles & Idsoe, Citation2021) is a flowchart detailing participants recruited to the study.
Figure 1. Flowchart of participants.
Intervention versus control group
Intervention group
The ACDC intervention is a course for adolescents with mild to moderate clinical symptoms of depression. During the intervention period, the ACDC participants learned to reflect on their own thinking style; in other words, to think of their own thoughts from a meta-perspective. By teaching them different skills and techniques, the course aimed to help adolescents be better able to cope with their depressive symptoms in future (Idsoe & Keles, Citation2016). The “toolbox” of skills and techniques the adolescents learned during the intervention (Idsoe et al., Citation2019) is based on a wide theoretical framework; from rational emotive behaviour therapy and cognitive behaviour therapy, but also from mega-cognitive theory, positive psychology and modern neurobiological perspectives (Keles & Idsoe, Citation2021). The ACDC course consisted of 10 sessions, usually delivered in a group format over eight weeks. After the main eight weeks, follow-up sessions were conducted about three and six weeks after the last session. Each session lasted about two hours, including breaks (Idsoe et al., Citation2019; Keles & Idsoe, Citation2021).
Usual care group
The UC group received the care they would normally access at the site from which they were recruited, with no restrictions. This usual care could involve the participants receiving a spectrum of different treatments from different care providers, such as psychologists, doctors, school nurses, and others. The different measures or types of therapy could involve pharmacotherapy, conversation therapy, or no therapy. None of the participants in the UC group received therapy from a CBT-based group programme that could overlap with the ACDC (Idsoe et al., Citation2019).
Measures
A dummy variable was used to evaluate the intervention effect, coded 0 for usual care and 1 for the ACDC intervention.
Intention to quit school was measured with a four-item scale developed by Studsrød and Bru (Citation2009). Sample items included: “If I could, I would have dropped out of school”, “I would rather work than go to school”, “I am tired of school”, and “I am considering quitting school”. The scale has the following response categories: “strongly agree”, “agree a little”, “disagree a little”, and “strongly disagree” (ibid). In the original scale, higher scores indicated lower intention to quit. For ease of interpretation, the items were reverse coded for this article; that is, higher scores here indicate higher intention to quit school. The Cronbach’s alpha for the total scale in the current study was .79 both for pretest (T1) and the six-month follow-up (T4). The measure has previously demonstrated satisfactory coefficients of reliability, as well as factorial validity (ibid).
Depressive symptoms were assessed by the Center for Epidemiologic Studies Depression Scale for Adolescents (CES-D; Radloff, Citation1977), which is a well-validated and applied self-report measure of depression. CES-D asks for the frequency of symptoms during the last week of depressed affect (7 items), lack of positive affect (4 items), somatic and retarded activity (7 items), and interpersonal problems (2 items). To complete the form, the participants chose one answer from four different levels of severity for each item: “rarely or none of the time”, “some or a little of the time”, “occasionally or a moderate amount of time” or “most or all of the time” (ibid). Total score ranged from 0 (no symptoms) to 60 (high level of and frequent symptoms) (Idsoe et al., Citation2019). Higher scores indicated higher levels and frequency of depressive symptoms. The Cronbach’s alphas for the total symptom scale in the current study varied between .88 and .92, respectively, across waves.
The control variables, age and gender, of the participants were collected at the screening interviews.
Procedure
The data were collected via self-report questionnaires. The first collection was a pretest of baseline measures for both variables (T1) and was captured at the screening interviews held from November to December 2015. Depressive symptoms were also measured a second time, just before the intervention started (T2): in January 2016 for the first cohort and January 2017 for the second cohort. After the intervention ended, depressive symptoms were measured at post-test (T3) in June 2016/2017, and intention to quit school was measured at six-month follow-up (T4).
Study ethics
This study has been approved by the Norwegian Regional Committee for Medical and Health Research Ethics (South East). Approval reference: 2015/1027 “Depresjonsmestring for ungdom”. In the first screening interview, all participants went through written information about the background and purpose of the project, knowing that they could be in the ACDC intervention or the UC group. The participants were also informed that their responses would be recorded in a safe way without their name or other recognizable information. Lastly, all participants were informed that participation in the original study was voluntary and that they could withdraw their consent or decline to participate at any time.
Statistical analyses
The analyses, in accordance with CONSORT, were based on the principle of intention to treat (ITT), both of which maintain the principle of randomization and make the groups comparable at a group level (Romundstad, Citation2021). The statistical software SPSS 21 was used for descriptive analyses, reliability analyses, and correlations, while Mplus 7.3 was used for analyzing the repeated data by examining autoregressive models of observed variables at pretest, post-test and the six-month follow-up stages. Because of the sample size restrictions, we were not able to run latent models. Due to the distribution of depressive symptoms in the adolescent population, the data are probably skewed, violating the criteria for the application of parametric tests. Because of this, MLR (maximum likelihood robust) estimator was preferred to accommodate non-normal item distributions (Muthén & Muthén, Citation2012). We employed the robust full information maximum likelihood (FIML) procedure, which uses all available datapoints and is consistent with the ITT approach. We consulted additional fit indices in accordance with the recommendations of West et al. (2012), who suggest that CFI > .95, NNFI > .90, and RMSEA < .05 represent a well-fitting model, while > .90, NNFI > .85 and RMSEA < .08 represent an adequately fitting model. Possible cluster effects must be accounted for when calculating appropriate sample sizes for a single-centre cluster-randomized effectiveness trial with an active control group because groups rather than individuals are randomized (Hayes & Bennett, Citation1999). Cluster effects of the main outcome (intention to quit school) were estimated using data from earlier studies on the ACDC material (Garvik et al., Citation2014; Idsoe & Keles, Citation2016) and intraclass correlations (ICCs) were calculated with the course groups as clusters. The calculated design effects smaller than 2 indicated no practical influence of clustering for our analyses, so the results of the analyses without correction for clustering are reported. The bootstrapping procedure with 1000 resamples was also employed to derive the 95% confidence interval (CI) for the mediating effect.
Results
Missing data and attrition
Within structural equation modelling (SEM), ITT can be conducted using FIML. This, however, is based on the assumption that the mechanism for missingness is completely random (MCAR—missing completely at random) (Enders, Citation2010). The same holds for listwise deletion (i.e., analyzing complete data only). However, in contrast to listwise deletion, FIML may also be applied if the mechanism is the less restrictive “missing at random” (MAR). MAR means that missing is considered random after controlling for measured variables that are correlated with missingness or with the outcome of interest. To examine the plausibility of MAR, we investigated several issues. First, we inspected the flowchart but found no systematic differences in attrition for the two study conditions. Second, the ethical committee gave us permission to contact the attrition group and to ask them questions about why they had dropped out. Based on their answers, we could not see any systematic differences between the two study conditions regarding reasons for dropping out. In addition, we also found that participants mainly dropped out for practical/administrative reasons. Modern literature on missing data regards such reasons for missingness to a large extent as random (Enders, Citation2010). Finally, we investigated all of our data and identified third variables that were related either to missingness or to the outcomes of interest. Adding them to our model either as control variables or as auxiliary variables increased the likelihood of having data MAR that may be analyzed using the FIML procedure, as implemented in Mplus (Enders, Citation2010), but not by analyzing complete data only.
We also conducted attrition analyses by comparing baseline values for the attrition group with the participant group (see Appendix). As can be seen from attrition analyses, no significant differences were detected, except for age at first pretest (i.e., the participants dropping out at T2 were older at T1 than those who remained at T2) and for depressive symptoms at post-test (i.e., the participants dropping out at T4 had higher depressive symptoms at T3 than those remained at T4).
Descriptive statistics
shows descriptive statistics for each variable for both ACDC and UC conditions. displays the correlations among study variables. Depressive symptoms and intention to quit school were significantly positively and moderately correlated across all time points; that is, higher scores on depressive symptoms covaried with higher scores on intention to quit school. Gender was only correlated with T1 pretest depressive symptoms (r = −.214, p < .001); in other words, girls had a higher level of depressive symptoms at the baseline level before the intervention started. On the other hand, age was correlated with T2 second pretest depressive symptoms; that is, older youths had a lower degree of depressive symptoms just before the intervention started.
Table 1. Descriptive statistics for the intervention (ACDC) and control group (UC) measures.
Table 2. Intercorrelations among variables for the overall sample (N = 228).
Structural equation model
We examined the hypothesized associations between the variables (i.e., intervention, depressive symptoms, and intention to quit school) with an auto-regressive path model (). Baseline depressive symptoms and intention to quit school were controlled in the mediation analyses, as well as gender (age was dropped from further analyses since it was not associated with any variables in the model). The post-intervention depressive symptoms (T3) were regressed on the intervention variable, and the six-month follow-up intention to quit school (T4) was regressed on post-intervention depressive symptoms (T3). The path diagram shows the longitudinal effect of the intervention on six-month follow-up intention to quit (T4) via depressive symptoms at post-test.
Figure 2. Path diagram of the longitudinal effect of the intervention on six-month follow-up intention to quit as a result of depression at post-test. Gender = Males coded 1 and females coded 2. Intervention = UC control was coded 0 and ACDC intervention 1. T1 = pretest, T2 = second pretest, T3 = post-test, and T4 = six-month follow-up. Standardized parameter estimates are reported only for the significant paths of the covariates. **p < .01, ***p < .001.
The results revealed that the indices of the proposed model were a good fit: SRMR = .050; RMSEA = .040, 90% CI [.000, .084]; CFI = .979; TLI = .964. In accordance with the first hypothesis, depressive symptom scores at T3 positively predicted participants’ level of intention to quit school at T4 (β = .291, p < .001). The results revealed that the group receiving the ACDC intervention experienced a greater decrease in symptoms of depression (β = −.402, p = .003). It is not the intervention itself that predicts the intention to quit (T4) score, but the effect between the intervention and the reduced depressive symptoms score at T3. The indirect effect of the intervention on intention to quit school was β = −.117 (p = .018). The model explained 22% of the variance in six-month follow-up intention to quit school variable. The bias-corrected, bootstrapped indirect unstandardized effects of the intervention on the intention to quit variable were also significant (95% CI [−.037, −.217]). This supports the second research question and implies a mediational process in such a way that the ACDC intervention led to a decrease in the adolescents’ depressive symptoms at the end of the intervention period, which, in turn, led to a decrease in the adolescents’ intention to quit school levels at six-month follow-up after the intervention.
Treatment fidelity and implementation
Intervention fidelity of the ACDC has already been described in detail by Keles et al. (Citation2022). Intervention fidelity of ACDC was examined across a comprehensive framework of five domains identified by the National Institutes of Health’s (NIH) Behavior Change Consortium (i.e., study design, training, intervention delivery, receipt of the intervention, and enactment of intervention skills). The results indicated that the level of fidelity did not reach 100% but still approached a high level of treatment fidelity. As for the dosage, weekly registrations conducted by the course facilitators revealed that, on average, the adolescents attended 6.5 of the 10 sessions. The intervention is copyrighted and distributed by Rådet for psykisk helse (https://psykiskhelse.no).
Discussion
The aim of this study was twofold: first, an exploration of the relation between depressive symptoms and intention to quit school and, second, an assessment of whether the ACDC intervention reduces the intention of youths to quit school by reducing their depressive symptoms.
Self-reported depressive symptoms were assessed at post-test directly after the intervention, and intention to quit school was measured at six-month follow-up. The control group received usual care during the intervention period. The results from the analyses showed that depressive symptoms were moderately related to intention to quit school. Furthermore, the mediation analysis showed that the ACDC intervention had a negative indirect effect on intention to quit school through its effect on lowering depressive symptoms after the intervention. That is, participants receiving the ACDC treatment decreased their intention to quit school at six months after the intervention ended due to the reduction in their depressive symptoms. Adolescents receiving the ACDC showed significantly lower depressive symptom scores at post-test compared to the UC group after controlling for the pretest levels of depressive symptoms (Idsoe et al., Citation2019).
Intentions are, in general, the best predictors of actual behaviour (Beck & Ajzen, Citation1991), and therefore the adolescents’ potential intention to quit school may act as an indicator of the possibility of their actually dropping out of upper secondary school. In the research we reviewed initially, it was suggested that depression is a potential precursor to developing intention to quit school, and thus it is considered that reducing depression may also reduce intention to quit school. Our findings support this notion. The link between mental health and school performance has been established by earlier studies (Frojd et al., Citation2008; Masi et al., Citation2001; Rose et al., Citation2011). Depressed adolescents can demonstrate lower academic achievement, a higher degree of social problems, and poor wellbeing and, as a result, have a greater risk of dropping out of school. In other words, depression can have a great impact on emotional, social, and academic development (Hammen, Citation2005). First onset of depression peaks between the ages of 15 and 21 (Hankin et al., Citation1998) and it is thus especially important to be aware of this fact in upper secondary school.
One of the challenges facing the implementation of treatment or programmes for depressed adolescents is that the nature of their problem may cause a lack of motivation in terms of seeking help (Idsoe et al., Citation2019). This may be especially so for boys; in the ACDC intervention in 2016, few boys attended. Unfortunately, boys are also overrepresented in school absence and drop-out statistics (Statistisk sentralbyrå, Citation2020; Utdanningsforbundet, Citation2022).
Intention to quit school is a multifaceted behaviour, often linked to problems with school attendance.
Lack of school attendance is a common problem in many countries, including Norway (Kearney, Citation2002). A vast majority of adolescents leave school without the competence and experience necessary for further studies and work. As a result, in the Norwegian context the principles of early help, early intervention and intensive training (Opplæringslova, Citation1998; Utdanningsdirektoratet, Citation2018) have been the focus of securing the rights of children in need of extra help from school and other support systems following the implementation of the Norwegian national curriculum (LK06) (Meld.St. Citation16, Citation2006–Citation2007). The latest government document from the Norwegian Ministry of Education established that early help and early intervention are equally important in upper secondary school, especially to prevent dropout (Meld.St. Citation6, Citation2019–Citation2020). The literature shows that psychoeducational programmes for treatment of mental health problems can be related to school absenteeism (Havik, Citation2018) and the results of this study indicate that it is crucial to implement the right measures for early detection of depression to prevent dropout and avoid severe depression in adolescence and adult life. The results of our study support the effectiveness of ACDC as an intervention to reduce intention to quit school in the long term.
Limitations
This study has several limitations that should be taken into account when interpreting the results. First, data are only self-reported, and therefore there is a chance of both under- and over-reporting of symptoms and intentions. The second limitation is the attrition rate. In longitudinal studies, there is always the likelihood that some participants will drop out of the study before all data is collected (Barry, Citation2005) (see , which details dropout during the intervention). These dropouts can distort the accuracy of the results. As a precaution to diminish attrition effects, we followed the principles of ITT, including FIML procedures as implemented in Mplus. Third, the study has limited power as a result of its small sample size and the fact that girls are overrepresented. The specific challenge of recruiting male participants can also indicate that they are particularly reluctant to seek help (Idsoe et al., Citation2019) or that the course leaders used recruitment strategies more appealing to girls.
Implications for practice
Our results underscore the fact that early detection of depressive symptoms among adolescents and early intervention for this group may actually have a positive effect on adolescents’ potential intention to quit school over time. As the ACDC can be delivered by a broad array of providers trained in a relatively short period of time, and can be easily implemented in the educational context, we believe that our findings may have important implications for the educational practice field.
Implications for future research
The results of analyses of ACDC material show significant reductions in depressive symptoms and intention to quit school over time. The present study has identified depressive symptoms as the mediator for the effect of the intervention on intention to quit school. However, future studies should use larger sample sizes and better gender balanced samples (Idsoe et al., Citation2019) in order to support the effectiveness of the intervention. Registry data for actual dropout rates for the students in the sample from the ACDC in 2016 will also be collected. This will make it possible to investigate whether ACDC may also prevent actual dropout from school and the relation of dropout to both depressive symptoms and intention to quit school.
Disclosure statement
No potential conflict of interest was reported by the authors.
Additional information
Funding
References
- APA. (2013). Diagnostic and statistical manual of mental disorders (DSM-5®). American Psychiatric Association.
- Barry, A. E. (2005). How attrition impacts the internal and external validity of longitudinal research. Journal of School Health, 75(7), 267–270. https://doi.org/10.1111/j.1746-1561.2005.00035.x
- Beck, L., & Ajzen, I. (1991). Predicting dishonest actions using the theory of planned behavior. Journal of Research in Personality, 25(3), 285–301. https://doi.org/10.1016/0092-6566(91)90021-H
- Costello, E. J., Copeland, W., & Angold, A. (2011). Trends in psychopathology across the adolescent years: What changes when children become adolescents, and when adolescents become adults? Journal of Child Psychology and Psychiatry, 52(10), 1015–1025. https://doi.org/10.1111/j.1469-7610.2011.02446.x
- Derdikman-Eiron, R., Hjemdal, O., Lydersen, S., Bratberg, G. H., & Indredavik, M. S. (2013). Adolescent predictors and associates of psychosocial functioning in young men and women: 11 year follow-up findings from the Nord-Trøndelag health study. Scandinavian Journal of Psychology, 54(2), 95–101. https://doi.org/10.1111/sjop.12036
- Derdikman-Eiron, R., Indredavik, M. S., Bratberg, G. H., Taraldsen, G., Bakken, I. J., & Colton, M. (2011). Gender differences in subjective well-being, self-esteem and psychosocial functioning in adolescents with symptoms of anxiety and depression: Findings from the Nord-Trøndelag health study. Scandinavian Journal of Psychology, 52(3), 261–267. https://doi.org/10.1111/j.1467-9450.2010.00859.x
- Enders, C. K. (2010). Applied missing data analysis. The Guilford Press.
- Fletcher, J. M. (2008). Adolescent depression: Diagnosis, treatment, and educational attainment. Health Economics, 17(11), 1215–1235. https://doi.org/10.1002/hec.1319
- Frojd, S. A., Nissinen, E. S., Pelkonen, M. U., Marttunen, M. J., Koivisto, A. M., & Kaltiala-Heino, R. (2008). Depression and school performance in middle adolescent boys and girls. Journal of Adolescence, 31(4), 485–498. https://doi.org/10.1016/j.adolescence.2007.08.006
- Frostad, P., Pijl, S. J., & Mjaavatn, P. E. (2015). Losing all interest in school: Social participation as a predictor of the intention to leave upper secondary school early. Scandinavian Journal of Educational Research, 59(1), 110–122. https://doi.org/10.1080/00313831.2014.904420
- Garvik, M., Idsoe, T., & Bru, E. (2014). Effectiveness study of a CBT-based adolescent coping with depression course. Emotional and Behavioural Difficulties, 19(2), 195–209. https://doi.org/10.1080/13632752.2013.840959
- Garvik, M., Idsoe, T., & Bru, E. (2016). Motivation and social relations in school following a CBT course for adolescents with depressive symptoms: An effectiveness study. Scandinavian Journal of Educational Research, 60(2), 219–239. https://doi.org/10.1080/00313831.2015.1017838
- Hammen, C. (2005). Stress and depression. Annual Review of Clinical Psychology, 1(1), 293–319. https://doi.org/10.1146/annurev.clinpsy.1.102803.143938
- Hankin, B. L., Abramson, L. Y., Moffitt, T. E., Silva, P. A., McGee, R., & Angell, K. E. (1998). Development of depression from preadolescence to young adulthood: Emerging gender differences in a 10-year longitudinal study. Journal of Abnormal Psychology, 107(1), 128–140. https://doi.org/10.1037/0021-843X.107.1.128
- Havik, T. (2018). Skolefravær: Å forstå og håndtere skolefravær og skolevegring (1 ed.). Gyldendal Akademisk.
- Hayes, R. J., & Bennett, S. (1999). Simple sample size calculation for cluster-randomized trials. International Journal of Epidemiology, 28(2), 319–326. https://doi.org/10.1093/ije/28.2.319
- Haynes, N. M. (2002). Addressing students’ social and emotional needs. Journal of Health & Social Policy, 16(1–2), 109–123. https://doi.org/10.1300/J045v16n01_10
- Helland, M. S., & Mathiesen, K. S. (2009). 13-15-åringer fra vanlige familier i Norge: Hverdagsliv og psykisk helse. Nasjonalt folkehelseinstitutt, 2009(1), 65. https://www.fhi.no/globalassets/dokumenterfiler/rapporter/2009-og-eldre/13-15-aringer-rapport-2009-1-pdf
- Honjo, S., Nishide, T., Niwa, S., Sasaki, Y., Kaneko, H., Inoko, K., & Nishide, Y. (2001). School refusal and depression with school inattendance in children and adolescents: comparative assessment between the Children's Depression Inventory and somatic complaints. Psychiatry and Clinical Neurosciences, 55(6), 629–634. https://doi.org/10.1046/j.1440-1819.2001.00916.x
- Idsoe, T., & Keles, S. (2016). Study protocol for a randomized controlled trial of a group cognitive-behavioral course for depressed adolescents. BMC Psychiatry, 16(1), 10. https://doi.org/10.1186/s12888-016-0954-y
- Idsoe, T., Keles, S., Olseth, A. R., & Ogden, T. (2019). Cognitive behavioral treatment for depressed adolescents: Results from a cluster randomized controlled trial of a group course. BMC Psychiatry, 19(1), 155. https://doi.org/10.1186/s12888-019-2134-3
- Kearney, C. A. (2002). Identifying the function of school refusal behavior: A revision of the school refusal assessment scale. Journal of Psychopathology and Behavioral Assessment, 24(4), 235–245. https://doi.org/10.1023/A:1020774932043
- Kearney, C. A. (2008). School absenteeism and school refusal behavior in youth: A contemporary review. Clinical Psychology Review, 28(3), 451–471. https://doi.org/10.1016/j.cpr.2007.07.012
- Keles, S., Bringedal, G. E., & Idsoe, T. (2022). Assessing fidelity to and satisfaction with the “Adolescent coping with depression course” (ACDC) intervention in a randomized controlled trial. Journal of Rational-Emotive & Cognitive-Behavior Therapy, 40(3), 583–602. https://doi.org/10.1007/s10942-021-00427-x
- Keles, S., & Idsoe, T. (2021). Six- and twelve-month follow-up results of a cluster randomized controlled trial of a CBT-based group course. Prevention Science, 22(4), 409–418. https://doi.org/10.1007/s11121-020-01160-0
- Lewinsohn, P. M., Hops, H., Roberts, R. E., Seeley, J. R., & Andrews, J. A. (1993). Adolescent psychopathology: I. Prevalence and incidence of depression and other DSM-III—R disorders in high school students. Journal of Abnormal Psychology, 102(1), 133–144. https://doi.org/10.1037/0021-843X.102.1.133
- Lewis, A. J., Kremer, P., Douglas, K., Toumborou, J. W., & Hameed, M. A. (2015). Gender differences in adolescent depression: Differential female susceptibility to stressors affecting family functioning. Australian Journal of Psychology, 67(3), 131–139. https://doi.org/10.1111/ajpy.12086
- Liu, R. T., Kleiman, E. M., Nestor, B. A., & Cheek, S. M. (2015). The hopelessness theory of depression: A quarter-century in review. Clinical Psychology: Science and Practice, 22(4), 345–365. https://doi.org/10.1111/cpsp.12125
- Markussen, E. (2010). Frafall i utdanning for 16–20 åringer i Norden (978-92-893-1999-7). https://www.norden.org/no/publication/frafall-i-utdanning-16-20-aringer-i-norden
- Masi, G., Tomaiuolo, F., Sbrana, B., Poli, P., Baracchini, G., Pruneti, C. A., Favilla, L., Floriani, C., & Marcheschi, M. (2001). Depressive symptoms and academic self-image in adolescence. Psychopathology, 34(2), 57–61. https://doi.org/10.1159/000049281
- Meld.St. 6. (2019–2020). Tett på - tidlig innsats og inkluderende fellesskap i barnehage, skole og SFO. https://www.regjeringen.no/no/dokumenter/meld.-st.−6-20192020/id2677025/
- Meld.St. 16. (2006–2007). … og ingen sto igjen - Tidlig innsats for livslang læring. https://www.regjeringen.no/no/dokumenter/stmeld-nr-16-2006-2007-/id441395/sec7
- Muthén, L. K., & Muthén, B. O. (2012). Mplus use’s guide (7th ed.). Muthén & Muthén.
- Opplæringslova LOV-1998-07-17-61 C.F.R. (1998).
- Radloff, L. S. (1977). The CES-D scale. Applied Psychological Measurement, 1(3), 385–401. https://doi.org/10.1177/014662167700100306
- Romundstad, P. R. (2021). ITT. In Store norske leksikon. https://sml.snl.no/ITT
- Rose, A. J., Carlson, W., Luebbe, A. M., Schwartz-Mette, R. A., Smith, R. R., & Swenson, L. P. (2011). Predicting difficulties in youth’s friendships: Are anxiety symptoms as damaging as depressive symptoms? Merrill-Palmer Quarterly, 57(3), 244–262. https://doi.org/10.1353/mpq.2011.0013
- Schulz, K. F., Altman, D. G., & Moher, D. (2010). CONSORT 2010 statement: Updated guidelines for reporting parallel group randomised trials. BMC Medicine, 8(1), 18. https://doi.org/10.1186/1741-7015-8-18
- Statistisk sentralbyrå. (2020). Gjennomgføring i videregående opplæring. https://www.ssb.no/utdanning/videregaende-utdanning/statistikk/gjennomforing-i-videregaende-opplaering
- Studsrød, I., & Bru, E. (2009). The role of perceived parental socialization practices in school adjustment among Norwegian upper secondary school students. British Journal of Educational Psychology, 79(Pt 3), 529–546. https://doi.org/10.1348/000709908X381771
- Sund, A. M., Larsson, B., & Wichstrom, L. (2011). Prevalence and characteristics of depressive disorders in early adolescents in central Norway. Child and Adolescent Psychiatry and Mental Health, 5(1), 28. https://doi.org/10.1186/1753-2000-5-28
- Utdanningsdirektoratet. (2018, 01.08.2018). Intensiv opplæring for elever fra 1. til 4. årstrinn. https://www.udir.no/laring-og-trivsel/tilpasset-opplaring/intensiv-opplaring/1/#
- Utdanningsforbundet. (2022, November 10). Frafall i videregående opplæring. https://www.utdanningsforbundet.no/var-politikk/utdanningsforbundet-mener/artikler/frafall/
- Vasalampi, K., Kiuru, N., & Salmela-Aro, K. (2018). The role of a supportive interpersonal environment and education-related goal motivation during the transition beyond upper secondary education. Contemporary Educational Psychology, 55, 110–119. https://doi.org/10.1016/j.cedpsych.2018.09.001
- West, S. G., Taylor, A. B., & Wu, W. (2012). Model fit and model selection in structural equation modeling. In R. H. Hoyle (Ed.), Handbook of structural equation modeling (pp. 209–231). The Guilford Press.
- World Health Organization. (2018). International statistical classification of diseases and related health problems (11th Revision). https://icd.who.int/browse11/l-m/en
- World Health Organization. (2020). Depression. https://www.who.int/news-room/fact-sheets/detail/depression
Appendix: Attrition Analyses
Comparison analyses (chi-square analyses for gender and t-test analyses for age and each outcome measure) for participants who dropped out versus those who remained in school in the following measurements are presented below.
CONSORT 2010 checklist of information to include when reporting a randomized trial*