Welkom bij THIM Hogeschool voor Fysiotherapie & Bohn Stafleu van Loghum
THIM Hogeschool voor Fysiotherapie heeft ervoor gezorgd dat je Mijn BSL eenvoudig en snel kunt raadplegen. Je kunt je links eenvoudig registreren. Met deze gegevens kun je thuis, of waar ook ter wereld toegang krijgen tot Mijn BSL. Heb je een vraag, neem dan contact op met helpdesk@thim.nl.
Om ook buiten de locaties van THIM, thuis bijvoorbeeld, van Mijn BSL gebruik te kunnen maken, moet je jezelf eenmalig registreren. Dit kan alleen vanaf een computer op een van de locaties van THIM.
Eenmaal geregistreerd kun je thuis of waar ook ter wereld onbeperkt toegang krijgen tot Mijn BSL.
Login
Als u al geregistreerd bent, hoeft u alleen maar in te loggen om onbeperkt toegang te krijgen tot Mijn BSL.
Aggression and depression are two of the most common forms of psychosocial difficulties in adolescents. High prevalence and disabling effects stress the need for more cost-effective interventions. Studies show that the two difficulties are closely linked regarding comorbidity, risk factors, consequences and theoretical background, making it interesting to investigate whether an intervention targeting one of them could also reduce the other. The aim of this study was therefore to assess the association between aggression and depression and to investigate whether a treatment that has been found effective in reducing depression—“Adolescents coping with depression course” (ACDC), a group-based cognitive behavioural intervention-, could also reduce aggression, either directly, or indirectly via reducing depression. The data came from a randomized control trial, with a pre- post- and follow-up longitudinal design. In total 228 youth participated in the study, of which 133 received the ACDC-intervention and 95 received usual care. Correlation analysis showed that aggression and depression were associated at all time points. Structural equation modelling showed that the intervention did have neither direct nor indirect effect on aggression. Our conclusion is that future CBT interventions with a primary aim to reduce depression might have to design intervention components that are more directly targeted toward aggression if also this potential comorbid difficulty should be addressed. Previous studies have found higher levels of comorbid aggression among depressed boys than girls. Few boys in our sample may thereby have reduced statistical power to detect potential influences on aggression.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Introduction
Among 10 to 20% of school aged children and adolescents suffer from symptoms of aggression and/or depression (Card, 2011; Sund et al., 2011). Both difficulties can have a serious negative impact on the youth's life if the health systems fail to provide effective treatments. Aggression and depression both increase the risk of poor academic performance, drug problems, school refusal and difficulties in maintaining good relationships with family and friends (Card, 2011; Garber, 2006). Despite the high prevalence of aggression and depression amongst youth, studies show that only a small percentage of adolescents with difficulties seeks professional help (Helland & Mathiesen, 2009; Verhulst & Ende, 1997; Zwaanswijk et al., 2003). This is a problem as this group is especially vulnerable to longer first episodes, relapse and comorbid diseases (Fletcher, 2008). The high prevalence as well as the serious potential outcomes stress the need for low-threshold evidence-based treatments.
Although aggression and depression are quite different in their form of expression, the two difficulties have similarities and are often closely linked. Studies show that there is a high comorbidity between depression and the reactive form of aggression (del Barrio Gándara et al., 2021; Card & Little, 2006; Fite et al., 2011). Reactive aggression is known as the “hot blooded” version of aggression. This form of aggression is a reaction to perceived provocation or threat that the aggressor interpreted as being intentionally hostile (Kempes et al., 2005). The comorbidity between reactive aggression and depression might be a result of social problems, which is both a risk factor and a consequence of the two difficulties. Some suggest that social problems therefore act as a mediating link between reactive aggression and depression (Fite et al., 2012).
On the contrary, the association with depression is not so established when it comes to the proactive type of aggression (del Barrio Gándara et al., 2021; Card & Little, 2006; Fite et al., 2012). Proactive aggression is understood as a more cold-blooded behaviour where the individual may use aggressive actions as a tool to gain things or social status. The background for a lower association between depression and proactive aggression may be linked to the fact that reactive-aggressive children/adolescents have poorer relationships with and experience more negative interactions with parents and peers than proactively aggressive children (Coie et al., 2006).
Reactive aggression and depression have a number of common features when it comes to symptoms and consequences. Depression and reactive aggression in young people both increase the risk of poor academic performance, drug problems, school refusal and difficulties in maintaining good friendships and family relationships (Card, 2011; Garber, 2006). Children and young people who show reactive aggression experience more social rejection and isolation than non-aggressive peers (Crick & Grotpeter, 1995; Crick et al., 2006; Marshall et al., 2015). Social rejection and isolation are again two problem areas that have a strong association with depression (Laursen et al., 2007; Witvliet et al., 2010). Whether it is aggression or depression that is the cause of social rejection and isolation is not so easy to know as the two difficulties overlap. For example, depression has a negative impact on self-regulation skills. This can lead to children and young people who are depressed misinterpreting the actions of others as negative and malicious and reciprocating the action with reactive aggression, which in turn leads to rejection by peers (Marshall et al., 2015). Conversely, aggression can be at the core of children and young people being rejected by their peers, which in turn is a risk factor for depression. According to research, whether it is aggression or depression that occurs first is contradictory. Some studies claim that depression can be the cause of incipient aggression, while other studies find evidence for the opposite (Blain-Arcaro & Vaillancourt, 2017; Burke & Loeber, 2010; Gordillo et al., 2022; Shin, 2010). Another theory is that aggression and depression mutually influence each other, which has a worsening effect over time (Marshall et al., 2015). As well as social relationships with peers, different familial conditions also appear to influence both aggression and depression. The biggest risk factor for depression is having a parent, especially a mother, who has or has had depression (Thapar et al., 2012). In comparison, maternal care helps to influence the likelihood that reactive-aggressive young people will later commit partner violence. For young people who grow up with less caring mothers, the probability that they will later commit partner violence is higher than for young people who do not (Brendgen et al., 2001).
In addition to prevalence, comorbidity and potential consequences, some theoretical frameworks describing the two difficulties share a lot of similarities. Beck’s cognitive theory for depression (Beck, 1979) and Dodge and Crick’s social information processing theory for aggression (Crick & Dodge, 1996) are both based on the idea that aggression and depression are results of maladaptive ways of interpreting social cues, and as a consequence reinforcing the negative emotions. Cognitive behavioural therapy (CBT) is a form of psychosocial treatment that aims to reduce these dysfunctional ways of thinking and behaving (Wright & Beck, 1983). Previous studies have found CBT to be an effective method of reducing symptoms of both aggression and depression (David-Ferdon & Kaslow, 2008; Smeets et al., 2015). However, these studies have examined the effect of CBT on aggression and depression separately. As the two difficulties are closely linked when it comes to comorbidity and theoretical framework, it would be interesting to see if a CBT intervention aimed at reducing depressive symptoms, also reduces symptoms of comorbid aggression.
In the current study we examined the association between aggression and depression using data from a randomized controlled trial of the Adolescents Coping with Depression Course (ACDC) which is a group-based CBT intervention for 14–20 year old adolescents with subclinical, mild or moderate depression symptoms (Idsoe & Keles, 2016). Previous studies on this intervention found that at posttest, the adolescents receiving the ADCD had significantly lower scores on depression symptoms compared to those receiving usual care. Even though the symptoms did not further decrease during follow-up, the discrepancy between ACDC and control groups remained the same (Keles & Idsoe, 2021). The overall effect size of the intervention was mild to moderate (Idsoe et al., 2019).
The aim of the current study was to investigate the association between aggression and depression and whether the ACDC had a direct or indirect effect on reducing aggressive symptoms at posttest and over a 6-month follow-up period. The study raises the following three research questions:
(1)
What is the association between aggression and depression?
(2)
Does the intervention predict aggression directly at posttest, and indirectly at follow-up?
(3)
Does the intervention predict aggression at follow-up indirectly via reducing depression at posttest?
Due to the higher prevalence of aggression among boys and of depressive symptoms among girls, as well as it becomes more apparent in adolescence compared with boys (Lewinsohn et al., 1993; Lewis et al., 2015), it was important to include gender and age as control variables in the current study.
Method
The data used in this study came from a two-arm parallel cluster randomized control trial (RCT).1 The study was conducted in Norway between 2015 and 2017 using self-report questionnaires, with course leaders being the unit of allocation and the youth participants being the unit of analysis (Idsoe & Keles, 2016; Idsoe et al., 2019). Neither the participants nor the course leaders knew which group they were allocated until the beginning of the intervention (Idsoe & Keles, 2016). Data measuring the same variables was collected over an extended period, with a pre- post- and follow-up longitudinal design. A number of considerations in regard to how the study was designed, conducted and reported were taken into account based on the Consolidated Standards of Reporting Trials (CONSORT) as a way to improve the quality of the study (Moher et al., 2010).
Participants and Course Leaders
The participants of the study were 228 youth attending 1st or 2nd grade of upper secondary school in Norway with an age ranging from 16 to 19 years old (Meanage = 16.70 years, SD = 1.14). 87.7% of the participants (N = 200) were girls. The youth had subclinical or mild to moderate depression, according to the criteria of DSM-5, without the presence of any of the exclusion criteria including bipolar disorder, ADHD/ADD, psychosis, substance use and brain damage (Idsoe et al., 2019). All the participants had sufficient language abilities to attend the course and provided informed consent.
Course leaders were recruited from places with referral systems such as public hospitals, school mental health offices, Educational Psychological Services and other out-patient services (Idsoe & Keles, 2016; Idsoe et al., 2019; Keles & Idsoe, 2021). To attend the five days certification course, there was a requirement that all the leaders had completed a minimum of 3-years college or university education. Randomization into either the ACDC or control conditions were done by administrative personnel at the Norwegian Center for Child Behavioral Development. 18 people randomized to the experimental group (ACDC) and 17 people randomized to the control group (UC) were eventually included in the study (Idsoe et al., 2019).
The ACDC/UC trainers provided information about the course by placing information on websites for youth, in the local newspaper, at schools and health centers (Idsoe & Keles, 2016). The participants were then recruited either by making direct contact with the ACDC trainers themselves, or being referred by practitioners such as health visitors, doctors, school counsellors, psychiatric clinics and municipal services (Idsoe & Keles, 2016; Idsoe et al., 2019). The course leaders did a screening of potential study participants using Beck Depression Inventory (BDI) (Beck et al., 1961). To be eligible to participate in the trial, the adolescents had to meet the cut-off criterion for mild to moderate depression with a BDI score of ≥ 10. In addition to the screening there was a brief clinical interview to determine whether the participants had any of the exclusionary diagnoses as mentioned above (Keles & Idsoe, 2021). To reduce post-randomization selection bias, the adolescents were informed about both conditions and signed consent without knowing which intervention they would receive (Idsoe et al., 2019) (Fig. 1).
Fig. 1
Participant flow chart for the sample in the ACDC study
Intervention Group: Adolescents Coping with Depression Course (ACDC)
Adolescents Coping with Depression Course (ACDC) is an early intervention CBT-based course for adolescents with subclinical or mild to moderate depression developed for the Norwegian context (Børve, 2012). The ACDC program is quite similar to the “Coping With Depression course” (CWA) but is mainly based on a more updated understanding of depression within CBT (Idsoe & Keles, 2016; Idsoe et al., 2019). In addition to CBT, which is the main program, the ACDC program contains also a few elements and methods from Rational Emotional Behavior Therapy (REBT), Meta-Cognitive Therapy (MCT) and Positive Psychology (PP) (Grieger & Ellis, 1977; Seligman, 2011; Wells et al., 2009). Based on these theoretical frameworks the ACDC course aims to reduce depressive symptoms by acquiring a set of skills and techniques that helps you modify and regulate your own thoughts. The ACDC course has a psycho-educational approach, meaning that the course leaders have a role of a facilitator more than a therapist, and the adolescents work quite independently through a standardized manual (Garvik et al., 2014) The ACDC course has a total of 10 sessions delivered in a group format. This includes 8 weekly sessions lasting approximately 2 h each session. About 3 and 6 weeks after the last session two follow-up sessions are conducted, lasting approximately 1,5 h. Each session has a specified topic and is ideally scheduled at the same time every week. More thorough details regarding course content are described in previous articles (Idsoe & Keles, 2016; Idsoe et al., 2019).
Control Group: Usual Care (UC)
Instead of being offered the intervention, participants in the control group received the usual care (UC) that the UC leaders typically would provide for this group of adolescents. This varies in accordance with procedures at different sites and included being referred to different care providers such as doctors, psychologists, school nurses and teachers. These care providers may offer a range of different standard treatments for instance conventions, pharmacotherapy or even no treatment at all. Both the UC-leaders and the participants in this group were asked to give a detailed explanation of what kind of treatment they received and from who. Even though there were no restrictions on what type of treatment the UC-group would receive, none of them reported CBT-based therapies or strategies that may overlap with ACDC Idsoe et al., 2019).
Measures
A dummy variable was used to evaluate the effect of the intervention. Usual care coded as 0 and the ACDC-intervention group coded as 1.
Aggression Symptoms
Reactive aggression symptoms among the participants were measured using an adapted version of Roland and Idsøe’s reactive aggression scale (Roland & Idsøe, 2001). The scale consists of 6 items. Examples of items were “I easily get angry” and “When I get angry, I easily lose control of myself”. The participants were asked to answer these statements using a four-point Likert scale. The answer options are 1. Strongly agree, 2. Somewhat agree, 3. Somewhat disagree and 4. Strongly disagree. Low scores on the scale originally indicated high levels of reactive aggression. However, this was reversed before the analyses to make the interpretation easier, hence the higher scores indicated higher levels of aggression Cronbach’s alpha for the total symptom scale in the current study ranged between 0.83 and 0.85.
Depressive Symptoms
Depressive symptoms were measured using the Center for Epidemiologic Studies’ Depression Scale for Adolescents (CES-D). CES-D is designed to measure depressive symptoms in the general population, using a short self-report scale (Radloff, 1977). The CES-D scale detects the presence and frequency of common depressive symptoms during the last week. Depressive symptoms are measured through a total of 20 items divided into different components; depressed mood, feelings of guilt and worthless-ness, feelings of helplessness and hopelessness, psychomotor retardation, loss of appetite and sleep disturbance (Radloff, 1977, p. 386). The person filling out the self-report scale answers a set of statements using a four-point Likert scale depending on how often they have felt a certain way during the past week. The total score ranges from 0 to 60 points, 0 meaning no symptoms and 60 meaning high levels of depressive symptoms. Cronbach’s alphas for the total symptom scale in the current study varied between 0.88 and 0.92 across waves.
Demographic Variables
The age and gender of the participants were registered at the screening interviews.
Procedure
The data in the study were collected using self-report questionnaires measured at different time points (Idsoe et al., 2019). The first data collection was a pretest of baseline measures (T1) including the variables used in this study, as well as several other measures. T1 was administered at the end of the screening interviews using pen and paper. As the main outcome measure, depression was measured at a second pretest (T2) two weeks prior to the trial. After the intervention period ended, participant received a post-intervention questionnaire (T3) measuring aggression and depression (as well as the other study measures). Two follow-up questionnaires were collected respectively 6 (T4) and 12-months (T5) after the intervention ended. In this study we used data from pretest, posttest and 6-month follow-up. All measures except T1 were conducted using electronic questionnaires.
Ethical Considerations
The study was approved by The National Regional Committee for Medical and Health Research Ethics (REK) prior to the investigation and was performed in accordance with the Declaration of Helsinki (Idsoe & Keles, 2016).
Information about the project was given to the participants at the initial screening interview. This included information about the background and purpose of the study, the study procedure itself and how the data would be used. The participants were informed about both conditions, and that they would be randomly assigned to either the ACDC-course or usual care. All the adolescents gave informed consent. Finally, the participants were informed that they could withdraw this consent at any time and without any given reason, as the study was completely voluntary.
Statistical Analyses
Analyses in the study were based on the principle of "intention to treat" (ITT) in line with recommended guidelines for analysis and reporting of randomized control studies (Moher et al., 2010). In ITT analysis, the data are analysed in accordance with the group they were originally assigned. This approach preserves the principle of randomization, allowing us to compare at group level (McCoy, 2017). Power analysis for the trial was calculated up front of the study and reported in our original trial article (Idsoe et al., 2019).
The statistical program SPSS was used for descriptive statistics, correlation analysis and reliability analysis. Mplus 8 was used for the structural equation modelling (Muthén & Muthén, 2017). Auto-regressive models were used to examine the direct and indirect effect of the intervention at posttest and 6-month follow-up outcomes. Gender and age were used as covariates in the analyses; however, age did not have any association with the other variables and was therefore dropped from further analysis.
To accommodate for missing data and non-normal item distributions in Mplus, the robust maximum likelihood (MLR) was preferred. In accordance with the ITT-analysis approach, we employed a procedure called full information maximum likelihood (FIML) that uses all available data points. Missingness in the data was carefully investigated by inspecting the answers from the attrition group regarding their reason for dropping out, controlling for auxiliary information and investigating missing data patterns. As a result of this, attrition was considered missing at random (MAR), and variables associated with missingness at pretest were entered to the models as either control or auxiliary variables. Potential cluster effects were accounted for in the analysis by examining intraclass correlations (ICCs), as recommended by CONSORT (Moher et al., 2010). Clustering must be considered if the design effect is greater than 2. Since this was not the case for any of the design effects in the data, the results of the analysis were reported without correction for clustering. The following fit indices were consulted: the comparative fit index (CFI), the root-mean-square error of approximation (RMSEA) and the standardized root-mean-square residual (SRMR) in addition to the chi-square statistics. CFI/TLI ≥ 0.95, RMSEA ≤ 0.05, and SRMR ≤ 0.06 is considered a well-fitting model and CFI/TLI ≥ 0.90, RMSEA ≤ 0.08, and SRMR ≤ 0.10 is considered an adequately fitting model by West et.al (2012).
Results
Descriptive Statistics
Table 1 shows the correlation among the study variables as well as descriptive statistics displaying both mean (M) and standard deviation (SD) for intervention and control groups. As an answer to the first research question, depression and aggression were significantly positively correlated at all time points, indicating that as levels of depression increase, levels of aggression also tend to increase, or vice versa. Gender was only correlated with depression at T1 (r = 0.21, p = < 0.01), meaning girls scored higher on depression symptoms at pretest. No gender differences were detected for the aggression variable.
Table 1
Descriptive statistics of the measures for intervention and control group and intercorrelations among the variables for the overall sample
1
2
3
4
5
6
7
8
ACDC intervention (N = 133)
UC control (N = 95)
M
SD
M
SD
1
Depression T1- pretest
(0.88)
0.65**
0.43**
0.28**
0.23**
0.15
0.21**
− 0.07
33.08
9.97
32.01
9.75
2
Depression T2- 2nd pretest
(0.88)
0.47**
0.21**
0.24**
0.17
0.13
− 0.16**
32.77
8.80
30.28
10.67
3
Depression T3- posttest
(0.92)
0.30**
0.35**
0.23*
0.07
0.00
26.85
11.82
29.55
10.77
4
Agression T1- pretest
(0.83)
0.70**
0.64**
0.05
0.03
2.27
0.76
2.28
0.73
5
Agression T3- posttest
(0.83)
0.69**
− 0.07
− 0.03
2.25
0.83
2.23
0.73
6
Agression T4- follow-up
(0.85)
0.4
− 0.02
2.18
0.81
2.19
7.6
7
Gender
(–)
− 0.16*
–
–
–
–
8
Age
(–)
16.55
1.10
16.92
1.16
*p < 0.05, **p < 0.01. Gender: Males are coded as 1 and females as 2. Cronbach’s Alpha presented in the parentheses. Notes. Ranges and anchors: Depressive symptoms (0 = No symptoms, 60 = High level of and frequent symptoms); Aggression symptoms (1 = strongly disagree, 4 = Strongly agree)
The sample mean age for the intervention group was 16.55 years and 16.92 for the control group. There was an overrepresentation of girls in the sample. Cronbach's alpha was > 0.80 for all the variables, indicating good internal consistency.
Structural Equation Modelling
An auto-regressive model was conducted to test whether the intervention had a direct or indirect effect on aggressive symptoms at posttest and 6-month follow-up. The model of research question 2 is illustrated in Fig. 2. Aggression posttest (T3) was regressed on the intervention variable, aggression pretest (T1) and gender, and aggression follow-up (T4) was regressed on aggression posttest (T3).
Fig. 2
First Model. Gender = Males coded as 1 and females coded as 2. Intervention: UC coded as 0 and ACDC coded as 1. Standardized parameter estimates were reported. ***p < .001
×
Although the chi-square was significant (p = 0.041), this model had an adequate fit according to RMSEA = 0.076, 90% CI [0.014, 0.134]; CFI = 0.961; TLI = 0.931; SRMR = 0.047. The results showed that the intervention has neither a direct effect on aggression symptoms at posttest (β = −0.003, p = 0.955) nor indirect effect at 6-month follow up (through posttest, indirect (β = 0.002, p = 0.955). This means that there is no significant difference between the ACDC-group and the UC-group regarding reduction in aggression symptoms.
The second auto-regressive model (Fig. 3) was conducted to test for the indirect effect of the intervention on aggression at 6-month follow-up via reducing depression at posttest. In addition to controlling for gender and baseline measures of aggression and depression, depression posttest (T3) was regressed on the intervention variable and aggression follow-up (T4) was regressed on depression posttest (T3).
Fig. 3
Second model. Gender = Males coded as 1 and females coded as 2. Intervention groups = UC was coded as 0 and ACDC as 1. Standardized parameter estimates are reported. *p < .05 ***p < .001
×
This model had a good fit, with non-significant chi-square (p = 0.137); RMSEA = 0.044, 90% CI [0.000, 0.087]; CFI = 0.980; TLI = 0.966; SRMR = 0.059. The model shows no significant indirect effect on aggression follow-up by reducing depression at posttest. That is, that the intervention had no indirect effect on reducing aggression symptoms by reducing depression at posttest (indirect β = 0.013, p = 0.316).
Discussion
The aim of this study was to explore the association between aggression and depression and whether the ACDC- intervention that has been found effective in reducing depression, also reduced aggression symptoms either directly, indirectly, or both. The study was based on the following three research questions: 1) What is the association between aggression and depression? 2) Does the intervention predict aggression directly at posttest, and indirectly at follow-up? 3) Does the intervention predict aggression at follow-up indirectly via reducing depression at posttest?
Results showed that aggression and depression were correlated at all time points. The correlations varied between 0.28 and 0.35 (significant at p < 0.01) indicating low to moderate correlation between aggression and depression This is in line with the previous studies on the association between aggression and depression (Dugré et al., 2020; Dutton & Karakanta, 2013; Gándara et al., 2021; Gordillo et al., 2022). However, this association is well documented for the reactive subtype of aggression, but not proactive aggression (del Barrio Gándara et al., 2021; Card & Little, 2006; Fite et al., 2012). These results were therefore expected, as the ACDC study only assessed reactive aggression.
In the first auto-regressive model we tested whether the intervention could predict aggression directly at posttest and indirectly at follow-up. Results showed that the intervention had no significant direct effect on aggression at posttest or indirect effect at follow-up through aggression at the previous time point. The same results were found in the second model. The intervention did not predict reduction in aggression symptoms indirectly, by reducing depression at posttest. These findings show that even though CBT has been proven effective at reducing both aggression and depression symptoms in adolescents, this intervention does not have the same effect on aggression as it does on depression. As previous studies have found great support for CBT being an effective treatment for aggression, this could mean that the ACDC intervention is too specifically targeted towards depression symptoms so that it might not target aggression symptoms. CBT interventions for aggression usually focus on the social cognitive deficits that lead to aggressive behaviour, as described in Crick and Dodge's social information processing theory (Dodge & Crick, 1990). This includes hostile attribution bias, biased interpretation of social cues and deficits in social problem-solving skills (Dodge, 1993; Dodge & Coie, 1987; Sukhodolsky et al., 2000). Even though the ACDC-program aims at teaching adolescents appropriate techniques to modify their thoughts, feelings and behaviour, the program does not directly focus on the sustaining factors of aggression as described above. This might explain why the ACDC-intervention does not reduce aggression symptoms in depressed adolescents. Despite evidence supporting the efficacy of CBT on aggression symptoms, few studies have examined whether reactive and proactive aggression moderates the outcome of the treatment (Smeets et al., 2015). As reactive, not proactive, aggression is linked to depression, more knowledge about CBT for the subtypes of aggression is needed to successfully interpret this as a part of depression programs like ACDC.
There are several limitations to this study that should be addressed, as they could influence the interpretation of the results. First is the sample size and overrepresentation of girls, causing limited statistical power. The gender difference in the sample might be a result of boys being less likely to admit that they have problems with their mental health and therefore not volunteering to the ACDC intervention (Idsoe et al., 2019). Even though depression is more frequent in girls than boys during adolescence, the skewness in the gender distribution is too big to be able to draw any conclusions on the efficacy of the intervention for boys. Studies also show that depressed men experience more symptoms of aggression than depressed women (Martin et al., 2013). There is a chance that the reported level of aggression is too low and reduction in aggression symptoms have not been successfully identified, due to the low percentage of boys participating in the current study,
The ACDC-data are based on self-reported symptoms by the participants themselves, increasing the risk of both under- and over-reported measures. As we know from previous studies, measures of depression based on self-reported questionnaires are a lot higher compared to measures on depression based on diagnostic interviews (Kessler et al., 2001). The study was conducted in a highly natural setting with course leaders being responsible for implementing the intervention, and not the researchers themselves. As a result of this limited control, there is always a risk of the intervention being inadequately applied which could influence the effectiveness of the intervention (Midgley et al., 2013). However, all the course leaders had a minimum of 3-years education and received a detailed 5-day certification course prior to the intervention, minimizing the risk of poor interpretation.
Lastly, in longitudinal studies there is always a risk of participants dropping out before the end of the data collection. This kind of attrition to the sample might have a negative effect on both internal and external validity, as it distorts the accuracy of the data (Barry, 2005). As a result of attrition, there was a loss of statistical power in the ACDC intervention. However, this attrition was considered missing at random (MAR) and was and handled with a full information maximum likelihood (FIML) procedure.
Implications for Practice
This study supports previous findings of a positive association between the two difficulties, aggression and depression (del Barrio Gándara et al., 2021; Dugré et al., 2020; Dutton & Karakanta, 2013). However, the intervention group did not predict any significant direct or indirect reduction in aggression symptoms at posttest or follow-up. More knowledge about CBT interventions for aggression is needed to understand how programs like the ACDC can be adapted to reduce a broader range of symptoms in addition to depression. For future research on the ACDC-intervention, a larger sample with a more even gender balance is needed to be able to generalize the effect of the intervention. To achieve this, difficulties related to recruitment and attrition of participants needs to be solved (Idsoe et al., 2019).
Declarations
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Met BSL Psychologie Totaal blijf je als professional steeds op de hoogte van de nieuwste ontwikkelingen binnen jouw vak. Met het online abonnement heb je toegang tot een groot aantal boeken, protocollen, vaktijdschriften en e-learnings op het gebied van psychologie en psychiatrie. Zo kun je op je gemak en wanneer het jou het beste uitkomt verdiepen in jouw vakgebied.
The sources for the description of the intervention and RCT is retrieved from the following articles: Idsoe et al., 2019; Idsoe & Keles, 2016; Keles & Idsoe, 2021).
