Within Intervention Change: Anger Regulation and Hostile Intent Attribution as Mechanisms to Reduce Children’s Aggressive Behavior

doi:10.21203/rs.3.rs-2198271/v1

Download PDF

Research Article

Within Intervention Change: Anger Regulation and Hostile Intent Attribution as Mechanisms to Reduce Children’s Aggressive Behavior

https://doi.org/10.21203/rs.3.rs-2198271/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Intervention programs can effectively reduce children’s aggressive behavior problems. However, surprisingly few studies have investigated through what underlying mechanisms these interventions exert their beneficial effects. This requires high-frequency measurements of change mechanisms throughout the course of treatment. We used such a design to examine adaptive anger regulation and hostile intent attribution as mechanisms of change in an intervention to reduce children’s aggressive behavior problems. In total, 76 boys with aggressive behavior problems (M_age = 10.58, SD = 1.52; 96.1% born in Netherlands) participated in an 11-session cognitive behavior therapy intervention. During each treatment session, we used brief 3-item measures to assess children’s self-reported anger regulation, hostile intent attribution and aggression, as well as parent-reported child aggression. We analyzed these high-frequency measurements using latent growth curve models, providing an empirical test of individual-level change mechanisms. Results showed that individual change in child-reported aggression was associated with individual change in adaptive anger regulation and hostile intent attribution. For parent-reported aggression, such associations were found only for hostile intent attribution. These findings provide support for anger regulation and hostile intent attribution as within intervention change mechanisms, supporting the idea that these mechanisms are effective intervention targets to reduce children’s aggressive behavior problems.

within intervention change

aggressive behavior problems

mechanisms of change

children

Aggressive behavior problems in childhood have a negative impact on children and their environment (Girard et al., 2019). When left untreated, the prognosis of children with aggressive behavior problems is poor, with increased risk for delinquency, substance abuse, and poor school adjustment, resulting in high costs to society (Evans et al., 2021; Loeber & Farrington, 2000). To prevent escalation of aggressive behavior problems, interventions are needed to treat these problems effectively as they arise in childhood (Weisz & Kazdin, 2017). Several intervention programs for children with aggressive behavior problems have yielded promising effects (Weisz & Kazdin, 2017). However, surprisingly few studies have investigated through what underlying mechanisms these interventions exert their beneficial effects (Kazdin, 2009). The aim of the current study was to investigate two possible mechanisms of change in an intervention to reduce children’s aggressive behavior problems: anger regulation and hostile intent attribution.

Identifying mechanisms of change is important for several reasons. First, identifying and understanding the mechanisms of change in therapy can improve our understanding of aggressive behavior problems in children (Kazdin & Nock, 2003). Second, current interventions can be optimized by ensuring that they target the relevant underlying mechanisms and do not contain unnecessary components (Kraemer et al., 2002). Third, identifying mechanisms of change can facilitate parsimony across the variety of interventions for different problems, by focusing on important mechanisms of change that cut across diverse problems (Kazdin & Nock, 2003). Fourth, mechanisms of change can be used as ‘tailoring variables’ by providing clinicians with the flexibility to adapt treatments to variations in the mechanisms underlying children’s problems (Collins et al., 2004; Kazdin & Nock, 2003).

Interventions to reduce children’s aggressive behavior problems have frequently targeted anger regulation and hostile intent attribution as possible mechanisms of change (for a review, see: Chorpita & Daleiden, 2009). Both mechanisms are part of the social information processing (SIP) model (Crick & Dodge, 1994; Lemerise & Arsenio, 2000), which assumes that biases or deficiencies in a series of ordered processing steps contribute to aggressive behavior. Indeed, empirical research has shown that children with aggressive behavior problems often have difficulties with anger regulation (for reviews see: Roberton et al., 2012; Röll et al., 2012) and display a stronger tendency to interpret ambiguously intended social behavior as stemming from hostile intent (for a review, see: Verhoef et al., 2019). From meta-analytical research, we know that targeting anger regulation and problem solving in treatments is associated with larger reductions in aggression (Landenberger & Lipsey, 2005; Sukhodolsky et al., 2016). However, to test anger regulation and hostile intent attribution as mechanisms of change, intervention studies are required that directly assess these mechanisms and investigate how treatment change is produced (Greca et al., 2009). Of the few studies examining mechanisms of change, one intervention study has found indications that children’s hostile attributions assessed at post-treatment may have functioned as a mediator between the intervention effect and children’s aggressive behavior at follow-up (although the mediation effect was not statistically significant; Lochman & Wells, 2002). Another intervention study with four measurements has shown that emotion regulation skills partially mediated reductions in child aggression (Burke & Loeber, 2016). Thus, there are some indications that hostile intent attribution and anger regulation may mediate intervention effects on aggression.

An important next step in mapping anger regulation and hostile intent attribution as mechanisms of change, is to use high-frequency measurements of these variables during an intervention study (Kazdin, 2007; Laurenceau et al., 2007). This has three advantages. First, repeated measurements allow us to not only capture group-level change in mechanisms and outcomes but also examine change and variability at the individual-level (Laurenceau et al., 2007). Investigating individual-level change is needed as this indicates whether changes in a mechanism within a specific child are related to changes in treatment outcomes for that child. This has not been possible with traditional mediation analyses using group-level analyses (e.g., regression or repeated measures analyses) over a few measurements, as those analyses can only examine whether differences between children on the mechanism predict differences between children on the outcomes. Second, high-frequency measurements enhance the chance of identifying a mechanism of change (Kazdin, 2007). For example, when the trajectory of a mechanism emerges as cubic or S-shaped and only few measurements are taken, conclusions are highly dependent on the moment of the measurements and cannot identify the shape of the trajectory (Laurenceau et al., 2007). Including multiple measurements decreases this risk and enables researchers to obtain an accurate and informative picture of how change unfolds over time (Laurenceau et al., 2007; Moldovan & Pintea, 2015). Third, high-frequency measurements better correspond with the theorized change processes for anger regulation and hostile intent attribution. That is, changes in children’s anger regulation and hostile intent attribution are expected to directly affect their subsequent (aggressive) response, rather than changing children’s aggressive behavior in a process that unfolds over months between measurement occasions (De Castro et al., 2015).

To examine mechanisms of change, timing and spacing of the repeated measurements is essential, and should be based upon theory and prior work of the expected mechanisms (Laurenceau et al., 2007). According to the SIP model, anger regulation and hostile intent attribution are situation-dependent fast mental processes, and may therefore likely be susceptible to rapid change. An intervention with weekly sessions should, in theory, be able to produce weekly changes in these mechanisms. Indeed, a recent study has revealed that weekly changes in adaptive anger regulation and hostile intent attribution covaried with weekly changes in children’s aggression (Alsem, Keulen, et al., 2022). The present study builds on these findings by examining, within an intervention study, whether causing changes in these mechanisms would go together with reductions in children’s aggression.

The current study examined anger regulation and hostile intent attribution as mechanisms of change to reduce children’s aggressive behavior problems. Our first aim was to examine whether adaptive anger regulation increased, whereas hostile intent attribution and child- and parent-reported aggression decreased over the intervention weeks. Our second aim was to investigate whether changes in anger regulation and hostile intent attribution were associated with changes in aggression over the weeks. We expected parallel change, given the theorized immediate effects of children’s anger regulation and interpretation on their subsequent (aggressive) behavior (De Castro et al., 2015).

We examined our research questions within the intervention YourSkills. In this manualized cognitive behavior therapy (CBT), children practice anger regulation and intent attribution skills during social interactions that are either simulated in virtual reality or roleplayed by a therapist (Alsem et al., 2021). In a previous study, we found that both the virtual reality and roleplay versions of YourSkills were more likely to reduce children’s aggressive behavior than care-as-usual for six out of seven aggression outcomes, with some indications that virtual reality outperformed roleplay (Alsem, van Dijk, et al., 2022). We assessed children’s self-reported anger regulation, hostile intent attribution, and aggression at each weekly treatment session. Parents also provided weekly reports of children’s aggression. These high-frequency measurements enabled us to provide an empirical test of individual-level change mechanisms.

Participants

Participants were 76 children, 8 to 13 years of age (M_age = 10.58, SD = 1.52), who participated in a randomized controlled trial. This trial investigated the effectiveness of YourSkills, a newly developed CBT (Alsem, van Dijk, et al., 2022). Children were recruited at fifteen clinical centers in the Netherlands providing mental health care for children who have severe problems that impair daily functioning and require treatment. Recruitment began in September 2019 and the study ended in July 2021. Therapists working at the clinical centers were asked to approach parents of boys whose casefiles met the inclusion criteria: age 8–13 years, aggressive behavior problems, estimated intelligence level above 80, no severe autism spectrum disorder, and no epilepsy or severe visual or auditory limitations.

The randomized controlled trial included 115 children, randomized to three groups: YourSkills virtual reality, YourSkills roleplay, or care-as-usual (for the flowchart see Alsem, van Dijk, et al., 2022). In this study, we only included children who were randomized to the two YourSkills groups, because our aim was to examine the mechanisms of change of interventions for children’s aggression. We excluded children in the care-as-usual group as we had no information about what mechanisms were targeted in the care-as-usual treatments. In addition, we only included children who participated in at least five of the intervention sessions, because children should have had the opportunity to learn the expected mechanisms of change.¹ Of the included children, 36 participated in YourSkills virtual reality and 40 in YourSkills roleplay. Most children (96.1%) were born in the Netherlands and in most families, both biological parents were born in the Netherlands as well (73.7%). In 14.5% of the families only one parent was born in the Netherlands and in 11.8% of the families both parents were born elsewhere. Parents attained middle levels of education (44.7%; ISCED 3–4), high education (39.5%; ISCED 5–8), or low education (15.8%; ISCED 0–2; UNESCO, 2012).

We obtained written informed consent from parents and 12- and 13-year-old children. Participation was voluntary and children and parents were assured of confidential use of their data. Children received a small gift after filling out the last assessment (e.g., a multicolor pen). This study was approved by the Ethics Committee of the University Medical Center Utrecht (NL67139.041.18) and the trial was registered in the Dutch Trial Register (NTR; https://www.trialregister.nl/trial/7959).

Procedure

The current study consisted of eleven weekly assessments scheduled at each of the eleven YourSkills treatment sessions; one introduction session with parents and ten sessions with children. Therapists scheduled all sessions and invited researchers to conduct the first assessment during the introduction session (i.e., our baseline assessment). This assessment took place in another room with the first author or a trained research assistant. Children first completed a short task and then filled out the weekly assessments used in this study, as well as some other questionnaires (20–30 minutes in total). Directly after the session ended, the researcher asked parents to fill out questionnaires on a laptop. The ten other assessments were integrated within the treatment sessions with the child: all sessions started with children filling out the brief measures of anger regulation, hostile intent attribution, and aggression in a paper booklet (5 minutes). Parents or caregivers who brought the child to the treatment session were asked to fill out a brief measure of aggression in the clinical center’s waiting room.

Yourskills Treatment

YourSkills is a manualized CBT, developed based on evidence-based treatments for children with aggressive behavior problems. The effectiveness of YourSkills is examined in a previous study (Alsem, van Dijk, et al., 2022). Results of this study showed that both the virtual reality and roleplay versions of YourSkills were more likely to reduce aggressive behavior than care-as-usual for six out of seven aggression outcomes, suggesting that YourSkills successfully reduced children’s aggressive behavior. When we directly compared virtual reality to roleplays, results favored virtual reality above roleplays on four of seven aggression measures (Alsem, van Dijk, et al., 2022).

The aim of YourSkills is to reduce children’s aggressive behavior problems by enhancing emotion regulation and social information processing skills. Children practice anger recognition, anger regulation, nonhostile intent attribution, and other social problem solving skills in social interactions (for an overview of the treatment sessions, see: Alsem et al., 2021). To enable children to practice their skills or new cognitions whilst being emotionally engaged, therapists create challenging social situations for children in virtual reality or roleplays. In each session, therapists first explain a new skill, then model the skill using roleplay, and then—depending on the YourSkills version—use virtual reality or roleplays to let children practice the skill in anger-provoking social situations, such as being disadvantaged, having authority conflicts or being rejected by a peer. The details of the virtual reality and roleplay version of YourSkills are described in detail elsewhere (Alsem, van Dijk, et al., 2022).

Measures

The psychometric quality of the weekly measures used in the current study were investigated in an earlier study. Results showed adequate internal consistency, convergent, and concurrent validity for the child-report version (Alsem, Keulen, et al., 2022).

Aggression

To assess children’s aggression over the weeks, we asked both parents and children to rate three items (e.g., “This week my child/I fought with someone”) on a scale from 1 (never) to 5 (very often). Ratings were averaged across items. The internal consistency in the current study was adequate over the weeks for both children (α_mean = .79, α_range = .75-.87) and parents (α_mean = .79, α_range = .69-.83).

Adaptive Anger Regulation

We assessed weekly adaptive anger regulation by asking children to rate three items (e.g., ‘This week I managed to do something against my anger’) on a scale from 1 (never) to 5 (very often). The internal consistency in the current study was adequate over the weeks (α_mean = .64, α_range = .58-.71).

Hostile Intent Attribution

To assess children’s weekly hostile intent attribution, we asked children to rate three items (e.g., ‘This week people were mean to me’). Children rated the items on a scale from 1 (never) to 5 (very often). The internal consistency in the current study was adequate over the weeks (α_mean = .71, α_range = .52-.83).

Analyses

We ran all analyses in Mplus 8.4. Before we tested our hypotheses, we explored whether there was variability in children’s individual scores on mechanisms and outcomes over the weeks. We ran four random intercept models (i.e., for anger regulation, hostile intent attribution, and child- and parent-reported aggression) to estimate the amount of variance at the within- and between-person level for each variable. We expected substantial individual within-person variation over the weeks, reflecting intervention effects on children’s adaptive anger regulation, hostile intent attribution, and aggression. We did not expect much variation at the between-person level because of our homogenous sample (i.e., participants all had aggressive behavior problems).

To test our first research question, whether children’s adaptive anger regulation increased and their hostile intent attribution and aggressive behavior decreased across the course of the intervention, we estimated four univariate latent growth models (LGMs) using an intercept factor and a linear slope factor. We expected that slope factors would significantly increase for adaptive anger regulation and decrease for hostile intent attribution and child- and parent-reported aggression. We modelled linear slopes because we expected change over the whole treatment period, and this approach is more parsimonious, preventing power problems in our relatively small sample.

Next, we investigated our second research question, whether individual-level increases in adaptive anger regulation and decreases in hostile intent attribution were associated with decreases in child- and parent-reported aggression. We estimated four bivariate LGMs, each estimating the slope factors of two variables at a time: 1) one of the two mechanisms and 2) children’s aggressive behavior, either child- or parent-reported. We expected significant correlations in between slope factors in each of the four models. As LGMs estimate individual-level change trajectories, these models are suitable to examine mechanisms of change (Laurenceau et al., 2007).

Last, as a control analysis, we checked whether children’s change trajectories differed between the virtual reality and the roleplay group, using Wald tests. If slopes differed between these two intervention groups, we corrected for this difference by adding intervention group as predictor to the slope factors. Results are presented in Appendix A.

For all LGM models, we examined the model fit using the Chi-square test statistic (χ²), the Comparative Fit Index (CFI), and the Root Mean Square Error of Approximation (RMSEA). However, we did not necessarily expect a good model fit because we had many parameters in the model (i.e., eleven weeks of assessment) and a relatively small sample size. We hoped to obtain adequate model fit: CFI above .90, RMSEA below .10 and the ratio between the χ ² test statistic and the degrees of freedom below 3 (this was not based on the χ² significance value to avoid problems with sample size; Schermelleh-Engel et al., 2003).

We inspected missingness in our measurements over the weeks. Most children completed questionnaires in all eleven weeks (78.9%) and all children completed questionnaires in at least six weeks. We compared children who completed all weeks (n = 60) with children who missed at least one week (n = 16) and found no significant differences in levels of adaptive anger regulation, hostile intent attribution, and aggression. To check for missing data patterns on the item-level across assessments, we conducted Little’s test. This test yielded a normed χ² (χ²/df) of 0.10, indicating that data was missing completely at random (Bollen, 1989). We therefore used default settings for multilevel data in Mplus to estimate missing data (i.e., maximum likelihood; Muthén & Muthén, 2007).

¹ We conducted sensitivity analyses including the five children that participated in less than half of the treatment sessions and results revealed similar conclusions.

Preliminary Analysis: Individual Variability in Change Mechanisms and Aggression

Our models revealed that there was significant variance at both the within- and between-level in adaptive anger regulation (57.9% within; 42.1% between), hostile intent attribution (58.0% within, 42.0% between), child-reported aggression (47.4% within; 52.6% between), and parent-reported aggression (51.9% within, 48.1% between). The within-person variances were substantial, and indicated, as expected, that children showed variability in their levels of adaptive anger regulation, hostile intent attribution, and aggression over the intervention weeks.

Research Question 1: Individual-level Change

Next, we examined whether we found the expected individual-level changes in adaptive anger regulation, hostile intent attribution, and child-and parent-reported aggression over the intervention weeks. Fit statistics of the LGMs are reported in Table 1. Not all fit statistics supported an adequate fit of our models. Although the chi-square ratio was adequate, CFI and RMSEA were not adequate for three out of four models. However, none of the modification indices improved the model and explained variances were substantial in all models (R² range .27 to .74). We continued with these models, because they provide a good test of our research questions and the lower fit is likely caused by the high amount of estimated parameters and small sample size (Stull, 2008).

As expected, results showed a significant linear increase in adaptive anger regulation, B = 0.04, SE = 0.01, β = .57, p < .001, and a significant linear decrease in hostile intent attribution over the intervention weeks, B = -0.02, SE = 0.01, β = − .43, p = .021. We also found the expected significant linear decreases in child-reported aggression, B = -0.03, SE = 0.01, β = − .51, p = .01, and parent-reported aggression over the intervention weeks, B = -0.05, SE = 0.01, β = − .83, p < .001. Descriptively, we inspected the percentage of children who changed in the expected direction (i.e., a slope factor above or below zero). We found that 72.4% of children increased in anger regulation, 64.9% decreased in hostile intent attribution, 71.2% decreased in self-reported aggression, and 79.0% decreased in parent-reported aggression. Thus, on average and for most children in the YourSkills intervention group, both mechanisms and outcomes changed in the expected direction.

Table 1

*Model Fit Statistics of the Individual-Level Change Models and Parallel Change Models for both Child- and Parent-Reported Aggression*
	χ² / df	CFI	RMSEA
Within intervention change
Adaptive anger regulation	1.78	.88	.10
Hostile intent attribution	1.79	.87	.10
Child-reported aggression	1.97	.88	.11
Parent-reported aggression	1.45	.93	.08
Parallel change child-reported aggression
Adaptive anger regulation	1.66	.84	.09
Hostile intent attribution	2.28	.72	.13
Parallel change parent-reported aggression
Adaptive anger regulation	1.58	.83	.09
Hostile intent attribution	1.68	.81	.10

Research Question 2: Parallel Change In Mechanisms And Outcomes

For child-reported aggression, results supported both anger regulation and hostile intent attribution as mechanisms of change. The bivariate LGM revealed that the slopes of adaptive anger regulation and child-reported aggression were strongly and significantly correlated, r = − .51, p = .018, indicating that children’s increases in adaptive anger regulation over the weeks were related to decreases in their aggressive behavior. Similarly, the slopes of hostile intent attribution and child-reported aggression were strongly significantly correlated, r = .73, p = .002. Figure 1 illustrates these parallel change trajectories, presenting the average change trajectory of each variable over the intervention weeks.

For parent-reported aggression, results supported only hostile intent attribution as mechanism of change. Our bivariate LGMs revealed that the slope of hostile intent attribution was strongly significantly correlated with the slope of parent-reported aggression, r = .62, p = .006, but not with the slope of adaptive anger regulation, r = − .20, p = .357. Thus, decreases in parent-reported aggression were related to decreases in children’s hostile intent attribution over the weeks, but not to increases in adaptive anger regulation.

The present study extends prior research on child interventions to reduce aggressive behavior problems, by investigating anger regulation and hostile intent attribution as mechanisms of change. We examined how interventions exert their beneficial effects, by providing an empirical test of individual-level change mechanisms. Therefore, we used high-frequency measurements of anger regulation, hostile intent attribution, and aggression throughout the course of treatment allowing us to not only capture group-level change in mechanisms and outcomes but also examine change and variability at the individual-level (Kazdin, 2007; Laurenceau et al., 2007). When modeling individual change across the eleven sessions of treatment, we found that children’s adaptive anger regulation increased and hostile intent attribution decreased over the intervention weeks. Importantly, changes in these mechanisms were strongly associated with child-reported aggression. For parent-reported aggression, such associations were found only for hostile intent attribution. These findings provide support for anger regulation and hostile intent attribution as within intervention change mechanisms.

Our high-frequency measurements allowed us to also investigate individual variation over the treatment course. We found that within-child variances were substantial, highlighting the necessity of an individual-child approach to capture variation in children over the weeks. The high within-child variances were in line with the social information processing model (Crick & Dodge, 1994; Lemerise & Arsenio, 2000), suggesting that anger regulation and hostile intent attribution are indeed situation-dependent fast mental processes, being susceptible to rapid changes. The relatively low between-children variances were not surprising, given our homogenous sample of boys with aggressive behavior problems. When we examined for how many children the intervention mechanisms and outcomes changed, we found that the majority of children changed in the expected directions (64.9–79.0% across mechanisms and outcomes). As not all children did benefit from targeting anger regulation and hostile intent attribution, future research might investigate whether other change mechanisms need to be targeted in CBT for these children, or that more intensive treatment programs are necessary (e.g., CBT combined with parent training). Our study showed that it is possible to identify the children who do change and who do not change or deteriorate during the treatment course. This could help clinicians to decide whether they need to adapt their treatment strategy.

Unexpectedly, we found no association between individual changes in child-reported anger regulation and parent-reported aggression. This could be due to a possible informant effect as the correspondence between the reportages of parents and children is found to be relatively low (De Los Reyes et al., 2015). Alternatively, it could be that some of the children that did not improve in anger regulation over the treatment course (27.6% of children remained stable or decreased in anger regulation), did reduce in parent-reported aggression. For this group it could be that only hostile intent attribution was a working mechanisms of change. Future research is needed to investigate the association between anger regulation as mechanisms of change for parent-reported aggression.

Our study had its limitations. First, we examined parallel change and can therefore not conclude that the mechanisms caused the decreases in aggression. To investigate causality, it is necessary to experimentally manipulate the mechanisms, for example by comparing children in an intervention group with children in a control group for whom the mechanisms were not targeted. Although our randomized controlled trial did include a control group, we could not use it for the current study because we were unable to conduct the weekly measures due to practical limitations in routine care. Since our intervention targeted anger regulation and hostile intent attribution, and we found in the trial that children in the intervention groups were more likely to reduce in aggression than children in the control group (Alsem, van Dijk, et al., 2022), it seems plausible that these mechanisms may have a causal effect. Second, our findings cannot be generalized to girls. We included only boys because the intervention was initially developed for this group, considering that girls’ development, forms, and outcomes of aggression may be different from boys (Underwood, 2002). Future research could examine whether girls with aggressive behavior problems benefit equally from changes in anger regulation and hostile intent attribution. Third, our sample size was too small to allow for the examination of more complex statistical models, such as parallel change quadratic growth models. Future research might also examine quadratic slopes and by doing so identify the shape of the growth trajectories (Laurenceau et al., 2007).

A strength of our study was the high-frequency of the measurements that enabled us to examine individual-level change in children’s mechanisms and outcomes (Laurenceau et al., 2007). We conducted our intervention study in a naturalistic setting by including multiple clinical centers, recruiting children in routine care, and examine effects under real-world conditions, promoting external validity. Using brief weekly measures in clinical practice is feasible, and these 3-items measures were well-validated in a previous study (Alsem, Keulen, et al., 2022). Last, we used a multi-informant approach to assess aggressive behavior problems, as the correspondence between informants in aggression reports is relatively low (De Los Reyes et al., 2015). This way, we could investigate whether child-reported changes in anger regulation and hostile intent attribution were also related to parent-reported changes in aggression.

The identification of hostile intent attribution and anger regulation as mechanisms of change has clinical implications. First, in clinical practice, it is important to know whether and how treatment elements work. These elements can be used by clinicians deciding which mechanisms to target when deficits in these mechanisms are expected to underlie children’s aggressive behavior problems (Collins et al., 2004; Kazdin & Nock, 2003). Our present findings suggest that anger regulation and hostile intent attribution provide good starting points for the majority of children, and frequent measurements may identified the few children for whom standard CBT does not influence these mechanisms, or does not change aggressive behavior. Second, existing interventions can be optimized by focusing on anger regulation and hostile intent attribution, while changing or omitting elements with smaller effects (Kraemer et al., 2002). Third, our study with high-frequency measurements highlights the clinical utility of weekly measures. These measures can be a valuable instrument to easily monitor mechanisms of change and treatment progress over the course of an intervention. Clinicians and clients have been shown to manage to administer these brief measures and use this for productive shared decision making during treatment (Campbell & Hemsley, 2009).

Our findings open directions for future research. First, as we found evidence for anger regulation and hostile intent attribution as mechanisms of change in interventions for children with aggressive behavior problems, it may be interesting to examine these mechanisms in interventions for other problems as well. Anger regulation and hostile intent attribution contribute transdiagnostically to diverse problems, such as externalizing and internalizing problems, and addressing them may thus facilitate parsimony across interventions (Granic, 2014; Kazdin & Nock, 2003). Second, it might be interesting to examine the influence of other mechanisms of change, such as therapeutic alliance or treatment engagement. There is a need to investigate the potential contribution of therapeutic variables to child treatment outcomes, preferably with mechanisms measured at multiple time points, examining their influence over the course of treatment (Shirk & Karver, 2003).

In conclusion, we have found support for adaptive anger regulation and hostile intent attribution as mechanisms of change for child-reported aggression. For parent-reported aggression, results supported only hostile intent attribution as mechanism of change. These findings provide evidence for the idea that targeting anger regulation and hostile intent attribution in interventions relates to reductions in individual children's aggressive behavior problems. Our study may inspire researchers to conduct intervention studies with high-frequency measurements to track client progress and to investigate through what mechanisms interventions exert their beneficial effects.

Funding. This research was supported by a grant from the Netherlands Organization for Scientific Research to the last author (grant number 453-15-004/511).

Conflict of Interest. The authors declare that they have no conflict of interest.

Acknowledgements. All authors contributed to the study conception and design. Material preparation and data collection were coordinated by Sophie Alsem. Data analyses were performed by Sophie Alsem. The first draft of the manuscript was written by Sophie Alsem and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Data Availability Statement. The raw data and analysis code are available at the Open Science Framework (Alsem, Verhulp, et al., 2022).

Alsem, S. C., Keulen, J., Verhulp, E. E., van Dijk, A., & De Castro, B. O. (2022). Capturing mechanisms of change: Weekly covariation in anger regulation, hostile intent attribution, and children’s aggression. Aggressive Behavior, 48, 232–240. https://doi.org/10.1002/ab.22019
Alsem, S. C., van Dijk, A., Verhulp, E. E., & De Castro, B. O. (2021). Using virtual reality to treat aggressive behavior problems in children: A feasibility study. Clinical Child Psychology and Psychiatry, 26(4), 1062–1075. https://doi.org/10.1177/13591045211026160
Alsem, S. C., van Dijk, A., Verhulp, E. E., Dekkers, T. J., & De Castro, B. O. (2022). Treating children’s aggressive behavior problems using cognitive behavior therapy with virtual reality: A multicenter randomized controlled trial. Manuscript Submitted for Publication.
Bollen, K. A. (1989). A new incremental fit index for general structural equation models. Sociological Methods & Research, 17(3), 303–316. https://doi.org/10.1177/0049124189017003004
Burke, J. D., & Loeber, R. (2016). Mechanisms of behavioral and affective treatment outcomes in a cognitive behavioral intervention for boys. Journal of Abnormal Child Psychology, 44(1), 179–189. https://doi.org/10.1007/s10802-015-9975-0
Campbell, A., & Hemsley, S. (2009). Outcome rating scale and session rating scale in psychological practice: Clinical utility of ultra-brief measures. Clinical Psychologist, 13(1), 1–9. https://doi.org/10.1080/13284200802676391
Chorpita, B. F., & Daleiden, E. L. (2009). Mapping evidence-based treatments for children and adolescents: application of the distillation and matching model to 615 treatments from 322 randomized trials. Journal of Consulting and Clinical Psychology, 77(3), 566–579. https://doi.org/10.1037/a0014565
Collins, L. M., Murphy, S. A., & Bierman, K. L. (2004). A conceptual framework for adaptive preventive interventions. Prevention Science, 5(3), 185–196. https://doi.org/10.1023/B:PREV.0000037641.26017.00
Crick, N. R., & Dodge, K. A. (1994). Review of social information processing in children social adjustment. Psychological Bulletin, 115(1), 74–101. https://doi.org/10.1037/0033-2909.115.1.74
De Castro, B. O., Merk, W., Koops, W., Veerman, J. W., & Bosch, J. D. (2005). Emotions in social information processing and their relations with reactive and proactive aggression in referred aggressive boys. Journal of Clinical Child and Adolescent Psychology, 34(1), 105–116. https://doi.org/10.1207/s15374424jccp3401_10
De Castro, B. O., Thomaes, S., & Reijntjes, A. (2015). Using experimental designs to understand the development of peer relations. Journal of Research on Adolescence, 25(1), 1–13. https://doi.org/10.1111/jora.12103
De Los Reyes, A., Augenstein, T. M., Wang, M., Thomas, S. A., Drabick, D. A. G., Burgers, D. E., & Rabinowitz, J. (2015). The validity of the multi-informant approach to assessing child and adolescent mental health. Psychological Bulletin, 141(4), 858–900. https://doi.org/10.1037/a0038498
Evans, S. C., Dίaz, K. I., Callahan, K. P., Wolock, E. R., & Fite, P. J. (2021). Parallel trajectories of proactive and reactive aggression in middle childhood and their outcomes in early adolescence. Research on Child and Adolescent Psychopathology, 49(2), 211–226. https://doi.org/10.1007/s10802-020-00709-5
Girard, L. C., Tremblay, R. E., Nagin, D., & Côté, S. M. (2019). Development of aggression subtypes from childhood to adolescence: A group-based multi-trajectory modelling perspective. Journal of Abnormal Child Psychology, 47(5), 825–838. https://doi.org/10.1007/s10802-018-0488-5
Granic, I. (2014). The role of anxiety in the development, maintenance, and treatment of childhood aggression. Development and Psychopathology, 26(4pt2), 1515–1530. https://doi.org/10.1017/S0954579414001175
Greca, A. M. L., Silverman, W. K., & Lochman, J. E. (2009). Moving Beyond Efficacy and Effectiveness in Child and Adolescent Intervention Research. Journal of Consulting and Clinical Psychology, 77(3), 373–382. https://doi.org/10.1037/a0015954
Kazdin, A. E. (2007). Mediators and mechanisms of change in psychotherapy research. Annual Review of Clinical Psychology, 3(1), 1–27. https://doi.org/10.1146/annurev.clinpsy.3.022806.091432
Kazdin, A. E. (2009). Understanding how and why psychotherapy leads to change. Psychotherapy Research, 19(4–5), 418–428. https://doi.org/10.1080/10503300802448899
Kazdin, A. E., & Nock, M. K. (2003). Delineating mechanisms of change in child and adolescent therapy: Methodological issues and research recommendations. Journal of Child Psychology and Psychiatry and Allied Disciplines, 44(8), 1116–1129. https://doi.org/10.1111/1469-7610.00195
Kraemer, H. C., Wilson, G. T., Fairburn, C. G., & Agras, W. S. (2002). Mediators and moderators of treatment effects in randomized clinical trials. Archives of General Psychiatry, 59(10), 877–883. https://doi.org/10.1001/archpsyc.59.10.877
Landenberger, N. A., & Lipsey, M. W. (2005). The positive effects of cognitive-behavioral programs for offenders: A meta-analysis of factors associated with effective treatment. Journal of Experimental Criminology, 1(4), 451–476. https://doi.org/10.1007/s11292-005-3541-7
Laurenceau, J. P., Hayes, A. M., & Feldman, G. C. (2007). Some methodological and statistical issues in the study of change processes in psychotherapy. Clinical Psychology Review, 27(6), 682–695. https://doi.org/10.1016/j.cpr.2007.01.007
Lemerise, E. A., & Arsenio, W. F. (2000). An integrated model of emotion processes and cognition in social information processing. Child Development, 71(1), 107–118. https://doi.org/10.1111/1467-8624.00124
Lochman, J. E., & Wells, K. C. (2002). Contextual social-cognitive mediators and child outcome: A test of the theoretical model in the Coping Power program. Development and Psychopathology, 14(4), 945–967. https://doi.org/10.1017/S0954579402004157
Loeber, R., & Farrington, D. P. (2000). Young children who commit crime: Epidemiology, developmental origins, risk factors, early interventions, and policy implications. Development and Psychopathology, 12(4), 737–762. https://doi.org/10.1017/S0954579400004107
Moldovan, R., & Pintea, S. (2015). Mechanisms of change in psychotherapy: Methodological and statistical considerations. Cognition, Brain, Behavior, 19(4), 299–311.
Muthén, L. K., & Muthén, B. (2019). Mplus. Muthén & Muthén.
Roberton, T., Daffern, M., & Bucks, R. S. (2012). Emotion regulation and aggression. Aggression and Violent Behavior, 17(1), 72–82. https://doi.org/10.1016/j.avb.2011.09.006
Röll, J., Koglin, U., & Petermann, F. (2012). Emotion regulation and childhood aggression: Longitudinal associations. Child Psychiatry and Human Development, 43(6), 909–923. https://doi.org/10.1007/s10578-012-0303-4
Schermelleh-Engel, K., Moosbrugger, H., & Müller, H. (2003). Evaluating the fit of structural equation models: Tests of significance and descriptive goodness-of-fit measures. Methods of Psychological Research Online, 8(2), 23–74. https://doi.org/10.1002/0470010940
Shirk, S. R., & Karver, M. (2003). Prediction of treatment outcome from relationship variables in child and adolescent therapy: A meta-analytic review. Journal of Consulting and Clinical Psychology, 71(3), 452–464. https://doi.org/10.1037/0022-006X.71.3.452
Stull, D. E. (2008). Analyzing growth and change: Latent variable growth curve modeling with an application to clinical trials. Quality of Life Research, 17(1), 47–59. https://doi.org/10.1007/s11136-007-9290-5
Sukhodolsky, D. G., Smith, S. D., McCauley, S. A., Ibrahim, K., & Piasecka, J. B. (2016). Behavioral interventions for anger, irritability, and aggression in children and adolescents. Journal of Child and Adolescent Psychopharmacology, 26(1), 58–64. https://doi.org/10.1089/cap.2015.0120
Underwood, M. K. (2002). Sticks and stones and social exclusion: Aggression among girls and boys. In P. K. Smith & C. H. Hart (Eds.), Blackwell handbook of childhood social development (pp. 533–548). Blackwell Publishers Ltd.
UNESCO. (2012). International Standard Classification of Education (ISCED 2011). In Prospects. https://doi.org/10.1007/BF02207511
Verhoef, R. E. J., Alsem, S. C., Verhulp, E. E., & De Castro, B. O. (2019). Hostile intent attribution and aggressive behavior in children revisited: A meta-analysis. Child Development, 90(5), 525–547. https://doi.org/10.1111/cdev.13255
Weisz, J. R., & Kazdin, A. E. (2017). Evidence-based psychotherapies for children and adolescents. Guilford Publications.

Appendix A: Group Differences Check

As a control analysis, we checked for possible differences in change trajectories between children who received the virtual reality versus roleplay version of YourSkills groups. Results showed no differences between the two groups in slopes of adaptive anger regulation, χ²_wald(1) = 0.01, p = .922, hostile intent attribution, χ²_wald(1) = 0.44, p = .510, and child-reported aggression, χ²_wald(1) = 3.82, p = .051. However, we did find group differences for parent-reported aggression, χ²_wald(1) = 5.69, p = .017. Children in the YourSkills virtual reality group showed a linear decrease in parent-reported aggression, B = -.07, SE = 0.01, β = -1.23, p < .001, whereas children in the YourSkills roleplay group remained stable, B = -.03, SE = 0.01, β = -.55, p = .078. We therefore checked whether our results changed if we added group as predictor for the slope factors of the mechanism and outcome variables in the models for parent-reported aggression. Results did not change: the correlation between slopes remained significant for hostile intent attribution and parent-reported aggression, r = .68, p = .006, and nonsignificant for adaptive anger regulation and parent-reported aggression, r = -19, p = .457.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Within Intervention Change: Anger Regulation and Hostile Intent Attribution as Mechanisms to Reduce Children’s Aggressive Behavior

Status:

Version 1

Abstract

Figures

Introduction

Method

Participants

Procedure

Yourskills Treatment

Measures

Aggression

Adaptive Anger Regulation

Hostile Intent Attribution

Analyses

Results

Preliminary Analysis: Individual Variability in Change Mechanisms and Aggression

Research Question 1: Individual-level Change

Research Question 2: Parallel Change In Mechanisms And Outcomes

Discussion

Declarations

References

Appendix

Additional Declarations

Status:

Version 1