Single Study on Collaborative and Proactive Solutions

Vol.:(0123456789)1 3

European Child & Adolescent Psychiatry (2018) 27:1181–1192 https://doi.org/10.1007/s00787-018-1181-5

O R I G I N A L CO N T R I B U T I O N

Child‑based treatment of oppositional defiant disorder: mediating effects on parental depression, anxiety and stress

Josepha Katzmann1  · Manfred Döpfner1,2,3 · Anja Görtz‑Dorten1,2

Received: 11 December 2017 / Accepted: 8 June 2018 / Published online: 13 June 2018 © Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract Previous research has shown that child-oppositional defiant disorder (ODD) and conduct disorders (CD) are associated with parental symptoms of depression, anxiety and/or stress, probably in a bidirectional relationship with mutual influences. It is, therefore, reasonable to assume that in child-centered treatment, a decrease in child-oppositional behavior problems constitutes (at least in part) a mechanism of change for a subsequent reduction in parental psychopathology. The aim of the present study (Clinical trials.gov Identifier: NCT01406067) was to examine whether the reduction in ODD symptoms due to child-based cognitive behavioral treatment (CBT) led to a reduction in parental depression, anxiety and stress. Eighty-one boys (age 6–12 years) with a diagnosis of ODD/CD were randomized either to a cognitive behavioral intervention group or an educational play group (acting as control group). Mediation analyses were conducted using path analysis. The stronger reduction in child ODD symptoms in the CBT group compared to the control group led to a decrease in parental depression and stress, as indicated by significant indirect effects (ab = 0.07 and ab = 0.08, p < 0.05). The proposed model for mechanisms of change was, therefore, confirmed for two of the three outcome parameters. Parental psychopathology and stress can be modified by child-centered CBT. The preceding reduction in ODD symptoms acts as a mediator for at least some of the changes in parental depression and stress. However, due to some limitations of the study, other possible explanations for the results found cannot be completely ruled out and are, therefore, discussed.

Keywords Oppositional defiant disorder · Conduct disorder · Parental psychopathology · Parental stress · Cognitive behavioral treatment · Mediation

Introduction

Disorders associated with aggressive behavior represent one of the most common groups of mental disorders in children and adolescents [1]. The DSM-5 describes disrup- tive, impulse-control, and conduct disorders as comprising

behaviors that violate the rights of others and/or bring the individual into conflict with societal norms or authority fig- ures [2]. Whereas children with conduct disorder (CD) show a persistent pattern of aggressive and antisocial behavior as well as serious violation of rules, those with oppositional defiant disorder (ODD) exhibit a pattern of angry/irritable mood, temper tantrums, and argumentative/defiant behavior, but without severe aggressive or antisocial behavior [2].

ODD and CD have been shown to be related to paren- tal psychopathology, in particular symptoms of depression and anxiety [3, 4] as well as parental stress [5]. Children of mothers with depressive symptoms show a higher preva- lence of behavior problems than children of non-depressed mothers [6]. Several studies have demonstrated the influence of maternal or parental depressive symptoms [7–16], mater- nal or parental anxiety [8, 9, 15, 17] and maternal or paren- tal stress [16, 18–21] on conduct or externalizing behavior problems in children and adolescents. However, due to the

* Josepha Katzmann [email protected]

1 Department of Child and Adolescent Psychiatry and Psychotherapy, School of Child and Adolescent Cognitive Behavior Therapy, University Hospital Cologne, Pohligstraße 9, 50969 Cologne, Germany

2 Institute of Child and Adolescent Psychotherapy of the Christoph-Dornier-Foundation for Clinical Psychology, University of Cologne, Cologne, Germany

3 Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, Medical Faculty of the University of Cologne, Cologne, Germany

1182 European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

cross-sectional nature of some of these studies, it is difficult to draw conclusions regarding causality [8–10, 13, 15, 19].

It is reasonable to assume both that parental mental health problems affect the child’s wellbeing and may result in oppo- sitional defiant- or aggressive behavior, and that, conversely, having a noncompliant, defiant, and aggressive child repre- sents a major stressor for parents, potentially resulting, for example, in feelings of depression or stress. Several studies found effects of child ODD symptoms on parental stress [22–25]. Instead of a unidirectional influence from paren- tal psychopathology to child ODD/CD symptoms, a bidi- rectional relationship with mutual influences is considered more likely [4, 8–10, 19, 26–28]. Indeed, several studies found evidence for a bidirectional model of the relationship between child and parental psychopathology [17, 29–33] and between child externalizing psychopathology and parental stress [5, 34, 35]. An experimental study found that deviant child behavior was causally related to parental stress and a negative parental mood [36]. A review of mediating mecha- nisms of mutual influences of maternal depression and child behavior problems revealed, among other things, that paren- tal discipline (inconsistent, lax, ineffective parenting) and dysfunctional parent–child interaction mediated the effects both from parent to child and vice versa [37]. However, the findings regarding a mutual influence of parental psychopa- thology and child externalizing problems are not consistent, as some studies found only unidirectional influences from parent to child [27, 38].

Several cognitive-behavioral- parent-based- or child- based interventions have been shown to be effective in the treatment of children with ODD/CD. Parent-management training and child-based interventions, including cognitive problem-solving training, anger-control training, or social skills training, have been evaluated as evidence-based treat- ments [39, 40]. Meta-analyses demonstrated at least small to medium effects of parent-based and child-based interven- tions in reducing or preventing child behavior problems [1, 41–43].

In addition to reducing child behavior problems, par- ent management training has also been found to exert positive effects on parental depression, anxiety and stress [44, 45], although the effects do not seem to be consist- ent [e.g., 46]. However, not only interventions that directly target the parents but also child-centered treatment has beneficial effects on parental psychopathology (depres- sion, stress and overall psychopathology) [47–49]. To date, however, the exact mechanism that leads from child- centered treatment to decreases in parental mental health problems remains unclear [47]. Regarding the bidirectional relationship between child and parental psychopathology, it is reasonable to assume that in child-centered treatment, a decrease in child behavior problems constitutes (at least in part) a mechanism of change for a reduction in parental

psychopathology, as has been similarly assumed for the effects of parent management training on parental mental health problems [45]. Indeed, two studies in the context of preventive parent management training might support this hypothesis: When investigating the possible mechanism of action of the effects of preventive parent training programs on parental psychopathology, both trials showed that com- pared to a control group, the effects of the parent training on maternal depression were mediated by a previous reduction in externalizing behavior problems of the child, which was in turn mediated by a change in parenting behavior [50, 51].

To the best of our knowledge, the mechanism of action for effects of child-centered treatment of ODD/CD on parental stress and psychopathology has not yet been evaluated. Therefore, the goal of the present study was to examine possible mechanisms of change in a child-based therapy on parental depression, anxiety, and stress in a randomized controlled trial. The treatment was based on the German Treatment Program for Children with Aggres- sive Behavior (THAV) [52]. THAV provides individual- ized treatment based on the specific problem-maintaining factors in each child, namely social cognitive information processing, impulse control, social problem-solving, social skills, and social interactions in these situations. The treat- ment combines patient-, parent-, teacher-, and peer-focused interventions. Patient-focused interventions are the main component, while parent-, teacher-, or peer-focused inter- ventions are added in accordance with the individual needs of the patient. Previous studies found that THAV leads to a stronger decrease in child ODD symptoms from pre- to post- treatment compared to a waiting phase [53] and compared to an active control group, and to a significant reduction in maternal stress [54].

The mechanism of change in a treatment can be exam- ined using mediation analyses. A mediating variable can be defined as “those characteristics of the individual that are changed by the treatment and that, in turn, produce change in the outcome of interest” [55, p. 248]. In the present study, we hypothesized that the effects of THAV for children with ODD on parental depression, anxiety and stress would be mediated by a preceding reduction in child symptoms, com- pared to an active control group. To our knowledge, this is the first study to address this hypothesis.

Methods

Participants

Patients were recruited via cooperation with outpatient units and private practices of child and adolescent psychiatry or child and adolescent psychotherapy, youth welfare offices, schools, and the media. Parents and children gave their

1183European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

informed consent to participate in the study after the pro- cedure had been fully explained. No incentives were given for taking part in the trial. The main study [54] investigated the efficacy of THAV especially on peer-related aggressive behavior. Therefore, only boys were included because the majority of patients with (peer related) overt aggressive behavior are male, while girls more often use relational aggression [56]. Further inclusion criteria for children were age 6–12 years, IQ ≥ 80 and an ICD-10 diagnosis [57] of CD (F91), ODD (F91.3), mixed disorder of conduct and emotions (F92), or hyperkinetic conduct disorder (F90.1) using the semi-structured interview for Disruptive Behavior Disorders (ODD, CD) (DCL-DBD) of the German Diag- nostic System for Children and Adolescents (DISYPS-II) [58]. Moreover, peer-related aggressive behavior had to cause persistent impairments in relationships with other children (clinical rating in the semi-structured interview) and the child had to have a high symptom score (Stanine score ≥ 7) on the German Symptom Checklist for Disruptive Behavior Disorder (SCL-DBD) total score of the DISYPS-II [58] at the pre-assessment immediately prior to the start of the treatment phase. Exclusion criteria were the presence of a primary comorbid disorder (e.g., autism) according to the judgment of the clinician, a planned change in medi- cation in a child receiving psychotropic medication, other child psychotherapy, parents suffering from a severe mental disorder and parents with insufficient knowledge of the Ger- man language.

Study design

The protocol of the study (Clinical trials.gov Identifier: NCT01406067) was approved by the ethics committee of the University Hospital, Cologne. Children were randomly assigned to an individually delivered THAV group or to an active control group (PLAY). In the current study, the treatment and control interventions were delivered by 13 experienced child therapists or therapists in training. The therapists received weekly group supervision by a senior child therapist (A. G.-D.).

Data were collected at pre-treatment, at three time points during treatment (every six sessions), immediately post- treatment after 24 sessions, and at a 12-month follow-up. The analyses in the present paper only included the data at pre-treatment (t1), after 18 sessions of treatment (t2), and at post-treatment (t3). The putative mediating variable in parent rating was assessed at t2 to ensure that change in the putative-mediating variable occurred before change in the outcome.

The study sample size was determined by power cal- culations for the main analyses of treatment effects [54] (N = 100; n = 50 in the THAV group, n = 50 in the PLAY

group to detect a moderate effect size with 80% power using a two-tailed t test with type I error rate set at 0.05).

Interventions

Treatment with THAV was delivered individually and com- prised 24 weekly child sessions (lasting for 45 min each) and additional sessions or shorter contacts with parents. THAV is a cognitive-behavioral intervention that is individually tailored to the needs of each participating child and his/her parents. The specific problems of each child are defined via a semi-structured interview at the beginning of the treatment, and a concept of the disturbance and a treatment rationale are developed together with the child and parents during the first sessions. The main modules of the child training are as follows: (a) psychoeducation and development of a therapeutic relationship; (b) social cognitive interventions; (c) anger-control training; (d) social problem-solving and skills training; and (e) relapse prevention. These topics are covered individually as needed. In the sessions for parents and teachers, the child’s target problems are identified, together with his/her competencies and the coercive inter- action process. Problem-maintaining social interactions are addressed in each module by interventions that aim to modify these interactions, for instance by teaching parents how to define social rules, how to use methods for reward- ing the child (e.g., token systems) when the child shows prosocial behavior, and how to use appropriate methods of punishment (e.g., timeouts) when the child shows aggressive behavior. Additionally, the sessions for the parents aim to identify and modify parental dysfunctional thoughts about their child, about their own aggressive behavior, impulse control, and conflict management. If possible, classmates are instructed by teachers or parents to support socially compe- tent conflict-solving behavior. A detailed description of the treatment components can be found in [54].

The active control condition comprised educational group play, with 3–5 children in each group. Each group received 12 biweekly sessions (lasting for 90  min each) over 24 weeks. These group sessions covered techniques to activate resources and the opportunity to practice prosocial interactions. No specific problem-solving techniques (e.g., development of alternative solutions, evaluation of solu- tions) or other cognitive interventions (e.g., identification of anger thoughts) were implemented. Moreover, the ses- sions did not include any skill training with role playing and rehearsal or interventions to support transfer to real life (e.g., therapeutic homework assignments). Parents attended two group sessions for parents (90 min each), providing psych- oeducation on appropriate general parenting strategies (e.g., establishing rules to reduce peer-related aggressive behav- ior, communication of effective commands, rewarding the child for prosocial behavior, or appropriate methods of

1184 European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

punishment for aggressive behavior). However, this general parenting advice was not tailored to the specific problems of the child, and the parents were not trained to implement these techniques in their daily parenting behavior.

Measures

Diagnosis of conduct disorders

The semi-structured interview for disruptive behavior dis- orders (DCL-DBD; including ODD and CD) [58] rated by child therapists was used to assess all ODD symptoms according to ICD-10 [57] criteria. In the current study, the DCL-DBD was administered before randomization, only to confirm the inclusion criterion of ICD-10 diagnosis of a dis- order associated with aggressive behavior. Child therapists rated each of the 25 items on a four-point Likert scale rang- ing from 0 (not at all) to 3 (very much). The DCL-DBD has been shown to be factorially valid and internally consistent (Cronbach’s α = 0.68–0.84) [59].

Child‑oppositional defiant behavior

Parent-rated ODD symptoms were assessed at t1, t2 and t3 according to ICD-10 [57] and DSM-IV [60] criteria using the corresponding subscale of the symptom checklist for dis- ruptive behavior disorders (SCL-DBD) [58]. Respondents (parents) rated each of the nine items on a four-point Likert scale ranging from 0 (not at all) to 3 (very much). The SCL- DBD has been shown to be factorially valid [61, 62]. We cal- culated item mean scores, which showed sufficient to good internal consistency in the present sample, with Cronbach’s α ranging from 0.76 to 0.88 at the three-assessment time points. Higher scores indicate higher levels of oppositional defiant behavior.

Parental psychopathology

Parents’ mental health problems were assessed at t1 and t3 via self-report using the German version of the Depres- sion–Anxiety–Stress Scale (DASS) [63, 64]. The 42-item questionnaire consists of three 14-item scales measuring the negative emotional states of depression, anxiety and stress, respectively, in the preceding week. Parents rated each of the items on a four-point Likert scale ranging from 1 (did not apply to me at all) to 4 (applied to me very much, or most of the time). For analysis, item mean scores were cal- culated for the three scales, with higher scores indicating a higher degree of pathology. The convergent and discrimi- nant validity of the DASS scales has been shown [63], as well as the superior validity of a three-factor model over models with one or two factors [65]. In the present sample, the scales showed good to very good internal consistencies

at pre- and post-treatment (Cronbach’s α ranging from 0.81 to 0.91). The subscales correlated with each other (r = 0.69 to r = 0.74, p < 0.01).

Statistical analyses

Baseline differences

Demographics and baseline psychosocial variables were compared between the two intervention groups using Chi- square tests (for categorical variables) or t tests for inde- pendent samples (for continuous variables).

Model fit indices

Model fit was determined using the root mean square error of approximation (RMSEA), Comparative Fit Index (CFI), and standardized root mean square residual (SRMR). The criteria for a good model fit were RMSEA values < 0.08, CFI values > 0.95, and SRMR values < 0.08 [66].

Mediation analyses

In a simple mediation model, the independent variable (X) affects the dependent variable (Y) through a mediating vari- able (M). The total effect of X on Y (expressed as the coef- ficient c) is the sum of the direct effect (c′) and the indirect or mediating effect (ab). The indirect effect consists of two paths, where X predicts M (a) and M predicts Y after control- ling for X (b). The indirect effect is defined as the product of the unstandardized regression coefficients a and b (i.e., ab), which can also be tested for significance [67]. Researchers consider that one can still conduct mediation analyses when the c coefficient for the total effect of X on Y is nonsignifi- cant [e.g., 67, 68]. Rucker et al. [68] gave several reasons why a significant indirect effect can occur in the absence of a significant total effect. Additionally, it is recommended to focus on the significance of the indirect effect ab as a sole indicator of mediation rather than also considering signifi- cant a and b paths for significant mediation [69].

Mediation analysis was conducted using structural equa- tion modeling with Mplus [70]. In our model, the independ- ent variable was group membership, operationalized as a dichotomous variable (0 = PLAY, 1 = THAV). We tested parent-rated parental depression, parental anxiety and paren- tal stress as three different outcome variables in one com- mon model and ODD symptoms of the child, assessed via parent rating at t2 (18 sessions after the start of treatment), as the putative mediating variable for the effects of THAV on parental psychopathology, respectively. The three DASS scales at t1 and t3 were allowed to covary.

To control for pre-treatment differences, the pre-treatment scores of the outcome measures and putative mediating

1185European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

variable were included as covariates in the model. Addition- ally, to control for effects of the extent of parental involve- ment in child treatment on parental mental health, we added the number of sessions with participation of the parents as a predictor for DASS outcome scores at t3. The model was estimated using maximum likelihood estimation. For signifi- cance testing of the indirect effects, direct effects, and total effects, we calculated bias-corrected bootstrap confidence intervals (CIs) with 10,000 resamples, as recommended by Preacher and Hayes [67]. The effects were regarded as sig- nificant if 0 was not included in the 95% CI. In accordance with Hayes [71], we report the effects as unstandardized regression coefficients.

Results

Attrition and treatment of missing values

Figure  1 shows the flow of the participants through the study. A total of 101 patients were randomized (block rand- omization: 50 THAV; 51 PLAY). Ten patients dropped out from the PLAY condition after randomization because of ethical objections of the therapists. For these patients, the more effective treatment according to the hypotheses of the trial was strictly indicated according to the regulations of the ethics committee, which required that the patients must not be disadvantaged due to their participation in the trial. This occurred if, for example, there was a high risk that a child would be excluded from school because of behavioral problems, meaning that school-based interventions were additionally necessary. Although these ten patients were excluded from the trial, they received the skills training and

Fig. 1 Flow of participants through the study

1186 European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

additional treatment components as needed. Besides these ten patients, all other patients completed the THAV and PLAY conditions. As our study was aimed at investigating the potential mediator for change in parental psychopathol- ogy, our analysis only included those families who com- pleted the program. Unfortunately, and despite our explicit instructions, in ten cases, the person who completed the questionnaire assessing parental psychopathology was not the same from pre- to post-test. To prevent any bias due to change of rater, we decided to exclude these families from the final analyses. Among the resulting final sample of 81 patients and their parents (THAV: n = 44, PLAY: n = 37), missing values were handled using full information maxi- mum likelihood estimation [72].

Sample description

Table 1 shows the demographic characteristics of the THAV and PLAY groups. There were no significant differences for the demographic characteristics of the families at pre- treatment between the THAV and PLAY groups as well as between the subsamples of mothers and fathers (the latter not depicted in the table). Compared to data from a nonclinical

sample of adults, the mean of parents’ raw scores on the three DASS scales all lay above the 90th percentile [65].

Treatment with THAV

In treatment with THAV, parent-focused interventions were always included. Parents attended an average of about eight sessions or shorter contacts (M = 8.2, SD = 4.0). Addition- ally, in about 10% of the cases 1–3 short teacher consulta- tions by telephone were added if parents agreed to contact the teacher and if child-aggressive behavior was present in the classroom and observable by the teacher. In only two cases (4%) it was possible that classmates were instructed by teachers or parents to support socially competent conflict- solving behavior. Treatment integrity and treatment adher- ence, both rated by the therapist, were good to excellent in the THAV group (for further details, see [54]).

Mediation model

The results of the mediation model with different outcome variables are presented in Fig. 2. The model showed a good to excellent fit to the data (χ2 = 19.77; df = 21; p = 0.54; CFI = 1.00; RMSEA = 0.00; SRMR = 0.06).

Table 1 Demographic characteristics of THAV and PLAY group and tests for between group differences at pre-treatment

Symptom severity (Stanine) = Stanine score in ODD subscale in the Symptom Checklist for Disrup- tive Behavior Disorders (SCL-DBD). Gender (female) = gender of parent participating in study. DASS scores = item mean scores on scale in the Depression-Anxiety-Stress-Scale (DASS)

Variable THAV (n = 44) PLAY (n = 37) Test statistic (df) p

M SD % M SD %

Child descriptives  Age (years) 8.8 1.8 8.6 1.6 t(79) = − 0.53 0.60  Intelligence (IQ) 108.0 16.2 104.8 12.2 t(79) = − 0.97 0.34  Diagnosis (ICD-10-Code) χ2(4) = 2.22 0.88   F91.1 0.0 2.7   F91.2 2.3 0.0   F91.3 77.3 73.0   F92.8 2.3 2.7   F90.1 18.2 21.6

 Symptom severity (Stanine) 8.3 0.8 8.1 0.8 t(79) = − 0.67 0.50  Medication (stimulants) 11.4 16.2 χ2(1) = 0.40 0.53

Parent descriptives  Gender (female) 90.9 86.5 χ2(1) = 0.40 0.53  At least part time employed   Mother 52.3 48.6 χ2(1) = 0.11 0.75   Father 79.5 75.0 χ2(1) = 0.24 0.63

 Single parent status 36.4 45.9 χ2(2) = 1.48 0.48  DASS scores   Depression 1.5 0.6 1.6 0.6 t(79) = 0.72 0.47   Anxiety 1.4 0.3 1.4 0.4 t(79) = 46 0.65   Stress 2.1 0.5 2.2 0.5 t(79) = 26 0.79

1187European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

For parental depression and anxiety, no significant total effect of treatment condition was observed at post-treatment when controlling for pre-treatment scores and number of ses- sions with participation of the parent (see total effect boxes in Fig. 2). However, there was a significant total effect for

parental stress, with a c3 coefficient of − 0.36 (CI − 0.64 to − 0.09), indicating that parents whose children were treated with THAV described fewer stress symptoms at post-treat- ment than did parents of children who participated in the PLAY group. These results concerning treatment effects on

Fig. 2 Mediation model with multiple outcomes (parental depression, anxiety and stress in parent rating) and ODD symptoms in parent rating as putative mediating variable. Notes: For reasons of simplic- ity, model results are depicted separately for each outcome while they were analyzed in one common model. a = unstandard- ized regression coefficient for effect of treatment on mediator. bi = unstandardized regression coefficient for effect of mediator on outcome. c′i = unstandardized regression coefficient for direct effect of treatment on outcome, controlling for putative media- tor. abi = indirect effect; product of a and bi. ci = unstandard- ized regression coefficient for total effect of treatment on outcome. Asterisk = significant effect based on bias-corrected confidence interval. Measures: Parental depression, parental anxiety, and parental stress: DASS. ODD symptoms: SCL- DBD

1188 European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

parental psychopathology and stress were already published elsewhere and are presented here only for the sake of com- pleteness [54].

Group membership significantly predicted changes in ODD symptoms of the children at t2, as rated by the parents (see a coefficients in Fig. 2). Parents in the THAV group described a greater improvement in child ODD symptoms than did parents in the PLAY group. Furthermore, ODD symptoms of the child at t2 predicted the extent of parental depression at post-treatment (see b1 coefficient in Fig. 2). This finding indicates that the more ODD symptoms of the child were observed by the parents at t2, the more feel- ings of depression the parents experienced themselves at post-treatment (when controlling for pre-treatment scores, respectively). The prediction of maternal anxiety and stress at post-treatment by ODD symptoms of the child at t2 were nonsignificant but in predicted direction.

The tests for indirect effects (mediating effects) of treat- ment group on outcome variables at t3 for parental depres- sion and stress via ODD symptoms of the child at t2 reached significance (see ab1 and ab3 coefficients in Fig. 2). This may indicate that parents in the THAV group experienced less depression and stress at post-treatment due to a preced- ing decrease in ODD symptoms of their children, compared to parents in the PLAY group. There was no indirect (media- tion) effect for change of ODD symptoms on parental anxi- ety. The tests for direct effects failed to reach significance for all three DASS outcome variables, as shown by the c’ coefficients in Fig. 2.

The DASS scores at posttreatment correlated significantly (depression and anxiety: r = 0.73, p < 0.01, depression and stress: r = 0.65, p < 0.01, anxiety and stress: r = 56, p < 0.01).

Discussion

Our findings suggest that the decrease in child ODD symp- toms may act as a mediating mechanism for the reduction in parental mental health problems (depression and stress) in patient-based cognitive-behavioral treatment (THAV) compared with a resource-activating educational PLAY group. Children who participated in THAV showed a greater decrease in ODD symptoms at t2, as rated by their parents, than children in PLAY. This, in turn, possibly resulted in a stronger decrease in parental depression and stress at post- treatment. It has to be noted, however, that even though we were able to detect significant indirect effects on parental depression and stress, the mean of the raw scores on the DASS at post-treatment still lay above the 90th percentile compared to nonclinical norms [65], indicating that parents were still affected after the intervention.

It is difficult to draw comparisons of our findings with previous research, as to our knowledge, the present study

is the first to examine possible mechanisms for the change in parental psychopathology after a patient-based interven- tion for children with ODD/CD. Two prevention studies with nonclinical samples reported comparable results: The decrease in child-externalizing behavior problems after a parent training acted as a mediator of the decrease in mater- nal depressive symptoms [50, 51]. As in other child-based treatment studies, the cognitive-behavioral intervention (THAV) had a positive effect on parental stress [47–49] compared to the active control group. Our results are also in line with previous findings concerning the influence of child conduct problems on parental psychopathology [e.g., 5, 30, 37].

As recommended, we chose the putative mediating vari- able on the basis of previous theories and research [73]. It is also worth pointing out that we assessed the mediator before the outcome variables, making it possible to exam- ine the timeline between changes in each variable, which is a crucial requirement for causal interpretation of indirect effect models [74].

However, with respect to the probable bidirectional rela- tionship between child and parental psychopathology, it is unfortunate that we did not assess parental depression and stress during treatment too. This prevents us from testing the reverse order of changes (a decrease in parental depression and/or stress leading to a decrease in child ODD symptoms) or even cross-lagged relationships. We consider that the pro- posed sequence of child ODD symptoms as a mediator for the change in parental depression and stress is theoretically reasonable because of the child-based focus of the interven- tion. Nevertheless, it would be desirable for future research to also test for other directions of causality.

It is important to consider that, while we found indirect effects for two of the three outcome variables (parental depression and stress), we were only able to detect a signifi- cant total effect of group membership on parental stress in favor of the THAV group. The total effect of the interven- tion on parental depression remained nonsignificant, as did the direct effects on depression and stress, after controlling for the mediating variable. At first glance, it might be not clear how these findings should be interpreted. More specifi- cally, it is not obvious why an indirect (mediating) effect of change in ODD symptoms on parental depression has been observed without there first being any effect to be mediated at all. According to Hayes and Rockwood [69], who refer to Kenny and Judd [75], a situation like this can occur due to the different power of the significance tests of indirect and total effect. This is one of the reasons why Hayes and Rock- wood [69] argue for the significance test of the indirect (i.e., mediating) effect as the sole indicator for mediation, regard- less of the (non-)significance of the other paths in the model.

Furthermore, when interpreting the total effect, one should keep in mind that it constitutes a sum of all possible

1189European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

mediating effects as well as the direct effect of the interven- tion on the outcome [69]. If two mediating effects of equal size differed in sign, they would add up to a zero total effect. Thus, it is possible that one mediating effect works in favor of the intervention group while the other one works in favor of the control group. In our study, this could mean that even though the reduction in ODD symptoms in the THAV group may have possibly led to a decrease in parental depres- sion, there was at least one other effect (not included in our model) in the control group, which also led to a decrease in parental depressive symptoms. One such putative effect might be the contact with other affected parents during the group sessions for parents. Unfortunately, we did not assess any corresponding variables that might have worked in favor of the control group.

Even if the change in ODD symptoms in the THAV group acted as a mediator for the decrease in parental mental health, the mechanism of change for the preced- ing decrease in ODD symptoms in THAV remains unclear. Two other studies showed a mediating effect of parenting [50, 51], which we did not assess in our study. However, it should be taken into account that the intervention in these two studies consisted of parent management training in a prevention trial, in contrast to our study, which comprised a child-based therapy in an individual setting with clinically referred children. It is, therefore, questionable whether the findings of the other mediation models can be easily trans- ferred to ours. In the THAV group, the number of sessions with participation of the parents varied according to indi- vidual needs. Regression analyses revealed that the number of sessions with parents’ participation did not predict ODD symptoms at t2 or parental depression, anxiety and stress at post-treatment (when controlling for pre-treatment scores, respectively). This finding may indicate that it was actually the child-centered part of the treatment, and not the module focusing on a change in parenting behavior, which led to changes. In the PLAY groups, two group sessions per parent were scheduled. Nevertheless, it would have been prefer- able to examine the role of parenting, particularly because this might also change due to child-centered treatment. This should be considered in further studies. In a related article, we have already examined and discussed other putative- mediating mechanisms for the decrease in ODD symptoms in our sample [76].

Additionally, it is not sure if the decrease in ODD symptoms does fully explain the effect of THAV on paren- tal depression and stress, even though the direct effects of group membership after controlling for the mediator are nonsignificant. Since it is unrealistic that an interven- tion only affects the outcome through one single mech- anism [69]. Other putative mediating variables need to be considered, such as the parents’ knowledge that their child is taking part in a treatment and the related hope

for improvement, as well as, again, parenting behavior or parenting-related cognitions. In a recent review, analogi- cal hypotheses were stated in the context of mediation for the decrease in parental stress after participation in parent management training [45].

Our study is further limited by the fact that mediating and outcome variables were all assessed in parent ratings. It is, therefore, possible that child and parental psychopathology were subject to a common bias of evaluation. Some studies reinforce this argument, while other researchers found that parents rate their child’s oppositional defiant behavior prob- lems independently from the extent of their own psychologi- cal distress [see 77 for a review]. A validation of the parents’ ratings via clinical rating would have been preferable and should be considered in future research. Nevertheless, even if we assume that we assessed a change in the parents’ per- ception rather than a change in actual behavior of the child, this does not, in our view, compromise the interpretation of our findings. If a biased evaluation of a child’s behavior problems leads to a decrease in the mental burden of the parents, this too would be a helpful effect. Regarding the bidirectional relationship of parents’ and children’s psycho- pathology, it may even be the case that, following a decrease in parental depression and stress due to a perceived reduc- tion in the child’s ODD symptoms, an actual reduction in the child’s behavior problems may occur, setting in motion a positive spiral of mutual influences. In a further analysis, we tested parental depression and stress at post-treatment (controlled for pre-treatment scores) as a mediator for effects of THAV compared to PLAY on ODD symptoms at the 12-month follow-up (controlled for post-treatment scores) and found a significant mediating effect of parental stress. This finding might be cautiously interpreted as an indication of the proposed sequence of mutual influences. However, because of a poor model fit, we did not consider it reliable.

Another—often noted—limitation is the underrepresenta- tion of fathers in the sample. Although we made no restric- tions concerning the gender of the parent participating in our study, the sample mostly comprised mothers. Therefore, our findings cannot be easily transferred to effects of child- based CBT on fathers’ psychopathology. Similarly, as our study only examined treatment effects on boys, no conclu- sions can be drawn concerning the effects of CBT on girls with ODD/CD. In future research, the examination of effects of THAV on girls’ aggressive behavior problems would be desirable. In another, yet unpublished trial, we investigated a treatment program very similar to THAV in a sample that included both boys (92%) and girls (8%) (Goertz-Dorten et al., in preparation).

In further studies, we would find it interesting to also include comorbidities of the aggressive behavior in the analysis, as aggressive behavior and ODD can also be a concomitant phenomenon of an anxiety disorder.

1190 European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

Finally, it is important to note that parental depression and stress correlated highly at pre- and post-treatment. It has already been shown via factor analysis that the three DASS subscales are nevertheless best viewed as different constructs [65]. However, the high correlations might point to a slightly different interpretation of our findings. It is also possible that the reduction in parental stress, following a preceding reduction in ODD symptoms in the child, led to a reduction in parental depression. It is not possible to exam- ine this hypothesis with our data because we did not assess parental depression and stress in a time-delayed order.

In conclusion, our study suggests an important relation- ship between child ODD symptoms and parental depression and stress, which needs to be considered even in the context of the child-based treatment of oppositional defiant- and aggressive behavior problems. The assumption that child- based interventions can also help to lessen the mental burden of the parents is highly relevant given, for example, that not all parents—and especially those who perceive psychologi- cal problems of their own—are able to participate in parent management training or are open to therapeutic advice. Even if parents are not the primary focus of the treatment, they might benefit with respect to their own psychopathology and that might be due to a change in child behavior problems.

Acknowledgements The study received financial support from the School of Child and Adolescent Cognitive Behavior Therapy at the University Hospital Cologne. The authors wish to thank all families who participated in this study.

Compliance with ethical standards

Conflict of interest Anja Görtz-Dorten and Manfred Döpfner receive royalties from publishing companies as authors of books and treatment manuals on child behavioral therapy, and of assessment manuals, in- cluding the treatment manual for THAV, which is evaluated in this trial. Josepha Katzmann declares that she has no conflict of interest.

Ethical approval All procedures performed in studies involving human participants were approved by the ethics committee of the University Hospital of Cologne and were therefore performed in accordance with the 1964 Declaration of Helsinki and its later amendments. This article does not contain any studies with animals performed by any of the authors. Informed consent was obtained from all individual participants prior to their inclusion in the study.

References

1. National Institute for Health and Care Excellence (2013) Anti- social behaviour and conduct disorders in children and young people: Recognition and management, national clinical guideline number 158. National collaborating centre for mental health and social care institute for excellence, commissioned by the national institute for health and clinical excellence, published by the british psychological society and the royal college of psychiatrists. https ://www.nice.org.uk/guida nce/cg158 . Accessed 9 Dec 2017

2. American Psychiatric Association (2013) Diagnostic and statis- tical manual of mental disorders, 5th edn. American Psychiatric Association, Washington

3. Beck CT (1999) Maternal depression and child behaviour prob- lems: a meta-analysis. J Adv Nurs 29(3):623–629. https ://doi. org/10.1046/j.1365-2648.1999.00943 .x

4. Connell AM, Goodman SH (2002) The association between psychopat hology in fat hers versus mot hers and chil- dren’s internalizing and externalizing behavior problems: a meta-analysis. Psychol Bull 128(5):746–773. https ://doi. org/10.1037//0033-2909.128.5.746

5. Mackler JS, Kelleher RT, Shanahan L, Calkins SD, Keane SP, O’Brien M (2015) Parenting stress, parental reactions, and externalizing behavior from ages 4 to 10. J Marriage Fam 77(2):388–406. https ://doi.org/10.1111/jomf.12163

6. Civic D, Holt VL (2000) Maternal depressive symptoms and child behavior problems in a nationally representative normal birthweight sample. Matern Child Health J 4(4):215–221. https ://doi.org/10.1023/A:10266 67720 478

7. Elgar FJ, Mills RS, McGrath PJ, Waschbusch DA, Brownridge DA (2007) Maternal and paternal depressive symptoms and child maladjustment: the mediating role of parental behavior. J Abnorm Child Psychol 35(6):943–955. https ://doi.org/10.1007/ s1080 2-007-9145-0

8. Hautmann C, Eichelberger I, Hanisch C, Plück J, Walter D, Döpfner M (2015) Association between parental emotional symptoms and child antisocial behaviour: what is specific and is it mediated by parenting? Int J Behav Dev 39(1):43–52. https ://doi.org/10.1177/01650 25414 53512 0

9. Vera J, Granero R, Ezpeleta L (2012) Father’s and mother’s perceptions of parenting styles as mediators of the effects of parental psychopathology on antisocial behavior in outpa- tient children and adolescents. Child Psychiatry Hum Dev 43(3):376–392. https ://doi.org/10.1007/s1057 8-011-0272-z

10. Kane P, Garber J (2009) Parental depression and child exter- nalizing and internalizing symptoms: unique effects of fathers’ symptoms and perceived conflict as a mediator. J Child Fam Stud 18:465–472. https ://doi.org/10.1007/s1082 6-008-9250-x

11. Dette-Hagenmeyer DE, Reichle B (2014) Parents’ depressive symptoms and children’s adjustment over time are mediated by parenting, but differentially for fathers and mothers. Eur J Dev Psychol 11(2):196–210. https ://doi.org/10.1080/17405 629.2013.84878 9

12. Kim-Cohen J, Moffitt TE, Taylor A, Pawlby SJ, Caspi A (2005) Maternal depression and children’s antisocial behavior. Arch Gen Psychiatry 62:173–181. https ://doi.org/10.1001/archp syc.62.2.173

13. McCauley Ohannessian C, Hesselbrock VM, Kramer J, Kuperman S, Bucholz KK, Schuckit MA, Nurnberger JI (2005) The relation- ship between parental psychopathology and adolescent psychopa- thology: an examination of gender patterns. J Emot Behav Disord 13(2):67–76. https ://doi.org/10.1037/t0392 6-000

14. Shaw DS, Owens EB, Giovannelli J, Winslow EB (2001) Infant and toddler pathways leading to early externalizing disorders. J Am Acad Child Adolesc Psychiatry 40(1):36–43. https ://doi. org/10.1097/00004 583-20010 1000-00014

15. Meadows SO, McLanahan SS, Brooks-Gunn J (2007) Parental depression and anxiety and early childhood behavior problems across family types. J Marriage Fam 69:1162–1177. https ://doi. org/10.1111/j.1741-3737.2007.00439 .x

16. Qi CH, Kaiser AP (2003) Behavior problems of preschool chil- dren from low-income families: review of the literature. Top Early Child Spec 23(4):188–216. https ://doi.org/10.1177/02711 21403 02300 40201

17. Elgar FJ, Waschbusch DA, McGrath PJ, Stewart SH, Curtis LJ (2004) Temporal relations in daily-reported maternal mood and

1191European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

disruptive child behavior. J Abnorm Child Psychol 32(3):237– 247. https ://doi.org/10.1023/B:JACP.00000 26138 .95860 .81

18. Deater-Deckard K, Pinkerton R, Scarr S (1996) Child care qual- ity and children’s behavioral adjustment: a four-year longitudi- nal study. J Child Psychol Psychiatry 37(8):937–948. https :// doi.org/10.1111/j.1469-7610.1996.tb014 91.x

19. Anthony LG, Anthony BJ, Glanville DN, Naiman DQ, Waanders C, Shaffer S (2005) The relationships between parenting stress, parenting behaviour and preschoolers’ social competence and behaviour problems in the classroom. Infant Child Dev 14:133– 154. https ://doi.org/10.1002/icd.385

20. Crnic KA, Gaze K, Hoffman C (2005) Cumulative parenting stress across the preschool period: relations to maternal parent- ing and child behaviour at age 5. Infant Child Dev 14:117–132. https ://doi.org/10.1002/icd.384

21. D e a t e r- D e ck a r d K , S c a r r S ( 1 9 9 6 ) Pa r e n t i n g st r e s s among dual-ear ner mothers and fathers: are there gen- der differences? J Fam Psychol 10(1):45–59. https ://doi. org/10.1037/0893-3200.10.1.45

22. Calkins SD (2002) Does aversive behavior during toddlerhood matter? The effects of difficult temperament on maternal percep- tions and behavior. Inf Mental Health J 23(4):381–402. https :// doi.org/10.1002/imhj.10024

23. Podolski C-L, Nigg JT (2001) Parent stress and coping in rela- tion to child adhd severity and associated child disruptive behav- ior problems. J Clin Child Psychol 30(4):503–513. https ://doi. org/10.1207/S1537 4424J CCP30 04_07

24. Williford AP, Calkins SD, Keane SP (2007) Predicting change in parenting stress across early childhood: child and maternal factors. J Abnorm Child Psychol 35(2):251–263. https ://doi.org/10.1007/ s1080 2-006-9082-3

25. Anastopoulos AD, Guevremont DC, Shelton TL, DuPaul GJ (1992) Parenting stress among families of children with attention deficit hyperactivity disorder. J Abnorm Child Psychol 20(5):503– 520. https ://doi.org/10.1007/bf009 16812

26. Rutter M (1990) Commentary: some focus and process considera- tions regarding effects of parental depression on children. Dev Psychol 26(1):60–67. https ://doi.org/10.1037/h0092 669

27. Elgar FJ, Curtis LJ, McGrath PJ, Waschbusch DA, Stewart SH (2003) Antecedent-consequence conditions in maternal mood and child adjustment: a four-year cross-lagged study. J Clin Child Adolesc Psychol 32(3):362–374. https ://doi.org/10.1207/S1537 4424J CCP32 03_05

28. Morgan J, Robinson D, Aldridge J (2002) Parenting stress and externalizing child behaviour. Child Fam Soc Work 7:219–225. https ://doi.org/10.1046/j.1365-2206.2002.00242 .x

29. Hammen C, Burge D, Stansbury K (1990) Relationship of mother and child variables to child outcomes in a high-risk sample: a causal modeling analysis. Dev Psychol 26(1):24–30. https ://doi. org/10.1037/0012-1649.26.1.24

30. Ge X, Conger RD, Lorenz FO, Shanahan M, Elder GEJ (1995) Mutual influences in parent and adolescent psychological distress. Dev Psychol 31(3):406–419. https ://doi.org/10.1177/01925 13X14 55305 5

31. Gross HE, Shaw DS, Moilanen KL (2008) Reciprocal associations between boys’ externalizing problems and mothers’ depressive symptoms. J Abnorm Child Psychol 36(5):693–709. https ://doi. org/10.1007/s1080 2-008-9224-x

32. Gross HE, Shaw DS, Moilanen KL, Dishion TJ, Wilson MN (2008) Reciprocal models of child behavior and depressive symp- toms in mothers and fathers in a sample of children at risk for early conduct problems. J Fam Psychol 22(5):742–751. https :// doi.org/10.1037/a0013 514

33. Gross HE, Shaw DS, Burwell RA, Nagin DS (2009) Transactional processes in child disruptive behavior and maternal depression: a longitudinal study from early childhood to adolescence. Dev

Psychopathol 21(1):139–156. https ://doi.org/10.1017/S0954 57940 90000 91

34. Baker BL, McIntyre LL, Blacher J, Crnic K, Edelbrock C, Low C (2003) Pre-school children with and without developmen- tal delay: behaviour problems and parenting stress over time. J Intellect Disabil Res 47(4/5):217–230. https ://doi.org/10.104 6/j.1365-2788.2003.00484 .x

35. Neece CL, Green SA, Baker BL (2012) Parenting stress and child behavior problems: a transactional relationship across time. Am J Intellect Dev Disabil 117(1):48–66. https ://doi. org/10.1352/1944-7558-117.1.48

36. Pelham WE, Lang AR, Atkeson B, Murphy DA, Gnagy EM, Greiner AR, Vodde-Hamilton M, Greenslade KE (1997) Effects of deviant child behavior on parental distress and alcohol con- sumption in laboratory interactions. J Abnorm Child Psychol 25(5):413–424. https ://doi.org/10.1023/A:10257 89108 958

37. Elgar FJ, McGrath PJ, Waschbusch DA, Stewart SH, Curtis LJ (2004) Mutual influences on maternal depression and child adjust- ment problems. Clin Psychol Rev 24(4):441–459. https ://doi. org/10.1016/j.cpr.2004.02.002

38. Kouros CD, Garber J (2010) Dynamic associations between maternal depressive symptoms and adolescents’ depressive and externalizing symptoms. J Abnorm Child Psychol 38(8):1069– 1081. https ://doi.org/10.1007/s1080 2-010-9433-y

39. Kazdin AE, Weisz JR (1998) Identifying and developing empiri- cally supported child and adolescent treatments. J Consult Clin Psychol 66(1):19–36. https ://doi.org/10.1037/0022-006X.66.1.19

40. Eyberg SM, Nelson MM, Boggs SR (2008) Evidence-based psy- chosocial treatments for children and adolescents with disruptive behavior. J Clin Child Adolesc Psychol 37(1):215–237. https :// doi.org/10.1080/15374 41070 18201 17

41. Bennett DS, Gibbons TA (2000) Efficacy of child cognitive- behavioral interventions for antisocial behavior: a meta-analysis. Child Fam Behav Ther 22:1–15. https ://doi.org/10.1300/J019v 22n01 _01

42. McCart MR, Priester PE, Davies WH, Azen R (2006) Differ- ential effectiveness of behavioral parent-training and cognitive- behavioral therapy for antisocial youth: a meta-analysis. J Abnorm Child Psychol 34(4):527–543. https ://doi.org/10.1007/s1080 2-006-9031-1

43. Matjasko JL, Vivolo-Kantor AM, Massetti GM, Holland KM, Holt MK, Dela Cruz J (2012) A systematic meta-review of evaluations of youth violence prevention programs: common and divergent findings from 25 years of meta-analyses and systematic reviews. Aggress Violent Behav 17(6):540–552. https ://doi.org/10.1016/j. avb.2012.06.006

44. Barlow J, Coren E, Stewart-Brown S (2002) Meta-analysis of the effectiveness of parenting programmes in improving maternal psychosocial health. Br J Gen Pract 52:223–233

45. Colalillo S, Johnston C (2016) Parenting cognition and affective outcomes following parent management training: a systematic review. Clin Child Fam Psychol Rev 19(3):216–235. https ://doi. org/10.1007/s1056 7-016-0208-z

46. Eichelberger I, Plück J, Hanisch C, Hautmann C, Jänen N, Döpfner M (2010) Effekte universeller Prävention mit dem Gruppenformat des Elterntrainings Triple P auf das kindliche Problemverhalten, das elterliche Erziehungsverhalten und die psychische Belastung der Eltern (Effects of the universal parent training program Triple P on child behavior problems, parenting strategies, and psycho- logical distress). Z Klin Psychol Psychother 39(1):24–32. https :// doi.org/10.1026/1616-3443/a0000 06

47. Kazdin AE, Wassell G (2000) Therapeutic changes in chil- dren, parents, and families resulting from treatment of children with conduct problems. J Am Acad Child Adolesc Psychia- try 39(4):414–420. https ://doi.org/10.1097/00004 583-20000 4000-00009

1192 European Child & Adolescent Psychiatry (2018) 27:1181–1192

1 3

48. Kazdin AE, Bass D, Siegel T, Thomas C (1989) Cognitive- behavioral therapy and relationship therapy in the treatment of children referred for antisocial behavior. J Consult Clin Psychol 57(4):522–535. https ://doi.org/10.1037/0022-006X.57.4.522

49. Kazdin AE, Siegel T, Bass D (1992) Cognitive problem-solving skills training and parent management training in the treatment of antisocial behavior in children. J Consult Clin Psychol 60(5):733– 747. https ://doi.org/10.1037//0022-006X.60.5.733

50. DeGarmo DS, Patterson GR, Forgatch MS (2004) How do out- comes in a specified parent training intervention maintain or wane over time? Prev Sci 5(2):73–89. https ://doi.org/10.1023/ B:PREV.00000 23078 .30191 .e0

51. Wong JJ, Gonzales NA, Montano Z, Dumka L, Millsap RE (2014) Parenting intervention effects on parental depressive symptoms: examining the role of parenting and child behavior. J Fam Psychol 28(3):267–277. https ://doi.org/10.1037/a0036 622

52. Görtz-Dorten A, Döpfner M (2010) Therapieprogramm für Kinder mit aggressivem Verhalten (THAV) (Therapy program for chil- dren with aggressive behavior (THAV)). Hogrefe, Göttingen

53. Goertz-Dorten A, Benesch C, Hautmann C, Berk-Pawlitzek E, Faber M, Lindenschmidt T, Stadermann R, Schuh L, Doepfner M (2015) Efficacy of an individualized social competence training for children with oppositional defiant disorders/conduct disorders. Psychother Res. https ://doi.org/10.1080/10503 307.2015.10945 87

54. Goertz-Dorten A, Benesch C, Berk-Pawlitzek E, Faber M, Haut- mann C, Hellmich M, Lindenschmidt T, Schuh L, Stadermann R, Doepfner M (2018) Efficacy of individualized social competence training for children with oppositional defiant disorders/conduct disorders: a randomized controlled trial with an active control group. Eur Child Adolesc Psychiatry. https ://doi.org/10.1007/ s0078 7-018-1144-x

55. Whisman MA (1993) Mediators and moderators of change in cog- nitive therapy of depression. Psychol Bull 114(2):248–265. https ://doi.org/10.1037/0033-2909.114.2.248

56. Card NA, Stucky BD, Sawalani GM, Little TD (2008) Direct and indirect aggression during childhood and adolescence: a meta- analytic review of gender differences, intercorrelations, and rela- tions to maladjustment. Child Dev 79(5):1185–1229. https ://doi. org/10.1111/j.1467-8624.2008.01184 .x

57. World Health Organization (2010) The ICD-10 classification of mental and behavioural disorders: Clinical descriptions and diag- nostic guidelines, 4th edn. World Health Organization, Geneva

58. Döpfner M, Görtz-Dorten A, Lehmkuhl G (2008) Diagnostik-Sys- tem für psychische Störungen im Kindes- und Jugendalter nach ICD-10 und DSM-IV, DISYPS-II (DISYPS-II: diagnostic system for psychiatric disorders in children and adolescents). Huber, Bern

59. Görtz-Dorten A, Döpfner M (2008) Diagnose-Checklisten aus dem Diagnostik-System für psychische Störungen im Kindes- und Jugendalter (DISYPS-II) – Gütekriterien und klinische Anwendung (Diagnostic-Checklists of the Diagnostic System for Mental Disorders in Childhood and Adolescence (DISYPS-II) – Psychometric criteria and clinical application). Klin Diagn Eval 1(4):378–394

60. American Psychiatric Association (2000) Diagnostic and statisti- cal manual of mental disorders, 4th edn. American Psychiatric Association, Washington

61. Görtz-Dorten A, Ise E, Hautmann C, Walter D, Döpfner M (2014) Psychometric properties of a german parent rating scale for oppo- sitional defiant and conduct disorder (FBB-SSV) in clinical and community samples. Child Psychiatry Hum Dev 45(4):388–397. https ://doi.org/10.1007/s1057 8-013-0409-3

62. Ise E, Görtz-Dorten A, Döpfner M (2014) Reliability and valid- ity of teacher-rated symptoms of oppositional defiant disorder and conduct disorder in a clinical sample. Psychopathology 47(5):312–318. https ://doi.org/10.1159/00036 2373

63. Lovibond PF, Lovibond SH (1995) The structure of negative emotional states: comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behav Res Ther 33(3):335–343. https ://doi.org/10.1016/0005- 7967(94)00075 -U

64. Köppe EC (2001) Glückliche Eltern – liebe Kinder? Auswirkun- gen von Partnerschaft und psychischer Symptomatik der Eltern auf das Verhalten ihrer Kinder. Unpublished doctoral dissertation, Technische Universität Carolo-Wilhelmina zu Braunschweig

65. Crawford JR, Henry JD (2003) The Depression Anxiety Stress Scales (DASS): normative data and latent structure in a large non-clinical sample. Br J Clin Psychol 42:111–131. https ://doi. org/10.1348/01446 65033 21903 544

66. Yu C-Y (2002) Evaluating cutoff criteria of model fit indices for latent variable models with binary and continuous outcomes. Unpublished doctoral dissertation, University of California, Los Angeles

67. Preacher KJ, Hayes AF (2008) Asymptotic and resampling strat- egies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods 40(3):879–891. https :// doi.org/10.3758/brm.40.3.879

68. Rucker DD, Preacher KJ, Tormala ZL, Petty RE (2011) Mediation analysis in social psychology: current practices and new recom- mendations. Soc Personal Psychol Compass 5:359–371. https :// doi.org/10.1111/j.1751-9004.2011.00355 .x

69. Hayes AF, Rockwood NJ (2016) Regression-based statistical mediation and moderation analysis in clinical research: observa- tions, recommendations, and implementation. Behav Res Ther. https ://doi.org/10.1016/j.brat.2016.11.001

70. Muthén LK, Muthén BO (2012) Mplus. 7.0 edn. Muthén & Muthén, Los Angeles

71. Hayes AF (2013) Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. The Guilford Press, New York

72. Enders CK (2001) The performance of the full information max- imum likelihood estimator in multiple regression models with missing data. Educ Psychol Meas 61(5):713–740. https ://doi. org/10.1177/00131 64401 61500 1

73. Johansson P, Høglend P (2007) Identifying mechanisms of change in psychotherapy: mediators of treatment outcome. Clin Psychol Psychother 14(1):1–9. https ://doi.org/10.1002/cpp.514

74. Kazdin AE (2009) Understanding how and why psychotherapy leads to change. Psychother Res 19(4–5):418–428. https ://doi. org/10.1080/10503 30080 24488 99

75. Kenny DA, Judd CM (2014) Power anomalies in testing media- tion. Psychol Sci 25:334–339. https ://doi.org/10.1177/09567 97613 50267 6

76. Katzmann J, Görtz-Dorten A, Hautmann C, Döpfner M (2018) Social skills training and play group intervention for children with oppositional defiant disorders/conduct disorder: mediating mecha- nisms in a head-to-head comparison. Psychother Res. https ://doi. org/10.1080/10503 307.2018.14255 59

77. Renk K, Oliveros A, Roddenberry A, Klein J, Sieger K, Roberts R, Phares V (2007) The relationship between maternal and paternal psychological symptoms and ratings of adolescent functioning. J Adolesc 30(3):467–485. https ://doi.org/10.1016/j.adole scenc e.2006.05.001

European Child & Adolescent Psychiatry is a copyright of Springer, 2018. All Rights Reserved.