**6.3 Results**

Figure 1 reports the prevalence rates of the CBCL-Externalizing problem scale and DSM-IV ADHD- and DSM-IV ODD-disorders for each assessment point. In this sample, the prevalence rate of *Externalizing* problems in the borderline and clinical range varied from 7.5 % (FU1) to 30.8 % (FU2). In primary school, about 22 % of the children were identified as having clinically significant externalizing behaviour problems. There was an increase over time in prevalence of 11 %. Significantly more children had externalizing problems during the primary school years than in the preschool years: 11.3 % vs. 30.8 % (FU2; *χ*2 (1) = 12.3, *p* < .001), 22.0 % (FU3; *χ*2 (1) = 16.3, *p* < .001, and 22.9 % (FU4; *χ*2 (1) = 10.4, *p* < .001). For *ADHD disorders*, prevalence rates between 6 % and 10 % were obtained. Chi-square analysis showed that the gain of prevalence in attention deficit-/hyperactivity disorder over time failed to reach statistical significance. Prevalence rates of *ODD disorder* ranged from 7.4 % (FU3) to 10.5 % (FU1). There were no significant increases in prevalence rates.

The incidence rates for the CBCL-Externalizing problems and DSM-IV ADHD- and DSM-IV ODD-disorders are shown in Figure 2. This represents the *cumulative incidence* for the time periods from first assessment (pre) to each follow-up (FU1-FU4). Incidence for externalizing problems ranged from 4.5 % to 22.5 %. The highest incidence was present in the period two years after the pre-assessment. Infants had an incidence of 15-16 %. Lower rates in the DSM-IV-ODD were found, ranging from 3.5 % to 6.1%. Incidence for oppositional defiant disorder decreased from kindergarten to primary school. These findings are comparable to the incidence of DSM-IV-ADHD. The lowest incidence was 2.3 % (pre-FU2), the highest incidence was 4.8 % (pre-FU4). In summary, the results indicate higher incidence rates on externalizing behaviour problems established by the empirical approach than externalizing behaviour problems based on diagnostic categories of DSM-IV.

The German Conduct Disorder Rating scale (FBB-SSV) is also part of the comprehensive Diagnostic System for Mental Disorders in Childhood and Adolescence (DISYPS-KJ)*.* The rating scale includes 23 items using the symptom criteria of both the ICD-10 and DSM-IV, as well as additional criteria (e.g. symptom onset). Studies have supported the instrument's psychometric properties, showing good reliability and validity in both clinical and nonclinical populations (Döpfner et al., 2008). Children were diagnosed with oppositional defiant disorder (ODD) if parents reported that four or more symptoms had persisted for at least 6 months. All questionnaire assessments were conducted at five assessment points:

Prevalence rates of the Externalizing problem scale during the preschool years were calculated for mothers based on the norms provided by Achenbach and Rescorla (2000). Prevalence rates were calculated as a proportion of children with *subclinical* behaviour (T = 60 - 63) and *clinically relevant* behaviour (T ≥ 64). From the 2-year follow-up, the CBCL 4-18 with representative German norms was used. Percentages of ADHD and ODD mental health problems were calculated for all assessment points. Incidence rates are based on cumulative incidence for time periods as follows: pre-FU1, pre-FU2, pre-FU3,

Figure 1 reports the prevalence rates of the CBCL-Externalizing problem scale and DSM-IV ADHD- and DSM-IV ODD-disorders for each assessment point. In this sample, the prevalence rate of *Externalizing* problems in the borderline and clinical range varied from 7.5 % (FU1) to 30.8 % (FU2). In primary school, about 22 % of the children were identified as having clinically significant externalizing behaviour problems. There was an increase over time in prevalence of 11 %. Significantly more children had externalizing problems during the primary school years than in the preschool years: 11.3 % vs. 30.8 % (FU2; *χ*2 (1) = 12.3, *p* < .001), 22.0 % (FU3; *χ*2 (1) = 16.3, *p* < .001, and 22.9 % (FU4; *χ*2 (1) = 10.4, *p* < .001). For *ADHD disorders*, prevalence rates between 6 % and 10 % were obtained. Chi-square analysis showed that the gain of prevalence in attention deficit-/hyperactivity disorder over time failed to reach statistical significance. Prevalence rates of *ODD disorder* ranged from 7.4 % (FU3) to 10.5 % (FU1). There were no significant increases in

The incidence rates for the CBCL-Externalizing problems and DSM-IV ADHD- and DSM-IV ODD-disorders are shown in Figure 2. This represents the *cumulative incidence* for the time periods from first assessment (pre) to each follow-up (FU1-FU4). Incidence for externalizing problems ranged from 4.5 % to 22.5 %. The highest incidence was present in the period two years after the pre-assessment. Infants had an incidence of 15-16 %. Lower rates in the DSM-IV-ODD were found, ranging from 3.5 % to 6.1%. Incidence for oppositional defiant disorder decreased from kindergarten to primary school. These findings are comparable to the incidence of DSM-IV-ADHD. The lowest incidence was 2.3 % (pre-FU2), the highest incidence was 4.8 % (pre-FU4). In summary, the results indicate higher incidence rates on externalizing behaviour problems established by the empirical approach than externalizing

behaviour problems based on diagnostic categories of DSM-IV.

pre-test, and 1, 2, 3 and 4 years after the first assessment (follow-up 1- 4).

**6.2.3 Statistical analysis** 

and pre-FU4.

**6.3 Results** 

prevalence rates.

Fig. 1. Prevalence rates of externalizing problems and DSM-IV-disorders

Fig. 2. *Incidence rates of externalizing problems and DSM-IV-disorders*

The developmental course of externalizing behaviour problems is presented in Table 2. Table 2 includes the developmental patterns of CBCL-Externalizing problems, DSM-IV ADHD- and DSM-IV ODD-disorders. Overall, we found three different patterns: the first group is *stable normal* – for each of the five assessments no borderline and clinical scores on the Child Behavior Checklist or no diagnoses of ADHD or no diagnoses of ODD were obtained. The second group of children is *temporary clinical* – for at least one assessment borderline and clinical scores on the Child Behavior Checklist or a diagnosis of ADHD or a diagnosis of ODD was seen. The third group (*stable clinical*) showed relevant externalizing symptoms on the Child Behavior Checklist or ADHD-diagnoses or ODD-diagnoses for at least four of five assessments.

On the CBCL Externalizing scale 61.0 % of the sample was stable normal; from preschool to primary school, mother's report resulted in T-scores < 60. In addition, 31.6 % of the children had deviant CBCL scores for at least one assessment point. The stable clinicalpattern (remaining deviant at least four times) occurred in 7.4 % of the sample. In

The Epidemiology of Child Psychopathology: Basic Principles and Research Data 155

for Europe in general. The findings of this study support the assumption that studies without a definition of impairment had significantly higher prevalence rates than those with a definition of impairment. Besides, the diagnostic approach on oppositional defiant disorder in the preschool and primary school years has yielded prevalence rates ranging from 7.4 to 10.5 %. These results are consistent with the lifetime prevalence of ODD reported by Nock et al. (2007), who found a rate of 10.2 %. Prevalence estimates in previous studies have yielded a wide range from 2 - 15% (e.g. Loeber, Burke et al., 2000). Our prevalence is concordant with those of another European study on preschool children. Furthermore, although considerable research exists on ADHD and conduct disorder, information

So far, no comparable studies reporting incidence rates on externalizing behaviour have been published. In the absence of sufficient comparison studies it is not yet clear whether the findings reported here can be generalized. Further research is urgently called for to answer this important question. Therefore, incidence rates and the developmental course of

On the CBCL about 7 % of preschool and primary school children showed a stable pattern of relevant externalizing problem behaviour. A recent study by van Lier et al. (2007) assessed the trajectories of parent-rated symptoms of conduct problems from age 4 to 18 years old also in a general population sample. In this broader age group slightly lower rates (4 - 5 %) of a high trajectory of ODD- and ADHD-symptoms were found. A thorough statistical analysis of trajectories through growth mixture modelling on a large sample size of Dutch children yielded these results. The discrepancies in findings from those in the present study were attributable to different age groups, data collection and recruitment procedures, and CBCL-versions. The results on the stable normal pattern are in line with data from the

The study had several strengths. First, it is one of the rare studies with preschool children conducted in a universal setting with a 4-year follow-up over that time span. Second, the time intervals between the assessments were shorter compared with other longitudinal studies examining the same topic. Third, we used two different diagnostic approaches to describe externalizing problems: one of the best-studied instruments for the evaluation of children´s psychopathology (Achenbach, 2009) and DSM-IV ADHD- and ODD-diagnostic criteria. In this context, the study met for the most part the methodological criteria previously suggested by Robins and Rutter (1990), since it investigated behavioural problems in a sample of the population assessed longitudinally through standardized procedures. The present study is not without limitations. A main limitation is the generalizability of findings. Our sample is relatively advantaged with only 1/3 of all potentially eligible families participating. This finding corresponds to the fact that the rates of families and children recruited for family-focused preventions are typically very low (e.g. Spoth & Redmond, 2000). When interpreting the results, it should be taken into account that the child mental health status was assessed by symptom checklist questionnaires and disorder rating scales. The use of maternal self-report on child behaviour ratings may have been affected by the mother's experience of stress, depressive symptomatology, or marital problems. Given a large number of subjects, the questionnaire approach is economical and offers useful information, but lacks the specificity and additional depth that structured psychiatric interviews might provide.

regarding ODD is limited.

externalizing problem behaviour are considered together.

literature (Bongers et al., 2003; Keiley et al., 2000).

comparison with rates of ADHD- and ODD-disorders in childhood, the percentage of children in the group "temporary clinical" was relatively high. Regarding the ADHDdisorders and as shown in Figure 3, 83.8 % of our sample was stable normal and showed no clinical relevant ADHD-symptoms over the course. Overall, 12 % of children met at least at one assessment time the criteria for a disorder of ADHD. For 3.7 % of the sample the *stable clinical* pattern was observed. These children met on at least four occasions the criteria for a DSM-IV ADHD disorder. We found similar results in terms of oppositional deviant disorder: Stable normal behaviour was seen in 80.1 % of the sample. The percentage of the *stable clinical* pattern (3.7 %) corresponds with that for the rate of attention deficit-/hyperactivity. Only in relation to the temporary clinical course was a slightly higher rate (16.2 % to 12.5 %) observed.


Table 2. Developmental course of externalizing problem (CBCL EXT) behaviour and DSM-IV-disorders. Data in percent.

In summary, the results showed that about 80 % to 84 % of the preschool children were stable normal with regard to the development of ADHD- and ODD-disorders. However, when only CBCL externalizing scores were taken into consideration the rate decreased from 80 % to 60 %. In about one third of the sample temporary clinical CBCL Externalizing scores were observed. In contrast to results for the diagnostic categories of DSM-IV these rates are two-fold higher. Regarding the stable clinical pattern from preschool to primary school, 4 % of the children fulfilled the criteria for an ADHD- or an ODD-DSM-IV disorder.
