Only 20/94 (21%) met cut-off criteria for ODD, and these represented the children clinically diagnosed with ODD as well as several other children with multiply comorbid mood, learning, developmental, and ADHD diagnoses.

Correlations (Tables 3a and 3b)

To investigate the relation between anxiety and oppositional dimensions, we recorded the CSI dimensional score for oppositional symptoms at the time of assessment (maximum possible score, 8) for each case. We then correlated this with generalized anxiety symptoms (maximum possible score, 8), using the dimensional score for generalized anxiety symptoms as a measure of global anxiety. The 2-tailed Pearson correlation between parent generalized anxiety and oppositional defiant symptom scores (Table 3a) was 0.356, highly significant at the 0.01 level. Although teacher oppositional scores tended to be lower on average, and the sample size was smaller, the correlation between anxiety scores and oppositionality was 0.387, significant at the 0.01 level (Table 3b).

Knowing that ADHD is correlated with ODD, we repeated the analysis, controlling for the presence of ADHD symptoms. For parent ODD and GAD scores controlled for either inattentive or hyperactive-impulsive ADHD symptoms or for total ADHD symptoms, the correlation remained significant at the 0.01 level (Table 3a). Similarly, for teacher-reported symptoms the correlation remained significant at the 0.01 level when controlled for inattentive symptoms and at the 0.05 level when controlled for either hyperactive-impulsive or total ADHD symptoms (Table 3b).

Sex effects were notable for teacher, but not for parent, oppositional symptom scores.

Mean parent and teacher scores for anxiety symptoms did not differ between boys and girls. Mean parent scores for ODD symptoms did not differ significantly between boys and girls (P = 0.71), whereas mean teacher ODD scores for boys were significantly higher than for girls (P = 0.017). Notably, only 3 of the 20 children with clinical cut-off levels for ODD symptoms on teacher ratings were girls, 2 of whom had MDD and 1 of whom had multiple anxiety disorders.

Discussion

Difficulties in managing the oppositional behaviours of children with anxiety are a common presenting concern at our mood and anxiety disorders clinic and have dominated discussion in parent groups run concurrently with cognitive-behavioural treatment groups for children with anxiety. Although resistance among these children to both psychological and pharmacologic treatment has been noted as a limiting factor in treatment studies, research on anxiety disorders in children has to date neglected this problem. Our survey of professionals in several outpatient clinics highlighted the apparent prominence of oppositionality as a clinical feature of children suffering from anxiety.

This study focused on preadolescents because this population has more pure anxiety disorders that are less likely to be complicated by depression, substance abuse, and adolescent behavioural problems. However, the preadolescent clinic population also provided diverse diagnoses that allowed comparison with children suffering primarily from anxiety. We examined anxiety symptoms rather than anxiety diagnoses because anxiety disorders were commonly mixed, and we expected that the actual level of anxiety symptoms would predict oppositional behaviours.