Structure and clinical correlates of obsessive-compulsive symptoms in a large sample of children and adolescents: a factor analytic study across five nations
Research output: Contribution to journal › Journal article › Research › peer-review
D. R. M. A. Højgaard, E. L. Mortensen, T. Ivarsson, K. Hybel, G. Skarphedinsson, J. B. Nissen, R. Valderhaug, K. Dahl, B. Weidle, N. C. Torp, M. Grados, A. B. Lewin, K. H. Melin, E. A. Storch, L. H. Wolters, T. K. Murphy, E. J. S. Sonuga-Barke, P. H. Thomsen
The underlying structure of obsessive-compulsive disorder (OCD) remains to be confirmed in child and adolescent populations. In this paper we report the first factor analytic study of individual OCD items from Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCS). OCD symptoms were assessed using the CY-BOCS symptom checklist in a sample of 854 patients with OCD (7-18 years of age) recruited from clinics in five countries. Pooled data were subjected to exploratory and confirmatory factor analysis (CFA) to identify the optimal factor structure. Various models were tested for age and gender subgroups. Also, the invariance of the solution across age and gender was tested and associations with demographic and clinical factors were explored. A three-factor model provided the best-fit solution. It consisted of the following factors: (1) harm/sexual, (2) symmetry/hoarding, (3) contamination/cleaning. The factor structure was invariant for age and gender across subgroups. Factor one was significantly correlated with anxiety, and factor two with depression and anxiety. Factor three was negatively correlated with tic disorder and attention-deficit/hyperactivity disorder (ADHD). Females had higher scores on factor two than males. The OCD symptom structure in children and adolescents is consistent across age and gender and similar to results from recent child and adolescents although hoarding may not be a separate factor. Our three-factor structure is almost identical to that seen in early studies on adults. Common mental disorders had specific patterns of associations with the different factors.
|Journal||European Child & Adolescent Psychiatry|
|Number of pages||11|
|Publication status||Published - Mar 2017|