Citation

Törn P, Pettersson E, Lichtenstein P, Anckarsäter H, Lundström S, Hellner Gumpert C, Larsson H, Kollberg L, Långström N, Halldner L. Eur. Child Adolesc. Psychiatry 2015; 24(9): 1049-1059.

Affiliation

Department of Medical Epidemiology and Biostatistics (MEB), Karolinska Institutet, Stockholm, Sweden, Peggy.torn@hotmail.com.

Copyright

(Copyright © 2015, Holtzbrinck Springer Nature Publishing Group)

DOI

10.1007/s00787-014-0658-0

PMID

25567653

Abstract

Bully victimization is a common problem among children with neurodevelopmental disorders, including attention deficit/hyperactivity disorder and autism spectrum disorder. Previous research was mostly cross-sectional and seldom accounted for co-morbid psychopathology, which makes it difficult to draw conclusions about causality and specificity of any association. Using a genetically informative prospective design, we investigated the association between various neurodevelopmental problems (NDPs) in childhood and bully victimization in adolescence, and the relative contributions of genetic and environmental factors to this association. We obtained parent-reports of NDPs at age 9/12 years and self-reported bully victimization at age 15 for 3,921 children participating in the The Child and Adolescent Twin Study in Sweden (CATSS). Structural equation modelling was used to control for NDP co-morbidity and bully victimization at baseline. Cholesky decomposition was used to analyse genetic and environmental contributions to observed associations. Because most of the NDPs were associated to later bully victimization, a common effect of all NDPs was summarized into a general NDP factor. Controlling for this general factor, only problems with social interaction and motor control uniquely predicted subsequent bully victimization in girls. General and unique associations were influenced by both genetic and unique environmental factors. NDPs in general and social interaction and motor problems in particular predicted later bully victimization. The longitudinal design and twin analyses indicated that these associations might be causal. Knowledge of these vulnerabilities may be important when designing risk assessment and prevention strategies.

Language: en