Longitudinal association of conduct and emotional problems with school exclusion and truancy: A fixed effect analysis of the UK Millennium Cohort Study

Aase Villadsen; Claire Cameron; John Evans; Jo Van Herwegen; Vivian Hill; Jane Hurry; Amelia Roberts; Dominic Wyse; Thure Johansen

doi:10.1111/camh.12681

Longitudinal association of conduct and emotional problems with school exclusion and truancy: A fixed effect analysis of the UK Millennium Cohort Study

Child Adolesc Ment Health. 2023 Nov 9. doi: 10.1111/camh.12681. Online ahead of print.

Authors

Aase Villadsen¹, Claire Cameron¹, John Evans¹, Jo Van Herwegen², Vivian Hill², Jane Hurry², Amelia Roberts², Dominic Wyse³, Thure Johansen⁴

Affiliations

¹ Social Research Institute, University College London, London, UK.
² Psychology and Human Development, University College London, London, UK.
³ Learning and Leadership, University College London, London, UK.
⁴ Treehouse Associates, Leicester, UK.

PMID: 37942837
DOI: 10.1111/camh.12681

Abstract

Background: There is a need for causally stronger research on the association between child mental health and school exclusion and truancy. This study examines school exclusion and truancy in relation to both conduct and emotional problems and considers these problems both as predictors and as outcomes of school exclusion and truancy.

Method: The sample included 15,236 individuals from the Millennium Cohort Study, a UK longitudinal birth cohort study. Conduct and emotional problems were assessed from childhood to adolescence (age 7, 11, 14 and 17 years), and reports of school exclusion and truancy were collected at age 11 and 14. Fixed effect analyses were used.

Results: Increases in conduct problems and emotional symptoms were associated with subsequent exclusion (OR 1.22, [95% CI 1.08-1.37] and OR 1.16, [1.05-1.29], respectively). Emotional symptoms, but not conduct problems, predicted truancy (OR 1.17, [1.07-1.29]). These estimates were similar for males and females. Exclusion was associated with an increase in conduct problems at age 14 (0.50, [0.30-0.69]), and for males, it was associated with an increase in emotional symptoms both at age 14 (0.39, [0.12-0.65]) and 17 (0.43, [0.14-0.72]). Truancy was associated with an increase in conduct problems at age 14 (0.41, [0.28-0.55]), and for females also at age 17 (0.22, [0.03-0.42]), and it was associated with increased emotional symptoms at age 14 (0.43, [0.25-0.62]) and 17 (0.44, [0.21-0.66]), which was similar for males and females.

Conclusion: Results indicate a bidirectional association between emotional symptoms and school exclusion and truancy, as an increase in these symptoms was associated with later truancy and exclusion, and emotional symptoms increased following both school events. For conduct problems, the association was bidirectional for school exclusion, but unidirectional for truancy as these symptoms did not lead to truancy, but an increase in conduct problems was observed after both exclusion and truancy.

Keywords: Millennium Cohort Study; School exclusion; conduct problems; emotional symptoms; truancy.

Grants and funding

Institute of Education, University of London