Prevalence and risk factors for child mental disorders in a population-based cohort of HIV-exposed and unexposed African children aged 7-11 years

SOURCE: European Child & Adolescent Psychiatry
OUTPUT TYPE: Journal Article
TITLE AUTHOR(S): T.J.Rochat, B.Houle, A.Stein, R.M.Pearson, R.M.Bland
Print: HSRC Library: shelf number 10404
HANDLE: 20.500.11910/12284

If you would like to obtain a copy of this Research Output, please contact Hanlie Baudin at


Despite being home to a large population of vulnerable children there is a dearth of population-based evidence on childhood mental disorders in sub-Saharan Africa. Parent and child mental health are rarely measured concurrently, despite potential for confounding with other risk factors, including parental HIV. Using the parent-report Child Behaviour Checklist (CBCL) we assessed children's mental health in a population-based cohort of 1536 HIV-negative children (31% HIV-exposed, 18% HIV-affected, 51% HIV-unexposed) aged 7-11 years. CBCL was scored using CBCL Rating-to-Score software. A binary indicator was determined using the clinical threshold of 65. We modelled mental disorders using logistic regression, including covariates associated with the mother, child, household, and parenting. Structural equation modelling techniques also derived continuous latent variables representing the underlying mental health and parent-relationship constructs. Prevalence of conduct disorders (11.8%) was high, regardless of HIV exposure, while HIV-affected children had increased odds of affective disorders. Maternal depression increased odds of externalising disorders; maternal anxiety was associated with affective and anxiety disorders. Mother-child relationship dysfunction increased odds of all disorders, including: affective [aOR = 5.1 (2.6-9.9)]; oppositional [aOR = 7.9 (4.0-15.5)]; conduct [aOR = 4.3 (2.6-7.2)] disorders. Food insecurity and male gender increased odds of somatic disorders; breastfeeding halved odds of conduct disorders. In the latent model, associations were substantially stronger for the mother-child relationship and externalising disorders (Oppositional 0.464 p < 0.001; Conduct 0.474 p = <0.001). Conduct disorders were high for all children regardless of HIV exposure. The mother-child relationship was strongly related to all child disorders, suggesting potential for concurrent interventions targeting child behaviours and the parent-child or mother-child relationship