Psychopathology and friendship in children and adolescents: disentangling the role of co-occurring symptom domains with serial mediation models

• Autor(a) principal: Manfro, Arthur Gus
• Data de Publicação: 2017
• Outros Autores: Pan, Pedro M. [UNIFESP], Gadelha, Ary [UNIFESP], Fleck, Marcelo, do Rosario, Maria C. [UNIFESP], Cogo-Moreira, Hugo [UNIFESP], Affonseca-Bressan, Rodrigo [UNIFESP], Mari, Jair [UNIFESP], Miguel, Euripedes C., Rohde, Luis A., Salum, Giovanni A.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UNIFESP
• Texto Completo: http://dx.doi.org/10.1007/s00787-017-0993-z; https://repositorio.unifesp.br/handle/11600/58290
• topic: Friendship; Psychopathology; Mediation

The consolidation of social friendship groups is a vital part of human development. The objective of this study is to understand the direct and indirect influences of three major symptomatic domains-emotional, hyperkinetic, and conduct-on friendship. Specifically, we aim to study if the associations of one domain with friendship may be mediated by co-occurring symptoms from another domain. A total of 2512 subjects aged 6-14 years participated in this study. Friendship was evaluated by the Development and Well-Being Assessment's friendship section. We evaluated two main constructs as outcomes: (1) social isolation and (2) friendship latent construct. Emotional, hyperkinetic, and conduct symptomatic domains were evaluated with the Strengths and Difficulties Questionnaire (SDQ). All SDQ domains were positively associated with social isolation and negatively associated with friendship latent construct in univariate analysis. However, serial mediation models showed that the association between conduct domains with social isolation was mediated by emotion and hyperkinetic domains. Moreover, the associations between emotional and hyperkinetic domains with friendship latent construct in non-isolated children were mediated by the conduct domain. Emotion and hyperkinetic domains were directly and indirectly associated with social isolation, whereas conduct was directly and indirectly associated with overall friendship in non-isolated children. Results suggest that interventions aimed to improve social life in childhood and adolescence may have stronger effects if directed towards the treatment of emotion and hyperkinetic symptoms in socially isolated children and directed towards the treatment of conduct symptoms in children with fragile social connections.