Direct and indirect effects of childhood conditions on health at older ages: the case of Brazil

• Autor(a) principal: Raquel Randow Bahia
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Institucional da UFMG
• Texto Completo: http://hdl.handle.net/1843/48755; https://orcid.org/ 0000-0002-6086-7939
• topic: Health of the Elderly; Socioeconomic Status; Childhood; Education

The expression 'the long arm of childhood' refers to the long-term effects of childhood conditions on health in old age. According to this theory, inequalities start early in life and can have enduring health consequences across an individual’s life (FERRARO; SCHAFER; WILKINSON, 2016). There is still an open agenda to investigate the impacts of exposures in childhood and the role of mediators on health at older ages in developing countries, like Brazil. The general aim is to examine the association between childhood socioeconomic and health circumstances and health outcomes in old age, in Brazil. The present study is based on a nationally representative, population-based survey, the Brazilian Longitudinal Study of Aging (ELSI-Brazil) (LIMA-COSTA et al., 2018). The method applied is the Structured Equation Modeling (SEM). We proposed three additive models for each of the four health outcomes: self-rated health, functional limitation, chronic conditions and a latent variable that includes all three observed variables. Model I measures the direct effect of childhood circumstances on health at old ages. Model II includes the indirect effect of education in the association between childhood and health at old ages. Model III adds health behaviors (physical activity, alcohol consumption, smoking and private health plans) to the second model. All three models control for gender and age. We confirmed the existence of ‘the long arm of childhood in Brazil’. The results show that childhood health and socioeconomic status impact directly and positively health outcomes later in life. Moreover, the models for self-rated health, chronic conditions and the latent old age health variable support the indirect association through the mediating role of education. On the other hand, the models for functional limitations at old ages did not support this hypothesis. None of the models show a significant and indirect association between diseases during childhood and health at older ages. Regarding the theoretical models of life course epidemiology, our results support more than one theoretical/conceptual model to explain the determinants of health among the elderly throughout the life trajectory, Critical Period Model as well as Chains of Risk.