Avaliação da associação entre religiosidade e a cognição de idosos no Brasil
Carregando...
Data
2023-02-28
Autores
Silva, Katherine Di Santi Corrêa da
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
RESUMO Objetivos: Avaliar a associação entre religiosidade e a cognição de idosos brasileiros e analisar possíveis mecanismos de mediação envolvidos. Metódos: Foram usados dados da linha de base do Estudo Longitudinal de Saúde do Idoso (ELSI-Brasil) de participantes com 60 anos ou mais. Com base em sete questões sobre religiosidade foram extraídas duas variáveis latentes relativas à religiosidade intrínseca e extrínseca por meio de análise fatorial exploratória. Foram testados oito modelos de equações estruturais com controle progressivo para 20 possíveis variáveis de confundimento e cinco modelos onde a associação entre as duas variáveis latentes e a cognição era mediada pelas seguintes variáveis: tabagismo, consumo de álcool, situação conjugal, sintomas depressivos, e frequência com que encontra amigos. Resultados: Foram incluídos 4607 participantes com mediana de idade (Intervalo Interquartil) de 68 (63 a 74) anos. O modelo com melhor grau de ajuste aos dados continha 20 covariáveis para controle de potenciais fatores de confundimento. Neste modelo foi observada associação significativa entre a religiosidade intrínseca e a cognição (β:0,09, IC95%:0,01 a 0,17, P=0,04), mas não entre a religiosidade extrínseca e o mesmo desfecho (β:-0,06, IC95%:-0,13 a 0,02, P=0,12). Não foram identificados efeitos de mediação relevantes para nenhuma das variáveis testadas. Conclusão: Foram encontradas evidências sugestivas de uma associação positiva modesta entre a religiosidade intrínseca e uma melhor performance cognitiva, a qual não parece ter sido mediada pelas variáveis relativas a hábitos de vida, contato social e saúde mental testadas.
ABSTRACT Aims: To evaluate the association between religiosity and cognition in Brazilian older adults and to analyse possible mediation mechanisms involved. Methods: We used data from the baseline of the Brazilian Longitudinal Study of Health of the Older People (ELSI-Brasil) for participants aged 60 years and older. Based on seven questions about religiosity, two latent variables related to intrinsic and extrinsic religiosity were extracted through exploratory factor analysis. Eight structural equation models were tested with progressive adjustment for 20 possible confounding variables and five models where the association between the two latent variables and cognition was mediated by the following variables: smoking, alcohol consumption, marital status, depressive symptoms, and how often they meet friends. Results: We included 4607 participants with a median age (Interquartile Range) of 68 (63 to 74) years. The model with the best fit to the data contained 20 covariates to control for potential confounding factors. In that model, a significant association was observed between intrinsic religiosity and cognition (β:0.09, 95%CI:0.01 to 0.17, P=0.04), but not between extrinsic religiosity and the same outcome (β :-0.06, 95%CI:-0.13 to 0.02, P=0.12). No relevant mediation effects were identified for any of the tested variables. Conclusion: We provide evidence suggesting a modest positive association between intrinsic religiosity and better cognitive performance, which does not seem to have been mediated by the variables related to life habits, social contact and mental health that we tested.
ABSTRACT Aims: To evaluate the association between religiosity and cognition in Brazilian older adults and to analyse possible mediation mechanisms involved. Methods: We used data from the baseline of the Brazilian Longitudinal Study of Health of the Older People (ELSI-Brasil) for participants aged 60 years and older. Based on seven questions about religiosity, two latent variables related to intrinsic and extrinsic religiosity were extracted through exploratory factor analysis. Eight structural equation models were tested with progressive adjustment for 20 possible confounding variables and five models where the association between the two latent variables and cognition was mediated by the following variables: smoking, alcohol consumption, marital status, depressive symptoms, and how often they meet friends. Results: We included 4607 participants with a median age (Interquartile Range) of 68 (63 to 74) years. The model with the best fit to the data contained 20 covariates to control for potential confounding factors. In that model, a significant association was observed between intrinsic religiosity and cognition (β:0.09, 95%CI:0.01 to 0.17, P=0.04), but not between extrinsic religiosity and the same outcome (β :-0.06, 95%CI:-0.13 to 0.02, P=0.12). No relevant mediation effects were identified for any of the tested variables. Conclusion: We provide evidence suggesting a modest positive association between intrinsic religiosity and better cognitive performance, which does not seem to have been mediated by the variables related to life habits, social contact and mental health that we tested.
Descrição
Palavras-chave
Cognição, Envelhecimento, Religião, Estudos transversais