The association of socioeconomic status and lifestyle behaviours on mental health appears well-established in the literature, as several studies report that better socioeconomic status such as higher levels of disposable income and employment as well as practising healthy lifestyles can enhance mental well-being. However, the reliance on cross-sectional correlations and lack of adequate statistical controls are possible limitations. This study aims to add the evidence of longitudinal association to the literature by using Japanese representative longitudinal household panel data. We employed panel data analytical techniques such as the random-effects conditional logistic regression (RE-CLR) and the fixed-effects conditional logistic regression (FE-CLR) models with possible time variant confounders being controlled. Our sample was comprised of 14,717 observations of 3,501 individuals aged 22- 59 years for five waves of the Japanese Household Panel Survey. We confirmed many of the factors associated with mental health reported in existing studies by analysing crosssectional data. These significant associations are also longitudinal (within) associations estimated by the FE-CLR models. Such factors include unemployment, low household income, short nightly sleeping duration, and lack of exercise. However, we also found that several factors such as disposable income, living alone, and drinking habits are not significantly associated with mental health in the FE-CRL models. The results imply the reverse causality that poor mental health conditions cause lower disposal income, possibly due to the inability to exhibit higher productivity, but an increase in disposal income would not necessarily improve mental health conditions. In this case, aggressive policy interventions to increase the disposal income of people of lower socioeconomic backgrounds would not necessarily be effective to minimize health inequalities.
ASJC Scopus subject areas
- Biochemistry, Genetics and Molecular Biology(all)
- Agricultural and Biological Sciences(all)