Marriage as a social tie in the relation of depressive symptoms attributable to air pollution exposure among the elderly

Abstract

Background: Air pollution is a risk factor for depression or depressive symptoms. However, few studies have examined an effect modifier as a protective factor against depressive symptoms associated with air pollution, including social support. Notably, less is known about a married relationship in the association between exposure to air pollution and depressive symptoms among the elderly.

Methods: This study included 2122 marrieds and 607 non-marrieds, recruited in 2014-2017 from different regions of South Korea. Depressive symptoms were measured by the Korean version of the Geriatric Depression Scale Short Form (SGDS-K). After adjustment for potential confounders using propensity score of being assigned to the marrieds, we examined the extent of whether the effects of exposure to air pollutants (PM10, PM2.5, and NO2) on depressive symptoms were different between marrieds and non-marrieds. Subgroup analyses by gender and residence area were also performed.

Results: Marrieds than non-marrieds were less likely to have depressive symptoms and had smaller SGDS-K associated with increased exposure to PM10 and PM2.5 concentrations, respectively. After stratification of subjects by gender and residence area, the interaction term appeared to be significant among men and the non-metropolitan group, indicating the protective effect of married relationships on depressive symptoms attributable to air pollution exposure in them.

Limitations: Although we adjusted the propensity score, our findings might be confounded by the contextual effect associated with married relationships.

Conclusions: A married relationship, as a social tie, may attenuate the effect of exposure to air pollution on depressive symptoms among the elderly. Nonetheless, additional research is worthwhile to explore the extent of other social relationships in the association between air pollution exposure and depressive symptoms.