Short-term exposure to air pollution and depressive symptoms: Findings from the French CONSTANCES cohort

Major depressive disorders pose a substantial public health burden, with emerging evidence linking these disorders to air pollution.

To investigate the association between short-term exposure to air pollution and depressive symptoms, focusing on particulate matter with a diameter ≤ 2.5 μm (PM_2.5) and nitrogen dioxide (NO₂).

Data from the enrollment phase of the French CONSTANCES cohort (2012-2016) were analyzed cross sectionally. For each participant, we assigned the median concentrations of PM_2.5 and NO₂, estimated by the CHIMERE model, over the seven days preceding completion of the Centre of Epidemiologic Studies Depression (CES-D) scale to their residential addresses. We assessed depressive symptoms using the CES-D and defined clinically significant depressive symptoms (binary variable) with the validated threshold of 19. We used negative binomial regression models (yielding Incidence Rate Ratio (IRR) and 95% confidence interval (CI)) and logistic regression models (yielding Odds Ratio (OR) and 95% CI), for an interquartile range (IQR) increase in exposures, adjusted for two sets of confounders.

105,976 participants (mean age, 47.3 ± 13.7 years; 53% women) were included. The 7-day median PM_2.5 and NO₂ exposures were 8.4 µg/m³ (IQR: 5.9-12.2) and 2.2 µg/m³ (IQR: 1.3-3.5), respectively. Short-term exposures to PM_2.5 and NO₂ were positively associated with a higher CES-D score (IRR = 1.009 (95% CI: 1.003-1.014) and IRR = 1.009 (95% CI: 1.004-1.014), respectively) and with higher rates of clinically-significant depressive symptoms (OR = 1.027 (1.008-1.046) and OR = 1.031 (1.012-1.051), respectively). These results were consistent across sensitivity analyses. There was an interaction between BMI and exposure to NO₂, with a significant effect only among subjects with obesity.

Our results suggest that air pollution may affect mental health, even at a subclinical level, warranting further research especially during pollution peaks. However, this study could not establish causality.