Background/aim Long-term exposure to traffic noise has been shown to be associated with cognitive function. Noise is associated with depressive symptoms, which in turn are adversely associated with cognitive performance. We aimed to analyse whether depressive symptoms mediate the association of long-term exposure to traffic noise assessed outdoors and indoors with cognitive performance.

Methods In a cross-sectional analysis cognitive function was assessed in 4086 participants at the first follow-up of German Heinz Nixdorf Recall study using five subtests. A global cognitive score (GCS) was calculated as a sum of all subtests. We assessed depressive symptoms using the Centre for Epidemiologic Studies Depression scale 15-item short form (CES-D). Long-term weighted 24 hour exposure to traffic noise was assessed according to the EU directive 2002/49/EC of the baseline residential addresses (LDEN), and corrected for type of window, bedroom orientation and ventilation (LDEN_IN). Multiple linear regression models adjusted for individual risk factors were calculated for the association of noise with cognitive function, with and without adjustment for depressive symptoms.

Results In the fully adjusted model, noise was negatively associated with cognitive subtests and with the GCS. For example, a 10 dB(A) increase in LDEN was associated with the GCS (β=-0.34 [95% confidence interval: −0.67; −0.01]). An association of indoor noise with cognitive performance was lower (i.e. for 10 dB(A) increase in LDEN_IN with GCS β=-0.18 [-0.25; −0.10]). Adjustment of noise for CES-D score ≥17 (indicating the presence of depressive symptoms) revealed an Odds Ratios of 1.22 [0.92; 1.62] and 1.04 [0.95; 1.14] for 10 dB(A) LDEN and LDEN_IN, respectively. However, the adjustment for CES-D score did not change the noise-cognitive outcome associations neither for outdoor nor for indoor noise (i.e. for LDEN and GCS β=-0.33 [-0.66; −0.01], and for LDEN_IN and GCS β=-0.18 [-0.25; −0.10]).

Conclusion We found associations between traffic noise and cognitive performance, which did not change after adjustment for depressive symptoms. Future longitudinal analyses might shed light on possible mediating pathways between noise, depression and cognition.