Objectives The aim is to examine occupational noise exposure as a risk factor for depression, utilising noise exposure as an objective measure of distressing working conditions that circumvents reporting bias.

Method In a 7-year cohort study we followed 109 378 industrial workers and 45 613 financial workers from 2001 or first year of employment thereafter until 2007. At start and end of follow up we recorded mean, full-shift noise exposure levels by personal dosimeters for 1077 workers from randomly selected companies. We assumed a linear relation with calendar year and predicted exposure levels by trade and occupation since 1980 and calculated cumulative noise exposure. Danish national registries provided complete employment histories since 1980, psychiatric diagnoses (1977–2001), and redemption of anti-depressants (Selective Serotonin Reuptake Inhibitors, SSRI) (1994–2007). Workers with psychiatric diagnoses or use of anti-depressants before 2001 were excluded.

Results During follow-up we identified 7754 incident users of SSRIs. Among women, risk of starting SSRI medication increased by cumulative noise exposure level OR=1.02 (95% CI: 1.01–1.02) per dB(A)-year when adjusted for age, calendar year and socioeconomic status. When excluding white-collar workers no effect was seen among women and no effect of noise was apparent among men overall.

Conclusions These preliminary results do not provide strong evidence that occupational noise exposure is a risk factor for depression. The increased OR seen among all women can be explained by differences in socioeconomic status between the blue- collar industrial workers and the white-collar financial workers since no trends were apparent in internal analyses among blue-collar workers.