RT Journal Article SR Electronic T1 O4A.2 Joint effects of night work and shift rotation on treated depression in a longitudinal cohort of manufacturing workers JF Occupational and Environmental Medicine JO Occup Environ Med FD BMJ Publishing Group Ltd SP A32 OP A32 DO 10.1136/OEM-2019-EPI.86 VO 76 IS Suppl 1 A1 Ferguson, Jacqueline A1 Elser, Holly A1 Costello, Sadie A1 Cullen, Mark A1 Eisen, Ellen YR 2019 UL http://oem.bmj.com/content/76/Suppl_1/A32.2.abstract AB Past research consistently identifies shift work – a type of scheduled work that includes both night work and schedule rotations – as risk factor for depression. However, relatively few studies have examined whether people working more nights or rotations are more likely to seek treatment for depression.A total of 5848 workers across 33 plants in the American Manufacturing Cohort (AMC) were followed 2003–2013. The first observed episode of treated depression was defined from insurance claims as two depression-related outpatient visits or two prescribed antidepressants within 365 days. Using detailed timeclock data, night work (≥3 hours between 23:00 and 6:00) was defined as the percent of shifts that included a night shift and was categorized into non-night work 0%; low >0%–30%; medium >30%–70%; and high 70%–100%. Shift rotations (≥6 hours between subsequent shift start times), were similarly defined and classified as none 0%, infrequent <10%, and frequent rotations≥10%. We examined the joint effects of night work and shift rotation on the time to first episode of treated depression using Cox proportional hazards regression.Shift workers were more likely to be treated for depression compared with non-rotating, non-night workers. Workers with medium exposure to night work had the highest association with treated depression compared with permanent non-night workers, and the joint effects were similar between those with frequent rotations (HR: 1.64, 95% CI: 1.00–2.67) and infrequent rotations (1.53, 0.99–2.37). Associations were slightly elevated among permanent night workers (1.27, 0.72–2.25) and workers with high exposure to night work regardless of rotations (1.36, 0.86–2.17).Recent shift work exposure may be associated with higher rates of treated depression, and night work, rather than rotation, may be driving the association. However, there may be selection out of night work due to underlying depression and we were unable to differentiate new-onset from preexisting disease.