Objectives The association between near-misses/minor injuries and moderate/severe injuries has yet to be investigated longitudinally. This study aimed to examine the longitudinal association between near-misses/minor injuries and moderate/severe injuries by the presence/absence of depressive symptoms using 1-year follow-up data obtained from a nationally representative sample of workers in Japan.

Methods Of the 18 231 eligible participants at time 1 (T1), 12 127 who responded to the 1-year follow-up survey at time 2 (T2) (response rate: 66.5%; 4370 females and 7757 males; mean age (SD), 45.3 (10.5) years) were included in the analysis. Multivariate logistic regression analyses were performed with the presence/absence of moderate/severe injuries at T2 as the dependent variable.

Results In total, 36.4% of participants reported depressive symptoms at T1. During the follow-up period, 1.6% of participants reported moderate/severe injuries in industrial settings. After adjusting for relevant variables, participants who reported near-misses (OR=1.7 (95% CI, 1.3 to 2.4)) and minor injuries (OR=2.5 (95% CI, 1.3 to 4.7)) at T1 were more likely to have moderate/severe injuries at T2 compared to those who reported no near-misses/minor injuries. However, this association was stronger in participants who did not have depressive symptoms at T1 than in those who had depressive symptoms.

Conclusions While the predictive value of near-misses/minor injuries for the occurrence of moderate/severe injuries by the presence/absence of depressive symptoms should be cautiously interpreted, our findings suggest that the development and utilisation of near-miss/minor injury reporting systems may help reduce the likelihood of moderate/severe injuries among workers, especially those without depressive symptoms.