Background: Mental symptoms are prevalent among populations, but their associations with premature mortality are inadequately understood. We examined whether mental symptoms contribute to cause-specific mortality among midlife employees, while considering key covariates. Methods: Baseline mail survey data from 2000-02 included employees, aged 40-60, of the City of Helsinki, Finland ( n = 8960, 80 % women, response rate 67 %). Mental symptoms were measured by the General Health Questionnaire 12-item version ( GHQ-12) and the Short Form 36 mental component summary ( MCS). Covariates included sex, marital status, social support, health behaviours, occupational social class and limiting long-standing illness. Causes of death by the end of 2013 were obtained from Statistics Finland ( n = 242) and linked individually to survey data pending consent ( n = 6605). Hazard ratios ( HR) and 95 % confidence intervals ( 95 % CI) were calculated using Cox regression analysis. Results: For all-cause mortality, only MCS showed a weak association before adjustments. For natural mortality, no associations were found. For unnatural mortality ( n = 21), there was a sex adjusted association with GHQ ( HR = 1.96, 95 % CI = 1.45-2.64) and MCS ( 2.30, 95 % CI = 1.72-3.08). Among unnatural causes of death suicidal mortality ( n = 11) was associated with both GHQ ( 2.20, 95 % CI = 1.47-3.29) and MCS ( 2.68, 95 % CI = 1.80-3.99). Of the covariates limiting long-standing illness modestly attenuated the associations. Conclusions: Two established measures of mental symptoms, i.e. GHQ-12 and SF-36 MCS, were both associated with subsequent unnatural, i.e. accidental and violent, as well as suicidal mortality. No associations were found for natural mortality due to diseases. These findings need to be corroborated in further populations. Supporting mental health through workplace measures may help counteract subsequent suicidal and other unnatural mortality among midlife employees.