Increased risks of infant mortality and of deaths due to congenital malformation in the offspring of male electronics workers

BACKGROUND: There is limited evidence on the association between paternal occupational exposure during preconception and infant mortality and deaths due to congenital malformation. This retrospective cohort study was conducted to determine if such an association existed among male workers employed in an electronics factory. METHODS: We linked the databases of labor insurance, birth registration, and national death registry, identified 7,202 male workers ever employed in this factory with 13,592 liveborn children and 81 deaths in the first year after, excluding 861 children with potential maternal exposure from the same workplace. Fathers employed in this factory during their preconceptional periods (3 months prior to the conception) were considered as exposed compared with those not employed during the same periods. Poisson regression models were constructed to adjust for potential confounding by child's sex, parity, multiple births, year of birth, parental age at delivery, and educational levels. RESULTS: Based on 24 exposed cases, the rate ratios (RRs) of infant mortality were increased to 5.06 (95% CI: 2.33-11.00) and 2.81 (95% CI: 1.44-5.51) among liveborn children whose fathers worked for >10 and 1-10 years, respectively, in this factory during preconception. Maternal delivery age less than 20 years, fathers with less than 10 years of education, and multiple births were associated with increased risks of infant mortality. When limited to 28 deaths with congenital malformation, Poisson regression model showed an increased risk for exposed pregnancies (RR 5 3.75; 95% CI: 1.29-10.94), especially among cardiac defects (RR = 5.06; 95% CI: 1.58-16.19). CONCLUSIONS: Our study suggests that paternal occupational exposures, possibly to organic solvents during preconception, might increase infant mortality and deaths due to congenital malformation, especially for cardiac defects. However, the small numbers of this study limited the generalization of its findings.