Objectives To study the relations between maternal cumulative exposure to extremely low frequency electromagnetic fields (ELF MF) and the risk of moderate prematurity and small for gestational age within the birth cohort Elfe.

Methods The Elfe study included 18,329 infants born at 33 weeks of amenorrhea or more in France in 2011 and was designed to follow children until 20 years of age. Gestational age and anthropometric data at birth was collected in medical records and small for gestational age was defined according to a French customised growth standard. During interviews, mothers were asked to report their job status during pregnancy. If employed, their occupation was coded according to the ISCO-88 classification and the date on which they stopped their work was recorded. Cumulative exposure to ELF MF during pregnancy was assessed, for both mothers who worked and those who didn’t work during pregnancy, using a recently updated job-exposure matrix (JEM). Cumulative exposure was considered as a categorical variable (<17.5, 17.5–23.8, 23.8–36.2, 36.2–61.6 or ≥61.6 µT-days), as a binary variable (<44.1 and ≥44.1 µT-days) and as a continuous variable. Associations were analysed by logistic regression, adjusting for mother’s lifestyle factors, sociodemographic characteristics and some mother’s medical history during and before pregnancy. Analyses were restricted to single births and to complete values for the pregnancy outcomes (n=16,733).

Results Cumulative exposure was obtained for 96.0% of the mothers. Among them, 37.5% were classified in the 23.8–36.2 µT-days category, but high exposures were rare: 1.3% in the ≥61.6 µT-days category and 5.5% in the ≥44.1 µT-days category. No significant association was observed between maternal cumulative exposure and moderate prematurity and small for gestational age.

Conclusion This large population-based study does not suggest that maternal exposure to ELF MF during pregnancy is highly associated with risk of moderate prematurity or small for gestational age.