Introduction Pesticide exposure, before and during the pregnancy, was described as dangerous for pregnancy and child development. Few studies investigated occupational pesticide exposure of women. The purpose of this study was to develop an untargeted approach to assess pesticide exposure of women of childbearing age, involved in agricultural activities and lived on farm.

Methods Thirty-nine women, aged 50 years or less and included in a prospective cohort of farmers were interviewed twice, during enrollment (1997 to 2000) and at 10-years of follow-up (between 2007 and 2010). First morning urine and individual data, including agricultural activities and pesticide use on crops, cattle, inside livestock buildings or on courtyard, were collected at each timepoint. An upgradable list of 205 quantifiable pesticides (including 16 metabolites) was formulated. Urine samples were analysed by using HPLC – tandem mass spectrometry with limits of quantification between 0.01 and 0.5 µg/L.

Results Almost all of women lived and worked on crop-livestock farms. Very few of them applied pesticide on crops but an half used herbicides on courtyard or insecticides on animal. Overall, 78 samples were analysed and 102 pesticides or metabolites were detected in at least one sample with a median of 9 molecules in samples at enrollment (min-max: 3–37) and at follow-up (1–57). Pentachlorophenol was detected in 75% of samples (median for contaminated samples: 0.582 and 0.506 µg/L at enrollment and at follow-up respectively). Four others pesticides were detected in more than 50% of samples at enrollment or at follow-up: Fluazifop-butyl (0.010 and 0.012 µg/L), Dinoterb (0.146 and 0.303 µg/L), Imazamethabenz-methyl (0.353 and 0.269 µg/L), and Hydroxyatrazine (0.034 and 0.0266 µg/L).

Conclusion We detected, without significant change after 10 years, active ingredients rarely analysed in studies of women occupationally exposed and some of them have been described as potential endocrine disruptors.