The initial vanguard cohort of the U.S. National Children’s Study was a pregnancy and birth cohort study that sought to enroll some women prior to pregnancy, and to assess exposures early in pregnancy. During the recruitment phase (2009–2010), geographically based sampling was used to recruit women early in pregnancy and women not currently pregnant, not using contraception and heterosexually active. We assessed the following outcomes for women enrolled preconception and early in pregnancy: yield of births; demographic characteristics of births for different enrollment groups; time to pregnancy for preconception women; and the timing of study visits for exposure assessment. 1399 women were recruited into the initial vanguard cohort: 429 preconception (198 trying for pregnancy, and 231 not trying) and 970 already pregnant. There were 1135 pregnancies (81 % of women) and 922 newborns enrolled (81 % of pregnancies) through September 2012. Preconception women represented 30.6 % of women enrolled, and contributed 14.5 % of births. Among women who gave birth, and who had enrolled preconception trying for pregnancy, 67.3 % were white non-Hispanic, compared to 50.0 % of preconception women not trying for pregnancy, and 61.5 % of pregnant women. Women enrolled preconception who were trying for pregnancy had higher cumulative probability of pregnancy at one year compared to women not trying (adjusted 86 % versus 56 %). Of 165 women enrolled preconception who became pregnant, 19 % had a study visit within 30 days of conception. By 10.5 weeks after conception, 75 % of women enrolled preconception had completed a pregnancy study visit; for women enrolled pregnant, the 75 % threshold was reached at 28.4 weeks. There were demographic differences in births from women enrolled preconception trying for pregnancy, preconception not trying for pregnancy, or during pregnancy. Time to pregnancy was shorter for women actively trying for pregnancy. Most women enrolled preconception did not have exposure assessment within 30 days of conception, but they did have exposure assessment much earlier during pregnancy than women who enrolled during pregnancy.

One pathogen that commonly causes gastrointestinal illnesses from the consumption of contaminated food is Escherichia coli O157:H7. In 2011 in Germany, however, there was a prominent outbreak of bloody diarrhea with a high incidence of hemolytic uremic syndrome (HUS) caused by an atypical, more virulent E. coli O104:H4 strain. To facilitate the identification of this lesser-known, atypical E. coli O104:H4 strain, we wanted to identify phenotypic differences between it and a strain of O157:H7 in different media and culture conditions. We found that E. coli O104:H4 strains produced considerably more biofilm than the strain of O157:H7 at 37 °C (p = 0.0470–0.0182) Biofilm production was significantly enhanced by the presence of 5% CO₂ (p = 0.0348–0.0320). In our study on the innate immune response to the E. coli strains, we used HEK293 cells that express Toll-like receptors (TLRs) 2 or 4. We found that E. coli O104:H4 strains had the ability to grow in a novel HEK293 cell culture medium, while the E. coli O157:H7 strain could not. Thus, we uncovered previously unknown phenotypic properties of E. coli O104:H4 to further differentiate this pathogen from E. coli O157:H7.