Fig 1. The Study of Women, Infant Feeding and Type 2 Diabetes after GDM Pregnancy (SWIFT) cohort and study design.

(A) SWIFT prospective cohort: 1,035 women diagnosed with gestational diabetes mellitus (GDM) in 2008–2011 were enrolled at 6–9 weeks postpartum (baseline). Of the 1,035 participants, 1,010 were confirmed via 2-hour 75-g oral glucose tolerance test to be without diabetes at baseline, underwent annual 2-hour 75-g OGTTs for 2 years post-baseline, and subsequently had their electronic medical records searched for clinical diagnoses of diabetes. Up to 8 years post-baseline, a total of 178 (18%) women developed type 2 diabetes (T2D) after baseline. Out of those 178 incident cases, 113 women were diagnosed as having diabetes clinically during the 2 years of in-person follow-up (blood samples of 98 women were available at this time point). (B) Metabolomics: using Biocrates p180 kits, 188 metabolites were measured in fasting plasma samples from participants at baseline (173 incident T2D cases and 485 nested pair-matched non-T2D controls) and follow-up (98 T2D cases and 239 nested pair-matched non-T2D controls). (C) The bioinformatics analysis pipeline includes 3 integral parts: (1) prospective analysis to predict future diabetes and to profile metabolic changes associated with future T2D onset at baseline, (2) cross-sectional analysis of the follow-up samples to reveal T2D-associated metabolic pathways, and (3) longitudinal analysis to shape a trajectory of T2D progression.