Dataset used for the paper " Temporal Trends in Maternal Food Intake Frequencies and Associations with Gestational Diabetes: The Cambridge Baby Growth Study" published as a manuscript in Nutrients (Nutrients 2019, 11(11), 2822; https://www.mdpi.com/2072-6643/11/11/2822/htm) and containing data from the Cambridge Baby Growth Study. Uncompressed Microsoft 2013 (.xlsx) file (1,240 rows including header; 62 columns including one for a dummy identifier) containing data relevant to the publication collected as part of the Cambridge Baby Growth Study (data collection 2001-2019). All the study participants were recruited from pregnancy clinics at the Rosie Maternity Hospital, Cambridge (2001-2009) around week 15 of pregnancy. A research 75 g oral glucose tolerance test (after an overnight fast) was performed on participants around week 28 of pregnancy. Gestational diabetes was characterised according to International Association of Diabetes in Pregnancy Study Groups criteria. Insulin concentrations were measured using enzyme-linked immunosorbent assays and glucose by a standard laboratory technique. HOMA modelling was performed using the online HOMA calculator (https://www.dtu.ox.ac.uk/homacalculator/). The index of multiple deprivation was estimated from residential postcodes. Missing data are presented by empty cells. For further information about the study please contact Dr. Clive Petry (address below; email: cjp1002@cam.ac.uk). For further general information about the Cambridge Baby Growth Study please contact Dr. Clive Petry (Department of Paediatrics, Box 116, Addenbrooke's Hospital, Cambridge Biomedical Campus, Hills Road, Cambridge CB2 0QQ, U.K.; email: cjp1002@cam.ac.uk).