Diabetes Type and Perinatal Adverse Outcomes Among Women in Pregnancy and Neonates in Japan: A Cohort Study Using a Large‐Scale Database

ABSTRACT Aims To compare maternal and neonatal outcomes across gestational diabetes, preexisting diabetes, overt diabetes in pregnancy and normoglycaemia in Japan. Materials and Methods Using the JMDC claims database, we conducted a retrospective cohort study of mother–neonate pairs from 2015 to 2024. Maternal glycaemic status was classified as normoglycaemia, gestational diabetes, preexisting diabetes or overt diabetes in pregnancy using diagnoses, prescriptions, procedures and timing. Multivariable logistic regression estimated adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for maternal pre‐eclampsia/eclampsia and severe neonatal complications (neonatal death, foetal distress or neonatal intensive care unit admission). Results Among 244 402 neonates from 205 594 mothers, gestational diabetes, preexisting diabetes and overt diabetes were identified in 20 116 (8.2%), 3287 (1.3%) and 901 (0.4%), respectively. Compared with normoglycaemia, gestational diabetes, preexisting diabetes and overt diabetes were associated with higher risks of maternal pre‐eclampsia/eclampsia, with aORs of 1.34 (95% CI: 1.21–1.48), 1.96 (1.64–2.33) and 1.89 (1.35–2.64), respectively. Corresponding aORs for severe neonatal complications were 1.45 (1.38–1.52), 1.95 (1.77–2.15) and 2.55 (2.14–3.03), respectively. Compared directly with preexisting diabetes, overt diabetes was associated with higher odds of severe neonatal complications (aOR: 1.31; 95% CI: 1.07–1.59), with no statistically significant difference detected in maternal pre‐eclampsia/eclampsia (aOR: 0.96; 95% CI: 0.67–1.40). Conclusions Maternal and neonatal risks varied by hyperglycaemic status. Overt diabetes was associated with a higher risk of severe neonatal complications than preexisting diabetes, with no statistically significant difference detected in maternal pre‐eclampsia/eclampsia. Further validation using longitudinal data around pregnancy is warranted.