A retrospective cohort followed 326,115 mothers with singleton pregnancies in a large Californian health system from their first delivery after January 2007 until diabetes onset, loss of membership or December 2022, with a median follow-up of 6.2 years. Gestational diabetes complicated 45,935 of 442,225 pregnancies, including 4,842 recurrences.
Gestational diabetes was associated with a nearly sixfold higher risk of later diabetes (HR 5.87, 95% CI 5.67 to 6.08). Recurrence took that to almost tenfold (HR 9.91, 95% CI 9.14 to 10.75). Asian women had the highest rate of gestational diabetes at 20.18%, the highest recurrence rate at 53.28%, and the highest risk after recurrence (HR 11.74, 95% CI 9.64 to 14.30). Read that recurrence figure again: more than half of Asian women who had gestational diabetes once had it again.
This lands hard in Indian practice, where gestational diabetes is common, onset is younger and BMI thresholds are lower. The number to act on is not the hazard ratio but the recurrence rate. A woman with one affected pregnancy planning another is more likely than not to have it again, which makes the window after the first delivery the one that matters: previous gestational diabetes → 6–12 week postpartum glucose test → annual glucose and weight review → pre-pregnancy assessment → screening at the first antenatal visit of the next pregnancy. The postpartum test is the most reliably skipped step in that whole chain.
- Book the 6–12 week postpartum glucose test before the woman leaves the maternity service, not at a later visit she may not attend.
- Put gestational diabetes on the permanent problem list so it surfaces at every future consultation.
- For any woman with previous gestational diabetes planning a pregnancy, screen at the first antenatal visit rather than waiting for 24–28 weeks.
- Ask South Asian and other Asian women about gestational diabetes in every past pregnancy, not just the most recent one.
- At annual review check weight, blood pressure and glucose together — the risk that follows is cardiometabolic, not glycaemic alone.
The statistics, in plain English
These hazard ratios are very large by the standards of observational research, and their intervals are tight — 9.14 to 10.75 for recurrence — because the cohort is enormous. Size buys precision, not freedom from confounding: women who develop gestational diabetes twice differ from those who never do in weight, family history and age, and adjustment can only go so far. The comparison group is women with no gestational diabetes at all, which is a low-risk group, so the ratio looks larger than the absolute difference in any single year would. Median follow-up was only 6.2 years, so these are the early cases; the lifetime figures will be higher still.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for diabetes & endocrinology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free