Mapping fine-scale spatial risk patterns of gestational diabetes over time in Australia: a nationwide geospatial study.

OBJECTIVE: To examine the geospatial distribution of gestational diabetes mellitus (GDM) over time in Australia. DESIGN: An ecological study was conducted using data from the National Diabetes Services Scheme (GDM cases). Data at Statistical Area Level 2 (SA2) level, a medium-sized spatial unit, on population denominators (women who gave birth) were obtained from the Australian Bureau of Statistics. The spatiotemporal distribution of GDM was explored at the SA2 level over three periods: 2016-2017, 2018-2019 and 2020-2021. Hotspot and cluster analyses were undertaken using Getis-Ord Gi* and local Moran's I statistics. SETTING: A nationwide study in Australia was conducted between 2016 and 2021. PARTICIPANTS: Women diagnosed with GDM and those who gave birth were included. OUTCOME MEASURES: Age-standardised and crude incidence of GDM per SA2. RESULTS: During 2016-2021, 1 718 963 eligible women who gave birth in Australia were included. Hotspot areas of GDM were consistently observed in Victoria (Southwest and North Melbourne); Western Australia (South and Southwest Perth); Australian Capital Territory (ACT) (East and North Canberra); Queensland (North Brisbane) and New South Wales (West and Southwest Sydney and Southeast New South Wales). ACT (South Canberra), North Tasmania, Northern Territory (North Darwin) and Victoria (South East Melbourne) had new hotspot regions recorded in the last two consecutive study periods. CONCLUSION: GDM incidence varies by geographical area over time, with hotspots in specific regions suggesting the need for geographically targeted policy interventions to curb the growing burden of GDM.