Adherence to gestational diabetes screening guidelines: Retrospective audit in three public hospitals in Ethiopia

INTRODUCTION: Gestational diabetes mellitus (GDM) is a major contributor to adverse perinatal outcomes. Most guidelines recommend universal GDM screening using standardized testing protocols, yet implementation in low- and middle-income countries remains limited. This study evaluated GDM screening uptake and guideline-concordant testing in routine antenatal care in Addis Ababa, Ethiopia. METHODS: We conducted a facility-based, cross-sectional study of pregnant women receiving antenatal care in three public hospitals in Addis Ababa, Ethiopia. A sample of 1,050 antenatal records was abstracted for women who were ≥28 weeks of gestation and/or had delivered during March to May 2025. Using a structured checklist, we extracted maternal and obstetric characteristics, documentation of any GDM screening, risk-based indications for screening, gestational age at testing, screening methods, and GDM diagnosis. We described uptake and adherence to guideline-recommended GDM screening. Logistic regression was used to estimate crude odds ratios (COR) and adjusted odds ratios (AOR) with 95% confidence intervals (CI) for factors associated with being screened. RESULTS: The mean (±SD) age of participants was 27.7 (± 5.0) years. Approximately one-third (29.4%) were pregnant for the first time. Overall, 139/1050 women were screened for GDM, yielding a screening rate of 13.2%. Median gestational age at screening was 34 weeks [IQR 30-37]; only 8.6% of screened women were tested in the recommended 24-28-weeks window. The oral glucose tolerance test was the predominant method (87.8%). Screening was almost exclusively risk-based (99.3%), where testing was prompted by clinical indications including polyhydramnios, advanced maternal age, and a prior history of macrosomic birth. Among screened women, 57/139 (41.0%) were diagnosed with GDM. In adjusted analysis, facility site (hospital C), older maternal age, higher parity and earlier gestational age at first antenatal care were independently associated with higher odds of being screened. CONCLUSION: In three public hospitals in Addis Ababa, GDM screening uptake was low. The screening practice was also delayed, and risk-based. Being screened was associated with hospital site, older maternal age, higher parity, and earlier first antenatal care attendance. These findings suggest addressing patient-level factors and facility-tailored implementation strategies to achieve universal, timely GDM screening in routine antenatal care workflows.