The diagnostic and prognostic performance of a selective screening strategy for gestational diabetes mellitus according to ethnicity in Europe

Context: The performance of standard selective screening strategies for gestational diabetes mellitus (GDM) may vary according to ethnicity. Objective: We aimed to evaluate the diagnostic and prognostic performance of a selective screening tool to determine whether it accurately predicts GDM and events in women of different ethnicities. The tool selectively screens based on patients having one or more of the following risk factors (RFs): body mass index >= 25 kg/m(2), age >= 35 years, family history of diabetes, and personal history of GDM or macrosomia. Design and Setting: We conducted an observational prospective study at a university hospital. Participants: We included 17 344 women of European (30.9%), North African (29.6%), Sub-Saharan African (22.2%), Caribbean (8.7%), Indian-Pakistani-Sri Lankan (5.5%), and Asian (3.3%) ethnicities who were without pregravid diabetes and had singleton deliveries (2002-2010). Main Outcome Measures: We universally screened GDM and GDM-related events (pre-eclampsia, birth weight >= 4000 g, or dystocia). Results: Independent of confounding factors, North African (odds ratio [OR], 1.35; 95% confidence interval [CI], 1.21-1.52; P < .001) and Indian-Pakistani-Sri Lankan (OR, 2.52; 95% CI, 2.13-3.00; P < .001) women had more GDM than Europeans, whereas Sub-Saharan African women had less (OR, 0.82; 95% CI, 0.71-0.94; P < .01). Having one or more RFs was associated with GDM among Europeans (OR, 1.45; 95% CI, 1.22-1.76), North African (OR, 1.33; 95% CI, 1.13-1.55), Sub-Saharan African (OR, 1.48; 95% CI, 1.20-1.83), and Caribbean (OR, 1.55; 95% CI, 1.12-2.14) women. Having one or more RFs was also associated with GDM-related events only in European (P < .01) and North African (P < .05) women, with the following incidences in Europeans: no GDM/no RF, 6.9%; no GDM/RF, 9.0%; GDM/no RF, 14.7%; and GDM/RF, 12.6%. Conclusion: Standard selective screening criteria were not predictive of GDM in women from India-Pakistan-Sri Lanka and Asia and were associated with GDM-related events only in European and North African women. However, the women with GDM, who were routinely treated, had a poor prognosis, even for those free of RFs. These results support universal screening, irrespective of ethnicity.