VARIABILITY IN DIAGNOSTIC EVALUATION AND CRITERIA FOR GESTATIONAL DIABETES

OBJECTIVE - To determine the frequency of screening for gestational di abetes mellitus (GDM) among a population receiving regular prenatal ca re and to assess the extent to which National Diabetes Data Group (NDD G) criteria for the diagnosis of GDM are used by practicing obstetrici ans. RESEARCH DESIGN AND METHODS - We studied participants in the Nurs es' Health Study II, a large prospective cohort study of 116,678 nurse s aged 25-42 years in 1989. A total of 422 women who reported a first diagnosis of GDM between 1989 and 1991 were sent supplementary questio nnaires regarding diagnosis and treatment, and medical records were re quested for a subset of 120 to validate self-reported GDM and assess c riteria used for diagnosis. A sample of 100 women who reported a pregn ancy not complicated by GDM were sent questionnaires addressing GDM sc reening and prenatal care. RESULTS - Among a sample of 93 women who re ported a pregnancy not complicated by GDM and responded to the supplem entary questionnaire, 16 (17%) reported no glucose loading test, 69% o f unscreened women had one or more risk factors for GDM. Among a sampl e of 114 women who self-reported GDM in a singleton pregnancy and whos e medical records were available for review, a physician diagnosis of GDM was confirmed in 107 (94%). Records and supplementary questionnair es indicated that oral glucose tolerance tests (OGTTs) were per formed in 96 (86%) of these women. Of women with a physician diagnosis of GD M whose OGTT results were available, 25% failed to meet NDDG criteria for this diagnosis, although all had evidence of abnormal glucose home ostasis. CONCLUSIONS - Screening for GDM is not universal, even among a group of health professionals in whom screening prevalence is likely to be higher than in the general population. Diagnostic criteria for GDM among obstetricians in practice remain nonstandard despite NDDG re commendations. Better understanding of the implications of differing d egrees of glucose intolerance and of varying GDM screening and managem ent strategies is required to make policy recommendations for appropri ate and cost-effective care.