Infant macrosomia is a serious medical concern. Pregnant women who do not meet the specific diagnosis for gestational diabetes may still have glucose-mediated macrosomia. In Santa Barbara County all pregnant women are screened for gestational diabetes at 24-28 weeks with a 50-g, 1-hr glucose challenge test (GCT). All patients who fail this test are placed on a standard euglycemic diet (40% carbohydrate, 20% protein, 40% fat) and perform home glucose monitoring of fasting and postprandial glucose levels. The objective of this study was to examine the effectiveness of this treatment program in decreasing infant macrosomia, maternal and infant morbidity, maternal complications, and operative delivery. We studied 103 women who had a positive GCT, but a negative 100-g, 3-hr oral glucose tolerance test (OGTT). The women were randomly assigned to either experimental or control groups with experimental women receiving dietary counseling and home glucose monitoring instruction (HBGM). HBGM diaries were reviewed weekly by clinic nurses. All women had hemoglobin A1c (HbA1c) tests at 28 and 32 weeks. Maternal and fetal charts were reviewed to determine delivery type and complications, indications for cesarean section (C-section), and infant gestational age, gender, Apgar scores, birth weight, morbidities, and congenital anomalies. Of the 103 women, 5 women required insulin treatment, 1 woman had an abortion, and 14 women were indeterminate regarding compliance or were control women who received diet counseling and HBGM. The results are based on 83 women--48 control and 35 experimental. There were no significant differences between the groups for age, parity, or weight at 28-30 weeks or 37 weeks to delivery, or HbA1c at 28 weeks. HbA1c was significantly higher in control women at 32 weeks. Birth weight expressed in grams or as a percentile specific for gender, ethnicity, and gestational age was significantly higher in control infants. Birth weight was significantly correlated with maternal intake weight, weight at 28-30 weeks, and weight at delivery and with HbA1c at 32 weeks' gestation. There were no significant differences between groups for maternal complications. Groups were significantly different for mode of delivery with experimental women having more induced vaginal deliveries but fewer repeat C-sections than control women. Groups were not different for primary C-sections. Women who fail the GCT, but not the OGTT and thus do not receive the diagnosis of GDM are still at risk for delivering a macrosomic infant and operative delivery. Our program of treatment for all women who fail the GCT improves outcome by reducing infant birth weight and the number of cesarean sections.