The purpose of this case-control study was to determine the relative importance of various predictors of newborn macrosomia, with particular reference to maternal constitutional factors and glucose intolerance of pregnancy. Macrosomia was defined by both absolute birthweight > or = 4,000 g and birthweight > or = 90th centile for gestational age. One thousand mother/newborn pairs [209 macrosomic (cases) and 791 non-macrosomic newborns (controls)] were recruited. Mothers with pre-gestational diabetes mellitus were excluded. Data on pregnancy and pregnancy variables were collected by review of prenatal, labour, and delivery and newborn assessment records and interview with the mother. Predictors that entered the stepwise multiple regression model in order of significance were: previous history of macrosomia, increasing maternal weight, nonsmoking status, multiparity, male newborn gender, gestational age of 40-42 weeks, North American Aboriginal ethnicity, maternal birthweight > 4,000 g, maternal height and maternal age < 17 years. Glucose screen positive/100-g oral glucose tolerance test (GTT) negative status was a significant predictor for macrosomia as defined by birthweight greater than the 90th percentile for gestational age, but not for absolute birthweight over 4,000 g. It was the least significant of all the factors examined. Treated gestational diabetes was not a significant predictor. By multivariate analysis, maternal constitutional factors are more powerful predictors of newborn macrosomia than maternal mild glucose intolerance. Treatment of mothers with GDM may be masking the effect of more pronounced carbohydrate intolerance.