The issue of parity and risk of pregnancy outcomes has been controversial for many decades. We conducted a study to evaluate the association between parity and pregnancy outcomes and to clarify the basis of the classification of risk based on parity.The study is a cross-sectional analysis on a large routinely collected data set of singleton births from 1992 to 1997 in New South Wales, Australia. Pregnancy outcomes were compared among parity groups.There was a significant association between different parity levels and pregnancy outcomes in terms of obstetric complications, neonatal morbidity, and perinatal mortality, after adjustment for potential confounders, including age, maternal smoking status, and socioeconomic status. In terms of obstetric complications, neonatal morbidity, and perinatal mortality, subjects can be classified into 3 groups according to parity: nulliparity, low multiparity (parity 1, 2, and 3), and grand multipara (parity 4 to 8). Compared with low multiparity, mothers and babies of nulliparity and grand multipara are at higher risk.This study supported the definition of grand multiparity that should start from parity 4 or the 5th baby. Attention should be paid to subjects of nulliparity and grand multiparity for their different problems.