Effect of parity on gestational age at delivery in multiple gestation pregnancies

Objective: To estimate the effect of parity on gestational age (GA) at birth in multifetal pregnancies. Study Design: Birth data from the public-access Matched Multiple Birth File produced by the National Center for Health Statistics from 1995 to 2000 were analyzed following IRB approval. GA, parity and demographic data were analyzed with parametric and nonparametric tests, including regression analysis, using SPSS. Result: Data from women with twin (n = 316 983), triplet (n = 11981), and quadruplet (n = 766) pregnancies were analyzed. A significantly higher proportion of nulliparous versus parous women were Caucasian (twins: 82 versus 77%; triplets: 91 versus 87%) and had more than 15 years of education (twins: 39 versus 24%; triplets: 55 versus 39%; quadruplets: 53 versus 35%). Mean GA was 5.6 days longer for twins, 5.4 days longer for triplets and 6.8 days longer for quadruplets born to parous versus nulliparous women. Caucasian and African-American parous women pregnant with twins or triplets delivered their babies at a later GA than their nulliparous counterparts at each level of education. GA at delivery increased as a function of age of the mother in nulliparous and parous women of twins or triplets, and at every age level, parous women delivered their babies at a later GA. A higher proportion of nulliparous women delivered before 24 weeks (twins: 2.9 versus 1.2%; triplets: 5.9 versus 2.5%; quadruplets: 8.3 versus 2.6%). The percentage of twins born at or after 32 weeks was 84.9% for nullipara and 90.1% for parous women; for triplets, corresponding figures were 61.4 and 69.6%; and for quadruplets the figures were 33.2 and 44.2%. The percentage of births at or after 36 weeks for nulliparous and parous women pregnant with twins was 54.8 and 63.2%, respectively. The majority of the gain in GA was observed between women who had no previous births and those who had one previous birth. In regression analysis, the effect of parity remained after controlling for demographic and risk factors known to affect GA. Conclusion: GA at delivery is significantly increased in parous women carrying a multifetal gestation after controlling for other factors that affect GA at birth.