1989-91 VS 1999-01. BARBARA LUKE, VICTOR HUGO GONZALEZ-QUINTERO, University of Miami School of Nursing and Health Studies, Coral Gables, Florida, University of Miami School of Medicne, Obstetrics and Gynecology, Miami, Florida OBJECTIVE: Our objective was to evaluate the characteristics of late preterm twin births during 1989-91 and 1999-01 to determine if risk factors and pregnancy outcomes had changed using national vital statistics data. STUDY DESIGN: The study population included all liveborn twins of 34-36 weeks gestation from the 1989-91 and 1999-01 Birth Cohort Linked Birth/Infant Death Data Set: 99,127 twins in 1989-91 and 156,984 twins in 1999-01. Maternal and newborn characteristics were compared using 2 or t tests. Adjusted odds ratios (AORs) and 95% confidence intervals calculated the infant mortality risk, controlling for maternal age, race, parity, education, smoking, adequacy of prenatal care, and infant gender. RESULTS: Between the two time periods, there was a significant shift to older maternal ages (35, 11.5% to 20.3%), more education (college graduates, from 19.8% to 33.2%), and earlier prenatal care (first trimester, from 78.7% to 85.6%), but also increases in smoking (from 6.7% to 8.9%), diabetes (from 2.9% to 4.3%), pregnancy-associated hypertension (from 7.5% to 10.6%), and cesarean delivery (from 53% to 58.7%). The risk of infant mortality fell significantly by 38% between the two time periods (AOR 0.62, 95% CI 0.57-0.68), with reductions in both neonatal and postneonatal deaths. CONCLUSION: The contemporary risk profile of women delivering late preterm twins has changed since the early 1990s, with decreases in many risk factors, but increases in others. Advances in medical care have resulted in significant reductions in neonatal and infant mortality during this period.