A revised birth weight reference for the United States.

Area of Science:

• Neonatalogy
• Public Health
• Biostatistics

Background:

• Current U.S. birth weight reference curves are based on 1991 data and use last menstrual period for gestational age estimation.
• The 2003 revision of the U.S. birth certificate includes an obstetric estimate of gestational age, often incorporating ultrasound data, offering a potentially more accurate measure.

Purpose of the Study:

• To generate updated birth weight curves for neonates in the United States.
• To base these curves on the obstetric estimate of gestational age as defined in the 2003 U.S. birth certificate.

Main Methods:

• Utilized National Center for Health Statistics data from 2011, including over 3.2 million live-born singleton neonates (500-6,000 g, no malformations).
• Constructed birth weight curves for gestational ages 24-42 weeks using the obstetric estimate of gestational age.
• Compared these 2011 curves with curves generated from 1991 data (over 3.6 million births) which relied solely on last menstrual period for gestational age.

Main Results:

• Birth weight percentile values were higher from 28 to 36 weeks of gestation in the 1991 dataset compared to the 2011 dataset.
• The 1991 reference curves overestimated birth weights for preterm neonates.
• For example, a 2,000 g birth weight was at the 50th percentile between 31-32 weeks in 1991, but shifted to 33-34 weeks in the 2011 data.

Conclusions:

• The study successfully generated revised birth weight curves for the United States.
• These updated curves, based on obstetric gestational age estimates, provide a more current and accurate national reference for newborn birth weights.