Jump to main content or area navigation.

Contact Us

EPA's Report on the Environment

Preterm Delivery

  • Learn more about how to use this interactive exhibit
  • Save the complete indicator as a printer-friendly PDF
  • Download this image
  • Download data for this exhibit

Click the legend to turn layers on or off. Hover your mouse over the display to reveal data.


Preterm delivery is defined as delivery prior to 37 completed weeks of gestation. The shorter the gestational age of an infant, the more likely it is for that infant to suffer adverse effects. Preterm birth along with low birthweight is the second leading cause of infant death (see the Infant Mortality indicator), and preterm infants are at higher risk for health and developmental problems (IOM, 2006). 

The determinants of preterm births are not fully known and the causes are often multi-factorial. Maternal high-risk conditions (e.g., infertility problems, vaginal spotting, inadequate maternal weight gain), previous history, socioeconomic status, smoking, alcohol consumption, and multiple gestation pregnancy are some of the known risk factors for preterm delivery. Environmental contaminants (e.g., lead, environmental tobacco smoke, air pollution) are possible risk factors for preterm birth (CDC, 2014a), which continue to be studied to better understand the strength of their associations with preterm delivery.

This indicator presents the proportion of U.S. infants born prior to 37 weeks of gestation, based on natality data reported to the National Vital Statistics System (NVSS), which is maintained by the Centers for Disease Control and Prevention’s (CDC’s) National Center for Health Statistics (NCHS). The NVSS registers virtually all deaths and births nationwide, with data coverage from 1933 to 2014 and from all 50 states and the District of Columbia. Beginning in 2014, NCHS transitioned from using the last normal menses (LMP) for estimating newborn gestational age to the obstetric estimate of gestation at delivery (OE) due to increasing evidence showing that the OE measure has greater validity compared with the LMP-based measure (NCHS, 2015). Due to the change in the NCHS’s metric for assessing gestational age, data from 2014 are not directly comparable to data from 2013 and earlier, and are therefore, examined separately in this indicator (CDC, 2014b).

The data presented here on preterm delivery were based on singleton births only. This was done to eliminate the effect of multiple births. The data are presented across three maternal age groups (under 20 years, 20-39 years, and 40 years and older).

What the Data Show

Based on LMP data, the proportion of all infants defined as preterm rose more than 20 percent from 1990 to 2006, but declined 11 percent from 2007 to 2013 (NCHS, 2012, 2015a). OE-based summary data, which are publicly available for 2007 to 2014, show an 8 percent decline in the preterm birth rate over this time period (NCHS, 2015b). While both the OE- and LMP-based estimates show declines in preterm birth rates, these variations result in different measures of preterm birth rates, and are therefore described separately below.

Exhibit 1 shows an overall increase in preterm deliveries of singleton infants from 1995 (9.8 percent) to 2006 (11.1 percent), then decreasing to 9.7 percent by 2013. The largest overall increase (1.3 percent) between 1995 and 2013 occurred among mothers in the 40 and over age group, with the percent of preterm births ranging from 12.0 percent (1995 and 1996) to 13.9 percent (2006 and 2007) but still at 13.3 percent by 2013. For the 20-39 year old maternal group, preterm deliveries ranged from a low of 9.2 percent (1996) to a high of 10.7 percent (2006) and then decreased to 9.4 percent by 2013. Among those under 20 years of age, preterm deliveries peaked at 13.9 percent in 2006 and decreased to their lowest percentage of 12.2 percent in 2013 (Exhibit 1).

In 1995, the percent of singleton preterm births was highest among black mothers (16.4 percent), followed by American Indian (11.5 percent), Asian/Pacific Islander (9.2 percent), and white (8.5 percent) mothers. From 1995 to 2013 preterm delivery among these mothers has shifted up and down slightly, with small percent differences (2.2 percent or less) across races over time (Exhibit 1). Specifically, from 1995 to 2013 the percent of preterm births ranged from 8.5 percent (1995 and 1996) to 10.1 percent (2005 and 2006) for white mothers, 14.2 percent (2013) to 16.4 percent (1995) for black mothers, 11.0 percent (1996) to 13.0 percent (2006) for American Indian mothers, and 8.7 percent (2012 and 2013) to 9.6 percent (2006 and 2007) for Asian/Pacific Islander mothers. Preterm delivery for Hispanic mothers ranged from 10.1 percent (1995, 1996, and 2013) to 11.2 percent (2006 and 2007), compared to 9.6 (2013) and 11.1 (2006) percent for non-Hispanic mothers between 1995 and 2013. From 1995 to 2013, the percent of preterm births was higher among Hispanic mothers except in 2005 when the percent was 11.0 percent for both Hispanic and non-Hispanic mothers. In the most recent year using LMP-based data (2013), the percent of preterm births was 10.1 percent for Hispanic mothers and 9.6 percent for non-Hispanic mothers.

In 2014, the first year individual OE-based NVSS data files are available, the proportion of all infants defined as preterm was 7.7 percent. The percent of singleton preterm births was highest among mothers in the 40 and over age group (10.8 percent), followed by mothers under 20 years of age (8.8 percent) and then mothers in the 20-39 age group (7.6 percent). Preterm births were most prevalent among black mothers (10.9 percent), followed by American Indian (8.9 percent), white (7.1 percent), and Asian/Pacific Islander (6.9 percent) mothers. Also in 2014, the percent of preterm births was the same (7.7 percent) among Hispanic and non-Hispanic mothers.


  • The primary measure used to determine the gestational age of the newborn from 1995 to 2013 is the interval between the first day of the mother’s last normal menses (LMP) and the date of birth (CDC, 2013). This measurement is subject to error for reasons such as imperfect maternal recall or misidentification of the LMP because of postconception bleeding, delayed ovulation, or intervening early miscarriage. In 2014, the measure used to determine the gestational age of the newborn transitioned to the obstetric estimate of gestation at delivery (OE), which is also subject to uncertainty (e.g., a physician could over- or underestimate the gestational age). Problems with reporting gestational age persist and may occur more frequently among some subpopulations and among births with shorter gestations (CDC, 2014b).
  • To prevent confusion about the significance of any differences among data obtained using LMP- or OE-based measures, data queries for this indicator are separate for years 1995-2013 and 2014.

Data Sources

The data used for this indicator were public-use natality data (1995-2014) obtained from the CDC’s National Vital Statistics System (CDC, 2015).


This page provides links to non-EPA websites that provide additional information about this topic. You will leave the EPA.gov domain, and EPA cannot attest to the accuracy of information on that non-EPA page. Providing links to a non-EPA website is not an endorsement of the other site or the information it contains by EPA or any of its employees. Also, be aware that the privacy protection provided on the EPA.gov domain (see Privacy and Security Notice) may not be available at the external link. Exit EPA Disclaimer

You will need the free Adobe Reader to view some of the files on this page. See EPA's PDF page to learn more.

Jump to main content.