Related Experiment Video
Updated: Aug 10, 2026

Transcutaneous Microcirculatory Imaging in Preterm Neonates
Published on: December 31, 2015
Changes in infant mortality among extremely preterm infants: US vital statistics data 1990 vs 2000 vs 2010
1Department of Pediatrics, University of Texas Medical Branch, Galveston, TX, USA.
Insights
Infant mortality for extremely preterm infants significantly decreased between 1990 and 2000. However, little progress was made in reducing infant mortality from 2000 to 2010, except for those born at 25 weeks gestation.
Area of Science:
- Neonatalogy
- Perinatology
- Public Health
Background:
- Infant mortality rates for extremely preterm infants (22-28 weeks gestation) vary significantly by gestational age.
- Understanding trends in mortality over time is crucial for assessing care improvements and informing parents.
Purpose of the Study:
- To analyze infant mortality trends (birth to 365 days) for infants born between 22 and 28 weeks gestation in the US.
- To compare mortality rates across three decades (1990, 2000, 2010), adjusting for maternal and infant characteristics.
Main Methods:
- Utilized US vital statistics linked birth and infant death certificate files for 1990, 2000, and 2010.
- Included births from 22 to 28 weeks gestation, with outlier birth weights removed.
- Calculated infant mortality rates and adjusted odds ratios using logistic regression.
Main Results:
- Infant mortality at 22 weeks gestation was 904/1000 in 1990, 835/1000 in 2000, and 866/1000 in 2010.
- An approximate 50% adjusted reduction in mortality odds was observed between 1990 and 2000.
- No significant mortality reduction occurred between 2000 and 2010, except at 25 weeks gestation (aOR=0.81).
Conclusions:
- A significant decrease in infant mortality for extremely preterm infants occurred between 1990 and 2000.
- Progress in reducing mortality for this population has stagnated between 2000 and 2010.
- Further research is needed to identify factors limiting mortality reduction in recent years.
Objective:
Infant mortality among extremely preterm infants (22 to 28 weeks gestation) varies considerably by gestational age. The reduction in mortality over a 20-year period, when examined in gestational age week increments, may give a more precise estimate of progress or lack thereof in caring for these infants and provide information to better inform practitioners and parents of the risk of mortality among these small infants. The objective of this analysis is to examine infant mortality (birth to 365 days) by week of gestation for infants 22 to 28 weeks gestation comparing mortality rates, adjusting for maternal and infant birth characteristics, among US births for the years 1990, 2000 and 2010.
Study Design:
US vital statistics period-linked birth and infant death certificate files for the years 1990, 2000 and 2010 were used. Maternal and infant characteristics for births at 22 to 28 weeks were abstracted from the files. A trimming procedure was used to remove records that had birth weights that exceeded the interquartile range of birth weights for a given week of gestational age. Infant mortality rates were calculated, and adjusted odds ratios for mortality were generated using logistic regression models.
Result:
A total of 15,593 live births, 22 to 28 weeks gestation were available for the year 1990; 17,095 for the year 2000; and 14,721 for the year 2010. Infant mortality rates ranged from 904 per 1000 live births at 22 weeks gestation in 1990, to 835 in 2000, to 866 in 2010. Across all gestational age groups there was an adjusted reduction in the odds ratio for mortality of ~50% from 1990 to the year 2000. However, between 2000 and 2010 there was no significant reduction in infant mortality except at 25 weeks gestation (adjusted odds ratio=0.81, 95% confidence interval=0.70, 0.93).
Conclusion:
Despite a significant reduction in infant mortality among extremely preterm infants between the years 1990 and 2000, there has been little progress in reducing mortality between the years 2000 and 2010.
Related Concept Videos
Regression Toward the Mean
Prevalence and Incidence
Prevalence indicates the proportion of individuals in a population who have a specific disease or health condition at a...
Relative Risk
Life Tables
Actuarial Approach
Consider the example of a high-risk surgical procedure with significant early-stage mortality. A two-year clinical study is conducted,...
Applications of Life Tables

