Marian MacDorman

1
Recent Trends in Infant Mortality in the United
States

• Marian MacDorman
• Division of Vital Statistics
• National Center for Health Statistics, CDC
• State Infant Mortality Collaborative Meeting
• Atlanta, December 8-9, 2006

2
US infant mortality rate 2000-2005
Source CDC/NCHS National vital statistics
system. 2000-2004 data are from the final
mortality file 2005 data is provisional.
3
United States international infant mortality
ranking

• Year US IMR Rank
• 1960 26.0 12
• 1980 12.6 19
• 1990 9.2 23
• 2000 6.9 28
• 2003 6.9 28

4
Neonatal, and postneonatal mortality rates, US,
2000-2004
Source CDC/NCHS National Vital Statistics
System, final mortality file.
5
Infant mortality rates by race US, 2000-2004
Source CDC/NCHS National Vital Statistics
System, final mortality file.
6
Infant mortality rates by race and Hispanic
origin of mother US, 1995, 2000, and 2003
Source CDC/NCHS National Vital Statistics
System, linked birth/infant death data set
7
Components of Infant Mortality

• The distribution of births by birthweight
• Birthweight-specific infant mortality rates (the
  mortality rate for infants at a given weight)

8
Percentage of preterm and low birthweight births
US, 2000-2005
Source CDC/NCHS National Vital Statistics
System, 2000-2004 are final birth data, 2005 data
are preliminary.
9
Percentage of very preterm and very low
birthweight births US, 2000-2005
Source CDC/NCHS National Vital Statistics
System, 2000-2004 are final birth data, 2005 data
are preliminary.
10
Infant mortality rates by birthweight U.S., 2003
grams
11
Percentage of live births and infant deaths by
birthweight in grams, 2003
12
Percentage of live births and infant deaths by
period of gestation in weeks, US, 2003
13
Infant mortality rates by birthweight, US, 2003

• Birthweight 2000 2001 2002
  2003
• lt1,500 244.3 244.4 250.8
  252.0
• 1,500-2,499 15.8 15.2 15.1
  15.0
• 2,500 2.5 2.4
  2.4 2.3

14
Percentage of live births and infant deaths by
birthweight in grams, 2003
15
Possible Factors Contributing to the 2001-2002
Increase in Very Low Birthweight Births

• Changes in reporting of births and/or fetal
  deaths Reporting changed during the 1980s and
  1990s, however little evidence of large changes
  in reporting from 2001-2002.
• Changes in the risk profile of births The
  increase occurred among non-Hispanic white,
  non-Hispanic black and Hispanic women. Most of
  the increase occurred among low-risk women aged
  20-34 years. Multiple births accounted for about
  ¼ of the increase however ¾ of the increase was
  among singletons.
• Changes in medical management of pregnancy
  There was stronger evidence that this played a
  role in the increase.

16
Percent of Total, Preterm, and Very Preterm
Births Delivered by Cesarean, US, 1996-2004
Very preterm
17
Can we estimate the potential effect of preterm
medical intervention on the percent of preterm
births?

• In 2004, 41 of preterm births were cesarean
  deliveries.
• For 7 states that revised their birth
  certificates 75 of preterm cesareans did not
  report a trial of labor.
• 41 X 75 31 of preterm deliveries were
  cesareans on women not in labor.
• Add the 14 of preterm births with induced labor.
• This yields an estimate of about 45 of preterm
  births were the result of medical intervention to
  deliver the infant early when the woman was not
  in labor.

18
Caveats

• Trial of labor is a new item and has had
  relatively limited validation, compared with
  other birth certificate items. Because hospitals
  are not as familiar with the item, it may be
  underreported.
• This item is only available from 7 states in
  2004. However, these 7 states are from every
  region of the country and, taken together, have
  similar rates of both cesarean section and
  preterm birth to the US total.
• Regardless of the exact level, it is clear that
  medical intervention does have an impact on the
  percent of preterm births.
• More data from additional states in 2005 and
  later will help to improve the precision of this
  estimate.

19
Percent of Total Deaths and Infant Mortality
Rates by Leading Causes United States, 2004
20
Conclusions

• The US infant mortality rate did not decline
  appreciably from 2000-2005
• The halt in the historic US infant mortality
  decline is due primarily to increases in the
  percent of very preterm and very low birthweight
  births.
• Increases in birthweight-specific infant
  mortality rates for very low birthweight infants
  have also played a role.

21
Conclusions (2)

• Increases in the percent of very low birthweight
  births occurred among all race/ethnic groups, and
  mostly among women aged 20-34, a group
  traditionally considered to be at low risk for
  poor pregnancy outcome.
• Traditional cause-of-death groupings
  substantially underestimate the impact of preterm
  birth on infant mortality rates
• Increases in preterm cesarean section and
  induction of labor have had a major impact on the
  increase in the preterm birth rate.