In 1997, the Office of Management and Budget (OMB) issued ‘‘Revisions to the
Standards for the Classification of Federal Data on Race and Ethnicity,’’ which revised the
‘‘1977 Statistical Policy Directive 15, Race and Ethnic Standards for Federal Statistics and
Administrative Reporting’’ [38,39]. These documents specify guidelines for collecting,
tabulating, and presenting race and ethnicity data within the federal statistical system. The 1997
revised standards incorporated two major changes designed to reflect the changing racial and
ethnic profile of the United States. First, the revision increased from four to five the minimum
set of categories to be used by federal agencies for identification of race. The 1977 standards
required federal agencies to report race-specific tabulations using a minimum set of four single-
race categories: American Indian or Alaska Native (AIAN), Asian or Pacific Islander (API),
black, and white. The five categories for race specified in the 1997 standards are: AIAN, Asian,
black or African American, Native Hawaiian or Other Pacific Islander (NHOPI), and white. The
revised standards called for reporting Asian persons separately from NHOPI. The revised
standards also require federal data collection programs to allow respondents to select one or
more race categories.
Beginning with the 2000 decennial census, the U.S. Census Bureau collected race and
ethnicity data in accordance with the 1997 revised standards; however, the National Vital
Statistics System, which is based on data collected by the states, will not be fully compliant with
the new standards until all of the states revise their birth certificates to reflect the new standards.
Thus, beginning with the 2000 data year, the numerators (births) for birth rates are incompatible
with the denominators (populations); see ‘‘Population denominators.’’ To compute rates, it is
necessary to ‘‘bridge’’ population data for multiple-race persons to single-race categories. This
has been done for birth rates by race presented in this report. Once all states revise their birth
registration systems to be compliant with the 1997 OMB standards, the use of bridged
populations can be discontinued.
Forty-one states and the District of Columbia, representing 90 percent of all U.S. resident
births, reported multiple-race data in 2012. Multiple-race reporting areas include: California,
Colorado, Delaware, Florida, Georgia, Idaho, Illinois, Indiana, Iowa, Kansas, Kentucky,
Louisiana, Maryland, Massachusetts, Michigan, Minnesota, Missouri, Montana, Nebraska,
Nevada, New Hampshire, New Mexico, New York, North Carolina, North Dakota, Ohio,
Oklahoma, Oregon, Pennsylvania, South Carolina, South Dakota, Tennessee, Texas, Utah,
Vermont, Virginia (revised after January 1
, 2012), Washington, Wisconsin, Wyoming, the
District of Columbia, Guam, and Northern Marianas, all of which used the 2003 revision of the
U.S. Standard Certificate of Live Birth, as well as Hawaii and Rhode Island, which used the
1989 revision. Puerto Rico, which revised its birth certificate in 2005, reported race according to
the 1989 certificate revision. Slightly more than 2 percent of mothers in the states reported more
than one race . Prior to 2012, the multiple-race reporting states varied; 6 states reported more
than one race in 2003, 15 states in 2004, 19 states in 2005, 23 states in 2006, 27 states in 2007,
30 states in 2008, 33 states and D.C. in 2009, 38 states and D.C. in 2010, and 40 states and D.C.
in 2011. Data from the vital records of the remaining 9 states and 1 territory followed the 1977
OMB standards in which only a single race is reported . In addition, these areas also reported
the minimum set of four races as stipulated in the 1977 standards compared with the minimum of
five races for the 1997 standards.
To provide uniformity and comparability of data during the transition period before
multiple-race data are available for all reporting areas, bridging the responses of those who
reported more than one race to a single race is necessary. The bridging procedure for multiple-
race mothers and fathers is based on the procedure used to bridge multiracial population
estimates; see ‘‘Population denominators’’ . Multiple race is imputed to a single race (AIAN,
API, black, or white) according to the combination of races, Hispanic origin, sex, and age of the
mother or father indicated on the birth certificate. The imputation procedure is described in detail
Where race of the mother is not reported, if the race of the father is known, the race of the
father is assigned to the mother. When information is not available for either parent, the race of
the mother is imputed according to the specific race of the mother on the preceding record with a
known race of mother. In 2012, race of mother was imputed for 5.8% of births (by occurrence).
Beginning with the 1989 data year, NCHS started tabulating its birth data primarily by
race of mother. In 1988 and prior years, births were tabulated by race of child, which was
determined from the race of the parents as entered on the birth certificate.
Trend data by race shown in this report are by race of mother for all years beginning with
the 1980 data year. Text references to white births and white mothers or black births and black
mothers are used interchangeably for ease in writing.
Age of mother
Information on age of mother is available for the entire United States. Beginning with
the 1989 U.S. Standard Certificate of Live Birth, a “Date of birth” item replaced the “Age (at
time of this birth)” item. Not all states revised this item, and, therefore, the age of mother either
is derived from the reported month and year of birth or coded as stated on the certificate. In
2012, age of mother was directly reported by one state (Virginia, before April 2012) and territory
(American Samoa). This information is recommended to be reported directly by the mother.
See the “Mother’s Worksheet for the Child’s Birth Certificate” available
Imputation of age of mother-From 1964 to 1996, mother’s age was edited for ages 10-49
years. Births reported as occurring to mothers under age 10 or over age 49 were assigned the
mean age of mothers based on data from a previous year with the same race, Hispanic origin, and
total birth order (total of live births and fetal deaths). Beginning in 1997, age of mother is
imputed for ages 9 years or under and 55 years and over. This procedure was used through 2006
for births in states using the 1989 Revision of the U.S. Standard Certificate of Live birth
(unrevised). Beginning in 2003 for births occurring in states using the 2003 revision of the birth
certificate (revised), a slightly wider age range is used; age of mother is imputed for ages 8 years
or under and 65 years and over (mother’s age 9 years is recoded as 10 years and ages 55-64 years
are recoded to an age from 50-54 years). Starting in 2007, the same procedures are used for
states using the unrevised certificate. A review and verification of unedited data for several
years including 2007 showed that the vast majority of births reported as occurring to women
aged 50 years and older were to women aged 50-54 years. The numbers of births to women aged
50-54 years have been too small historically to compute age-specific birth rates. These births
have been included with births to women aged 45-49 years for computing birth rates.
Extreme values of age-Data for single year of age of mother 9-11 and 55-64 years are not
shown in the public use data files. Births to mothers 9-11 years are collapsed into the categories
“12 years or under;” births to mothers 50-64 years into the category “50-54 years.”
Populations for age-specific rates-Age–specific birth rates are based on populations of
women by age, prepared by the U.S. Census Bureau. In census years the decennial census
counts are used. In intercensal years, estimates of the population of women by age are published
by the U.S. Census Bureau in Current Population Reports. The 2012 population estimates are
from the 2010 Census as of July 1, 2012, from responses to questions on age at last birthday and
month and year of birth, with the latter given preference. In the 1960, 1970, 1980, 1990, and
2000 Census of Population, age was also derived from month and year of birth. Age in
completed years was asked in censuses before 1960. This was nearly the equivalent of the
question of the pre-1989 birth certificates, which the 1950 test of matched birth and census
records confirmed, by showing a high degree of consistency in reporting age in these two sources
. More recently, reporting of maternal age on the birth certificate was compared with
reporting of age in a survey of women who had recently given birth. Reporting of age was very
consistent between the two sources .
Median and mean age of mother --Median age is the value that divides an age
distribution into two equal parts, one-half of the values being less and one-half being greater.
Median ages of mothers for 1960 to the present have been computed from birth rates for 5–year
age groups rather than from birth frequencies. This method eliminates the effects of changes in
the age composition of the childbearing population over time. Changes in the median ages from
year to year can thus be attributed solely to changes in the age–specific birth rates. Trend data
on the median age are shown in Table 1-5 of “Vital Statistics of the United States, 2003, Volume
1, Natality” , which is available at: http://www.cdc.gov/nchs/products/vsus.htm#natab2003.
Trend data on the mean age of mother, derived directly from frequencies of births by age,
are available at: http://www.cdc.gov/nchs/products/vsus.htm#natab2003, and for recent years, in
Table I-1, available at http://www.cdc.gov/nchs/data/nvsr/nvsr62/nvsr62_09_tables.pdf.
Not stated age or date of birth of mother --In 2012, age of mother was not reported on
0.01 percent of the records. Beginning in 1964, birth records with date of birth of mother and/or
age of mother not stated have had age imputed according to the age of mother from the previous
birth record of the same race and total-birth order (total of fetal deaths and live births). (See
NCHS Instruction Manuals, Part 12) [45, 46].
Age of father
Information on age of father is available for the entire United States
It is derived from
the father’s date of birth and is recommended to be reported directly by the mother. See the
“Mother’s Worksheet for the Child’s Birth Certificate”
at: http://www.cdc.gov/nchs/data/dvs/momswkstf_improv.pdf. Information on age of father is
often missing for children born to unmarried mothers, greatly inflating the number in the “Not
stated” category in all tabulations by age of father. If the age is under 10 years, it is considered
not stated and grouped with those cases for which age is not stated on the certificate. The
percent of records for which father’s age is not stated is shown in Table B.
In computing birth rates by age of father, births tabulated as age of father not stated are
distributed in the same proportions as births with known age within each 5–year-age
classification of the mother. This procedure is followed because, while father’s age is missing
on 12.8 percent of all births in 2012, age of father is missing for one-third (32.0%) of births to
teenaged (aged 15-19 years) mothers. This distribution procedure is done separately by race.
The resulting distributions are summed to form a composite frequency distribution that is the
basis for computing birth rates by age of father. This procedure avoids the distortion in rates that
would result if the relationship between age of mother and age of father were disregarded. Births
with age of father not stated are distributed only for rates, not for frequency tabulations.
Information on marital status is available for the entire United States. It is recommended
to be reported directly by the mother. See the “Mother’s Worksheet for the Child’s Birth
Certificate” at: http://www.cdc.gov/nchs/data/dvs/momswkstf_improv.pdf. National estimates
of births to unmarried women are based on two methods of determining marital status. For 1994
through 1996, birth certificates in 45 states and the District of Columbia included a question
about the mother's marital status. For the other states, marital status is inferred from information
on the birth certificate. Beginning in 1997, the marital status of women giving birth in California
and Nevada was determined by a direct question in the birth registration process. New York
City also changed its procedures for inferring marital status in 1997 to the same procedures in
effect in New York State, a separate registration area. Beginning June 15, 1998, Connecticut
discontinued inferring the mother’s marital status and added a direct question on mother’s
marital status to the state’s birth certificate. Michigan added a direct question in 2005 to the
birth registration process, but used inferential procedures to update information collected using
the direct question. Beginning in 2007, Michigan added a direct question on mother’s marital
status to the state’s birth certificate.
Inferential procedures-In 2012, inferential procedures were used to compile birth
statistics by marital status in full or in part for New York. In New York, a birth is inferred as
nonmarital if either of these factors, listed in priority-of-use order, is present: a paternity
acknowledgment was received or the father’s name is missing. In recent years, a number of
states have extended their efforts to identify the fathers when the parents are not married in order
to enforce child support obligations. The presence of a paternity acknowledgment, therefore, is
the most reliable indicator that the birth is nonmarital in the states not reporting this information
directly; this is now the key indicator in the nonreporting states. Details of the changes in
reporting procedures and the impact of the procedures on the data are described in previous
Imputation of marital status-The mother’s marital status was not reported in 2012 on 0.04
percent of the birth records in the 49 states, the District of Columbia, and New York City where
this information is obtained exclusively by a direct question. Marital status was imputed for
these records. If status was unknown and the father’s age was known, then the mother was
considered married. If the status was unknown, and the father’s age unknown, then the mother
was considered unmarried. This represents a change from the procedures in effect for 2002 and
previous years. Prior to 2003, marital status for all records with marital status not reported was
imputed as “married.” Because of the small number of records affected (1,728 in 2012), the
change in imputation procedures had essentially no impact on measures of nonmarital births.
When births to unmarried women are reported as second or higher order births, it is not
known whether the mother was married or unmarried when the previous deliveries occurred
because her marital status at the time of these earlier births is not available from the current birth
Mother--Information on educational attainment of the mother is available for the revised
reporting area, representing 86.3 percent of all U.S. births in 2012. The following eight
educational categories are collected in a checkbox format (Figure 1):
grade or less
grade, no diploma
High school graduate or GED completed
Some college credit but no degree
The instructions are to check the box that best describes the highest degree or level of
school completed at the time of the delivery. This information is recommended to be reported
directly by the mother. See the “Mother’s Worksheet for the Child’s Birth Certificate,” available
at: http://www.cdc.gov/nchs/data/dvs/momswkstf_improv.pdf. See the NCHS manual for
detailed descriptions of editing and computation methods  and Documentation Table 2 for
2012 data. The percent of records for which mother’s education is not stated is shown in Table
Father--Information on educational attainment of the father is available for the revised
reporting area, representing 86.3 percent of all U.S. births in 2012. The question is identical to
that of the mother. From 1995-2008, NCHS did not collect information on the educational
attainment of the father. Information on education of father is often missing on birth certificates
of children born to unmarried mothers, greatly inflating the number in the “Not stated” category.
While the overall percentage of “not stated” records was 15.7 percent for the reporting area
(Table B), four areas (Delaware, the District of Columbia, South Carolina, and Wisconsin) had
more than 25 percent of records with missing data for this item.
Live-birth order and parity
Information on live-birth order and parity are available for the entire United States.
Live-birth order and parity classifications refer to the total number of live births the mother has
had including the 2012 birth. Fetal deaths are excluded.
Live-birth order indicates what number the present birth represents; for example, a baby
born to a mother who has had two previous live births (even if one or both are not now living)
has a live-birth order of three. Parity indicates how many live births a mother has had. Before
delivery, a mother having her first baby has a parity of zero, and a mother having her third baby
has a parity of two. After delivery the mother of a baby who is a first live birth has a parity of
one, and the mother of a baby who is a third live birth has a parity of three.
Live-birth order and parity are determined from two items on the birth certificate,
“Number of previous live births now living” and “Number of previous live births now dead.”
This information is recommended to be collected directly from the medical record using the
facility worksheet. Detailed instructions and definitions for these items are presented in the
Guide to Completing the Facility Worksheets for the Certificate of Live Birth and Report of
Fetal Death (2003 Revision) . See also the NCHS manual for detailed descriptions of editing
and computation methods . The percent of records for which live-birth order is not stated is
shown in Table B.
In computing birth rates by live-birth order, births tabulated as birth order not stated are
distributed in the same proportion as births of known live-birth order.
Information on birth interval is available for the revised reporting area, representing 86.3
percent of all U.S. births in 2012. Birth intervals are computed for all births of second or higher
order. The interval is computed from the infant’s date of birth (month and year) and the date of
the last live birth (month and year). In a plural delivery, the 2
and higher order birth within a
set is classified at an interval of 0-3 months. This information is recommended to be collected
directly from the medical record using the facility worksheet. Detailed instructions and
definitions are presented in the Guide to Completing the Facility Worksheets for the Certificate
of Live Birth and Report of Fetal Death (2003 Revision) . See the NCHS manual for detailed
descriptions of editing and computation methods . The percent of records for which birth
interval is not stated is show in Table B. See Documentation Table 11 for 2012 data.
Medical and Public Services Utilization
Information on the timing and number of prenatal care visits is available for the revised
reporting area, representing 86.3 percent of all U.S. births in 2012. This information is collected
from the question “Date of first prenatal visit” (with a checkbox for “no prenatal care”) and
“Total number of prenatal visits for this pregnancy.” The public use file includes the month of
the first prenatal visit (ranging from months 1-10 of the pregnancy) as well as the trimester of the
first prenatal visit (1
, or 3
Prenatal care information is recommended to be collected directly from the prenatal care
record using the facility worksheet. Detailed instructions and definitions for these items are
presented in the Guide to Completing the Facility Worksheets for the Certificate of Live Birth
and Report of Fetal Death (2003 Revision) . See also the NCHS manual for detailed
descriptions of editing and computation methods . The percent of records for which month
prenatal care began and number of prenatal visits are not stated are shown in Table B. See
Documentation Table 8 for 2012 data on month prenatal care began.
In general, data on prenatal care utilization based on the 2003 revised birth certificate are
substantially lower than those based on the 1989 birth certificate. For the first year revised
certificates are implemented, the percentage of women reported to begin care in the first
trimester typically falls in a jurisdiction by at least 10 percentage points .
Did mother get WIC food for herself during this pregnancy?
Information on receipt of WIC (The Special Supplemental Nutrition Program for
Women, Infants, and Children) food for the mother during this pregnancy is available for the
revised reporting area, representing 86.3 percent of all U.S. births in 2012. The WIC program,
run by the U.S. Department of Agriculture, is intended to help low-income pregnant women,
infants, and children through age 5 receive proper nutrition by providing vouchers for food,
nutrition counseling, health care screenings and referrals . This information is recommended
to be reported directly by the mother. See the “Mother’s Worksheet for the Child’s Birth
Certificate” for the wording of the question for the mother, available
at: http://www.cdc.gov/nchs/data/dvs/momswkstf_improv.pdf. See the NCHS manual for
detailed descriptions of editing and computation methods  and Documentation Table 5 for
2012 data. The percent of records for which WIC receipt is not stated is shown in Table B.
Information on obstetric procedures is available for the revised reporting area,
representing 86.3 percent of all U.S. births in 2012. Four obstetric procedures are separately
identified on the revised 2003 certificate in a checkbox format (Figure 1):
Documents you may be interested
Documents you may be interested