Nature and sources of Data in these reports are based on information from
all death certificates filed in the 50 States and the District of Columbia.
The U.S. Standard Certificate of Death was revised in 1989; for additional
details see the 1989 revision of the U.S. standard certificates under reports
(16) and Technical Appendix of Vital Statistics of the United States, 1989,
Volume II, Mortality, part A (17).
Mortality statistics are based on information coded by the States
and provided to the National Center for Health Statistics (NCHS) through
the Vital Statistics Cooperative Program (VSCP) and from copies of the
original certificates received by NCHS from the State registration offices.
In 1996 all the States and the District of Columbia participated in this
program and submitted part or all of the mortality data for 1996 in electronic
data files to NCHS. The 42 States in the VSCP that submitted precoded medical
data for all deaths are Alabama, Alaska, Arkansas, California, Colorado,
Connecticut, Delaware, Florida, Georgia, Hawaii, Idaho, Indiana, Iowa,
Kansas, Louisiana, Maine, Mary-land, Massachusetts, Michigan, Minnesota,
Mississippi, Montana, Nebraska, Nevada, New Hampshire, New Mexico, New
York State (excluding New York City), North Carolina, North Dakota, Oklahoma,
Oregon, Pennsylvania, Rhode Island, South Carolina, South Dakota, Texas,
Utah, Vermont, Virginia, Washington, Wisconsin, and Wyoming. Of these 42
States, Maine, Montana, North Dakota, and Wyoming contracted with a private
company to provide precoded medical data to NCHS. The remaining eight States,
New York City, and the District of Columbia submitted copies of the original
certificates from which NCHS coded the medical data. For 1996 all States
submitted precoded demographic data for all deaths.
Data for the entire United States refer to events occurring within
the United States. Data shown for geographic areas are by place of residence.
Beginning with 1970 mortality statistics for the United States exclude
deaths of nonresidents of the United States. All data exclude fetal deaths.
Mortality statistics for Puerto Rico and Virgin Islands exclude deaths
of nonresidents of Puerto Rico and Virgin Islands, respectively. For Guam,
however, mortality statistics exclude deaths that occurred to a resident
of any place other than Guam or the United States.
Quality of race data-A number of studies have been conducted on the reliability
of race reported on the death certificate by comparing race on the death
certificate with that reported on another data collection instrument, such
as the census or a survey. Differences may arise because of differences
in who provides race information on the compared records. Race information
on the death certificate is reported by the funeral director as provided
by an informant or in the absence of an informant, on the basis of observation.
In contrast, race on the census or on the Current Population Survey (CPS)
is obtained while the individual is alive and is self reported or reported
by another member of the household familiar with the individual and, therefore,
may be considered more valid. A high level of agreement between the death
certificate and the census or survey report is essential to assure unbiased
death rates by race.
Results from several studies (18–20) show that a person self-reported
as American Indian or Asian on census or survey records was sometimes reported
as white on the death certificate. The net effect of misclassification
is an underestimation of deaths and death rates for races other than white
Other races and race not stated-Beginning in 1992 all records
coded as ‘‘Other races’’ (0.02 percent of the total deaths in 1996) were
assigned to the specified race of the previous record. Records for which
race was unknown, not stated, or not classifiable (0.08 percent) were assigned
the racial designation of the previous record.
Infant and maternal mortality rates For 1989–96, as in previous
years, infant and maternal deaths continue to be tabulated by the race
of the decedent. However, beginning with the 1989 data year, the method
of tabulating live births by race was changed from race of parents to race
of mother as stated on the birth certificate. This change affects infant
and maternal mortality rates because live births are the denominators of
these rates (21,22). To improve continuity and ease of interpretation,
trend data by race in this report have been retabulated by race of mother
for all years beginning with the 1980 data year. Quantitatively, the change
in the basis for tabulating live births by race results in more white births
and fewer black births and births of other races. Consequently, infant
and maternal mortality rates under the new tabulating procedure tend to
be about 2 percent lower for white infants and about 5 percent higher for
black infants than when they are computed by the previous method of tabulating
live births by race of parents. Rates for most other minority races also
are higher when computed by race of mother (17,22).
Infant mortality rates for specified race are biased because of
inconsistencies in reporting race between the birth and death certificates
for the same infant. Infant mortality rates by specified race are less
subject to reporting bias when based on linked files of infant deaths and
live births (15). The linked data set computes infant mortality rates using
the race of the mother from the birth certificate in the numerator and
denominator. Race information on the birth certificate is considered to
be more accurate than that on the death certificate because, on the birth
certificate, race is generally reported by the mother at the time of delivery.
Estimates of reporting bias have been made by comparing rates based on
the linked files with those in which the race of infant death is based
on information from the death certificate (17).
The mortality statistics presented here were compiled in accordance with
the World Health Organization (WHO) regulations, which specify that member
nations classify causes of death by the current Manual of the International
Statistical Classification of Diseases, Injuries, and Causes of Death (23).
Causes of death for 1979-96 were classified according to the manual. For
earlier years causes of death were classified according to the revisions
then in use 1968-78, Eighth Revision; 1958–67, Seventh Revision; and 1949-57,
Sixth Revision. Changes in classification of causes of death due to these
revisions may result in discontinuities in cause-of-death trends. Consequently,
cause-of-death comparisons among revisions require consideration of comparability
ratios and, where available, estimates of their standard errors. Comparability
ratios between the Eighth and Ninth Revisions, between the Seventh and
Eighth Revisions, and between the Sixth and Seventh Revisions may be found
in other NCHS reports (24-26). National Vital Statistics Report, Vol. 47,
No. 9, November 10, 1998 89
Besides specifying the classification, WHO regulations outline the form
of medical certification and the procedures to be used in coding cause
of death. Cause-of-death data presented in this publication were coded
by procedures outlined in annual issues of the NCHS InstructionManual (27–29).
Before data for 1968, mortality medical data were based on manual
coding of an underlying cause of death for each certificate in accordance
with WHO rules. Effective with data year 1968, NCHS converted to computerized
coding of the underlying cause and manual coding of all causes (multiple
causes) on the death certificate. In this system, called Automated Classification
of Medical Entities (ACME) (30), the multiple cause codes serve as inputs
to the computer software that employs WHO rules to select the underlying
cause. Many States have implemented ACME and provide multiple cause and
underlying cause data to NCHS in electronic form; for those States that
have not, NCHS coded the mortality medical data using ACME.
The ACME system is used to select the underlying cause of death
for all death certificates in the United States. In addition, NCHS has
developed two computer systems as inputs to ACME. Beginning with 1990 data,
the Mortality Medical Indexing, Classification, and Retrieval system (MICAR)
(31,32) was introduced to automate coding multiple causes of death. In
addition, MICAR provides more detailed information on the conditions reported
on death certificates than is available through the International Classification
of Diseases (ICD) code structure. Then, beginning with data year 1993,
Super MICAR, an enhancement of the MICAR system, was introduced. Super
MICAR allows for total literal entry of the multiple cause-of-death text
as reported by the certifier. This information is then automatically coded
by the MICAR and ACME computer systems. Records that cannot be automatically
processed by MICAR or Super MICAR are manually multiple cause coded and
then further processed through ACME.
For 1996 approximately 29 percent of the Nation's death records
were multiple cause coded using SuperMICAR and 71 percent, using MICAR
only. This represents data from 22 States that were coded by SuperMICAR
and data from 28 States, the District of Columbia, and New York City that
were coded by MICAR.
In this report tabulations of cause-of-death statistics are based
solely on the underlying cause of death. The underlying cause is defined
by WHO as the disease or injury that initiated the sequence of events leading
directly to death or as the circumstances of the accident or violence that
produced the fatal injury. It is selected from the conditions entered by
the physician in the cause-of-death section of the death certificate. When
more than one cause or condition is entered by the physician, the underlying
cause is determined by the sequence of conditions on the certificate, provisions
of the ICD, and associated selection rules. Generally, more medical information
is reported on death certificates than is directly reflected in the underlying
cause of death.
Codes for HIV infection
Beginning with data for 1987, NCHS introduced categories *042-*044
for classifying and coding Human immunodeficiency virus (HIV) infection.
The asterisks before the category numbers indicate that they are not part
of the Ninth Revision, International Classification of Diseases (ICD-9).
Deaths classified to HIV infection are included, but not shown separately,
in the category ‘‘All other infectious and parasitic diseases’’ in the
List of 72 Selected Causes of Death and in the category ‘‘Remainder of
infectious and parasitic diseases’’ in the List of 61 Selected Causes of
Infant Deaths. Before 1987 deaths involving HIV infection were classified
to ‘‘Deficiency of cell-mediated immunity’’ (ICD-9 No. 279.1), contained
in the category ‘‘All other diseases’’; to ‘‘Pneumocystosis’’ (ICD-9 No.
136.3), contained in the category ‘‘All other infectious and parasitic
diseases’’; to ‘‘Malignant neoplasms, including neoplasms of lymphatic
and hematopoietic tissues’’; and to a number of other causes. As a consequence
cause-of-death data beginning with 1987 are not strictly comparable with
data for previous years.
Alzheimer's disease deaths
Beginning with data year 1979, WHO introduced a separate category
for Alzheimer's disease in the ICD-9 (331.0). Alzheimer’s disease mortality
is believed to be underreported. Research shows that a substantial number
of death certificates for persons diagnosed with dementia do not mention
the condition. Underreporting is believed to be due to failure to report
a condition considered to contribute to death, judgment that this condition
did not contribute to death, or failure to diagnose the condition (33).
Codes for firearm deaths
Causes of death attributable to firearm mortality include ICD-9No.
E922, Accident caused by firearm missile; Nos. E955.0-E955.4, Suicide and
self-inflicted injury by firearms; E965.0-E965.4 and E970, Assault by firearms
and legal intervention; and E985.0-E985.4, Injury by firearms, undetermined
whether accidentally or purposely inflicted. Injury by firearm causes exclude
explosives and other causes in directly related to firearms.
Codes for drug induced deaths
Causes of death attributable to drug induced mortality include ICD-9
No. 292, Drug psychoses; No. 304, Drug dependence; Nos.305.2-305.9, Nondependent
use of drugs not including alcohol and tobacco; Nos. E850-E858, Accidental
poisoning by drugs, medicaments, and biologicals; Nos. E950.0-E950.5, Suicide
by drugs, medicaments, and biologicals; No. E962.0, Assault from poisoning
by drugs and medicaments; and Nos. E980.0-E980.5, Poisoning by drugs, medicaments,
and biologicals, undetermined whether accidentally or purposely inflicted.
Drug induced causes exclude accidents, homicides, and other causes indirectly
related to drug use. Also excluded are newborn deaths associated with mothers
Codes for alcohol induced deaths
Causes of death attributable to alcohol induced mortality includeI
CD-9 No. 291, Alcoholic psychoses; No. 303, Alcohol dependence syndrome;
No. 305.0, Nondependent abuse of alcohol; No. 357.5, Alcoholic polyneuropathy;
No. 425.5, Alcoholic cardiomyopathy; No.535.3, Alcoholic gastritis; Nos.
571.0–571.3, Chronic liver disease and cirrhosis, specified as alcoholic;
No. 790.3, Excessive blood level of alcohol; and No. E860, Accidental poisoning
by alcohol, not elsewhere classified. Alcohol induced causes exclude accidents,
homicides, and other causes indirectly related to alcohol use. This category
also excludes newborn deaths associated with maternal alcohol use.90 National
Vital Statistics Report, Vol. 47, No. 9, November 10, 1998
Quality of reporting and processing cause of death
One index of the quality of reporting causes of death is the proportion
of death certificates coded to the Ninth Revision; Chapter XVI; Symptoms,
signs, and ill-defined conditions (ICD-9 Nos.780-799). Although deaths
occur for which the underlying causes are impossible to determine, this
proportion indicates the care and consideration given to the certification
by the medical certifier. This proportion also may be used as a rough measure
of the specificity of the medical diagnoses made by the certifier in various
areas. In 1996,1.13 percent of all reported deaths in the United States
were assigned to Symptoms, signs, and ill-defined conditions compared with
1.18 percent in 1995. The percent of deaths from this cause for all ages
combined generally has remained stable since 1990.
The cause-of-death rankings in figure 4 and tables B, 7, 8, and
9 are based on the List of 72 Selected Causes of Death, HIV Infection (ICD-9
Nos. *042-*044) and Alzheimer's disease (ICD-9 No. 331.0). HIV infection
and Alzheimer's disease were added to the list of rankable causes effective
with data years 1987 and 1994, respectively. The cause-of-death ranking
for infants in tables 27, 28, and 29 are based on the List of 61 Selected
Causes of Infant Death and HIV Infection. HIV infection was added to the
list of rank able causes of infant death beginning with the 1994 data year.
The group titles Major cardiovascular diseases and Symptoms,
signs, and ill-defined conditions are not ranked from the List of 72 Selected
Causes; Certain conditions originating in the perinatal period and Symptoms,
signs, and ill-defined conditions are not ranked from the List of 61 Selected
Causes of Infant Death. In addition category titles that begin with the
words ‘‘Other’’ and ‘‘All other’’ are not ranked to determine the leading
causes of death. When one of the titles that represents a subtotal is ranked
(for example, Tuberculosis), its component parts are not ranked (in this
case, Tuberculosis of respiratory system and Other tuberculosis).
For 1996 mortality data for the Hispanic origin population are based
on deaths to residents of 49 States and the District of Columbia. Oklahoma
was excluded because its death certificates did not include an item to
identify Hispanic or ethnic origin. Infant mortality rates for the Hispanic
origin population are based on numbers of resident infant deaths reported
to be of Hispanic origin and numbers of resident live births by Hispanic
origin of mother for the same 49 States and the District of Columbia. In
computing infant mortality rates, deaths and live births of unknown origin
are not distributed among the specified Hispanic and non Hispanic groups.
The percent of infant deaths of unknown origin was 1.6 and the percent
of live births of unknown origin was 1.5 for the 49 States and the District
of Columbia for 1996.
Small numbers of infant deaths for specific Hispanic origin
groups result in infant mortality rates subject to relatively large random
variation (see ‘‘Random variation’’). Infant mortality rates by Hispanic
origin are less subject to reporting bias when based on linked files of
infant deaths and live births (15).
In 1990 the 49 States and the District of Columbia accounted
for 99.6 percent of the Hispanic population in the United States, including
about 99.5 percent of the Mexican population, 99.8 percent of the Puerto
Rican population, 99.9 percent of the Cuban population, and 99.7 percent
of the Other Hispanic population (34).
Quality of Hispanic origin data-A study (19) examined the reliability
Hispanic origin reported on 43,520 death certificates with that reported
on a total of 12 CPS’s conducted by the U.S. Bureau of the Census for the
years 1979–85. In this study, agreement was 89.7 percent for any report
of Hispanic origin. The ratio of deaths for CPS divided by deaths for death
certificate was 1.07 percent, indicating net underreporting of Hispanic
origin on death certificates as compared with self-reports on the surveys.
The sample was too small to assess the reliability of specified Hispanic
Age specific and age adjusted death rates by marital status are shown
in table 21. Mortality data by marital status is generally of high quality.
A study of death certificate data using the 1986 National Mortality Follow
back Survey showed a high level of consistency in reporting marital status
(20). Age adjusted death rates by marital status were computed based on
the age specific rates and the standard population for ages 25 years and
over. While age-specific death rates by marital status are shown for the
age group 15-24years, they are not included in the computation of the age-adjusted
rate because of their high variability, particularly among the widowed
population. Also, the age groups 75–84 and 85 years and over are combined
because of high variability in death rates in the 85year-and-over age group,
particularly for the never married population.
Beginning with the 1989 data year, an item indicating decedent’s
educational attainment was added to the certificates of numerous States.
Mortality data on educational attainment for 1996 are based on deaths to
residents of the 45 States and the District of Columbia whose data were
approximately 80 percent or more complete on a place-of-occurrence basis.
Data for Kentucky were excluded using this criterion. Data for Georgia,
Oklahoma, Rhode Island, and South Dakota were excluded because the item
was not on their certificates. For the first time, age specific and age
adjusted death rates by educational attainment are shown in this report
(table 22). Age-adjusted death rates by educational attainment were computed
based on the age-specific rates and the standard population for ages 25–64
years. Data for age groups 65 years and over are not shown because
reporting quality is poorer at older than younger ages (35). Rates by educational
attainment are affected by differences in measuring education for the numerator
and the denominator. The numerator is based on number of years of education
completed as reported on the death certificate whereas the denominator
is based on highest degree completed as reported on census surveys (36).
Injury at work
Information on deaths attributed to injury at work is derived from
a separate item on the death certificate that asks the medical certifier
whether the death resulted from an injury sustained at work. The item is
on the death certificate of all States. Number of deaths, age-specificrates,
and age adjusted rates by injury at work are shown for the first time in
this report (table 23). Deaths and age specific death rates by National
Vital Statistics Report, Vol. 47, No. 9, November 10, 1998 91
injury at work are shown for ages 15 years and over. Age adjusted
death rates by injury at work were computed using age specific death
rates and the standard population for ages 15 years and over.
U.S. abridged life tables are constructed by reference to a
standard life table (37). The life table provides a comprehensive
measure of the effect of mortality on life expectancy. It is composed
sets of values showing the mortality experience of a hypothetical
group of infants born at the same time and subject throughout their
lifetime to the specific mortality rates of a particular time, usually
Causes of death contributing to changes in life expectancy
Causes of death contributing to changes in life expectancy were
estimated using a life table partitioning technique. The method
partitions changes into component additive parts. This method identi-fies
the causes of death having the greatest influence, positive or
negative, on changes in life expectancy (4,5).
Population bases for computing rates
The population used for computing death rates in this report
(furnished by the U.S. Bureau of the Census) represents the
population residing in the specified area. Death rates for the United
States for 1996, shown in table I, are based on population estimates
as of July 1, 1996 (38). The estimates are based on the 1990 census
level counts. The 1990 census level counts by race were modified to
be consistent with U.S. Office of Management and Budget categories
and historical categories for death data (39). The population estimates
for Mexicans, Puerto Ricans, Cubans, and Other Hispanics, shown in
table II, are based on the CPS adjusted to resident population control
totals (40) for 49 States (excluding Oklahoma) and the District of
Columbia and, as such, are subject to sampling variation. Population
estimates by marital status, shown in table III, are also based on
CPS (40) adjusted to resident population control totals for all 50
States and the District of Columbia and are also subject to sampling
variation (see ‘‘Random variation’’).
Population estimates by educational attainment, shown in table IV,
are also based on the CPS (40) adjusted to resident population control
totals for 45 States and the District of Columbia and are also subject
to sampling variation (see ‘‘Random variation’’).
Population estimates for each State, Puerto Rico, Virgin Islands,
and Guam, shown in table V, are based on demographic analysis and,
therefore, are not subject to sampling variation (41–44).
Computation of rates
Rates in this report are on an annual basis and, except for infant
and maternal mortality rates, are per 100,000 estimated population
a specified group or area. Comparisons made in the text among rates,
unless otherwise specified, are statistically significant at the 0.05
of significance. Lack of comment in the text about any two rates does
not mean that the difference was tested and found not to be
significant at this level.
Infant mortality rates are the most commonly used index for
measuring the risk of dying during the first year of life. They are
calculated by dividing the number of infant deaths in a calendar year
by the number of live births registered for the same period and are
presented as rates per 1,000 or per 100,000 live births. Infant mortality
rates use the number of live births in the denominator to approximate
the population at risk of dying before the first birthday.
In contrast to infant mortality rates based on live births, infant death
rates are based on the estimated population under 1 year of age. Infant
death rates that appear in tabulations of age specific death rates
calculated by dividing the number of infant deaths in a calendar year
by the midyear population of infants under 1 year of age (estimated
births occurring in the 12 month period ending with June) and are
presented as rates per 100,000 population in this age group. Because
of differences in the denominators, infant death rates may differ from
infant mortality rates.
Maternal mortality rates are computed on the basis of the number
of live births. The maternal mortality rate indicates the likelihood
pregnant woman dying of maternal causes. They are calculated by
dividing the number of maternal deaths in a calendar year by the number
of live births registered for the same period and are presented as
per 100,000 live births. The number of live births used in the denomi-nator
is an approximation of the population of pregnant women who are
at risk of a maternal death.
Age adjusted death rates are used to compare relative mortality
risk across groups and over time. However, they should be viewed as
a construct or an index rather than as direct or actual measures of
mortality risk. Statistically, they are weighted averages of the age
death rates, where the weights represent the fixed population
proportions by age (45). The age adjusted rates presented in this report
excluding those by marital status were computed by the direct method,
that is, by applying the age specific death rates for a given cause
death to the U.S. standard population (relative age distribution of
enumerated population of the United States), which is shown in table
along with the corresponding weights used for computing the relative
standard error (RSE).
Age adjusted rates by marital status were computed by applying
the age specific death rates to the U.S. standard population for ages
25 years and over. Although death rates by marital status are shown
for the age group 15–24 years, they are not included in the calculation
of age adjusted rate because of their high variability, particularly
the widowed population. Also, the age groups 75–84 and 85 years and
over are combined because of high variability in death rates in the
year-and-over age group, particularly for the never married population.
The standard population and corresponding weights used for computing
age adjusted rates and relative standard errors by marital status are
shown in table VII.
Age adjusted rates by education were computed by applying the
age specific death rates to the U.S. standard population for ages 25–64
years. Data for age groups 65 years and over are not shown because
reporting quality is poorer for older than for younger ages (35). The
standard population and corresponding weights used for computing
age adjusted rates and relative standard errors by education are shown
in table VIII.
Age adjusted rates for Puerto Rico, Virgin Islands, and Guam were
computed by applying the age specific death rates to the U.S. standard
92 National Vital Statistics Report, Vol. 47, No. 9, November 10, 1998
January 16, 2008 -- CDC Just released :
- Age-adjusted death rates for 113 selected causes by race and sex: United States,
- Death rates for 113 selected causes by race and sex: United States, 2005
- Number of deaths from 113 selected causes by race and sex: United States, 2005
- Number of deaths from 113 selected causes by age: United States, 2005
- Death rates by age and age-adjusted death rates for the 15 leading causes of
death in 2005: United States, 1999-2005
- Life expectancy at selected ages by race and sex: United States, 2005
- Number of deaths and death rates, by age, race, and sex: United States, 2005
- Number of deaths, death rates, and age-adjusted death rates, by race and sex:
United States, 1940, 1950, 1960, 1970, and 1980-2005
- Abridged life table for the total population, 2005