About Infant Mortality Data
The purpose of the Infant Mortality module is to generate interactive statistics
and maps of infant mortality data. The intended audience is researchers, public
health officials, students, and the general public.
The Infant Mortality module is an infant death cohort. This means that the data
used is a combination or merge of infant death data taken from the death certificate
and birth data taken from the birth certificate data. Once the infant deaths have
been matched to their respective birth certificates then a dataset is created all
based on the year that the child has died. A child can be born in the same year
or the preceding year of his/her death. The denominator for any rates calculated
is the number of births that occur in the year of the child's death. Demographics
data such as race, ethnicity, county, and zip code are generally based on the death
certificate information of the infant.
The infant mortality rates for indicators originating from the birth certificate
data are calculated by taking the number of deaths with that characteristic over
the number of live births with that same characteristic. For example the infant
mortality rate of mother's less than 18 would be: the annual number of deaths to
mothers less than 18 over the annual live births to mothers less than 18.
Infant mortality rates for the causes of death are per 1000 annual live births.
The causes of death present a challenge in this module due to the changes in the
International Classification of Diseases (ICD) in 1999. Since the rankable causes
of infant death change with each edition of ICD codes, trendlines are not available
for years that overlap the changes in editions of the ICD 9 and the ICD10. Both
the ICD9 and ICD10 codes that were used in the classification of each disease are
placed next to the cause of death. Users can however select tables and bar charts
that do allow for all years of data.
A new feature that is specific to this module is the "top ten causes of death" variable,
which allows users to generate the top ten causes of infant death specific to their
For both datasets (births and deaths) the data is residents of South Carolina and
users can select the region of analysis. Users can generate on-the-fly maps and
tables of the frequencies, rates, and percentages along with trendlines and bar
charts of the selected data. When rates have been selected 95% Confidence intervals
can also be generated in the tables.
The interpretation of tables and maps is the users responsibility. Many supporting
pages are provided with definitions, examples of rate calculations, Division of
Biostatistics and Health GIS contact information, and frequently asked questions
to assist the user in interpreting this data. Rates with small numbers are unreliable
and should be used cautiously.