Data Suppression

Birth and Infant Mortality Statistics Data

Data Source: National Center for Health Statistics (NCHS), National Vital Statistics System (NVSS)

The Birth and Infant Mortality Statistics data contained in the data warehouse are provided in accordance with the data use agreement between HRSA and NCHS. Descriptions of the supersession rules employed for data sets: (1) years 2007 and later and (2) years 1997 through 2006 are presented below:

Citation:

• National Center for Health Statistics. Linked Birth/Infant Death (2007, 2008, 2009), as compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program.
• National Center for Health Statistics. Mortality/Natality (2007, 2008, 2009), as compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program.

Infant mortality records are identified from the NCHS mortality source data where the NCHS source data element [Age Recode 12] is equal to 01 (Under 1 year (includes not stated infant ages)).

The Birth and Infant Mortality Statistics data are composed of two types of information: aggregated event data and rate data. The event data are aggregated over a single year period from the Birth and Infant Mortality Statistics data provided by NCHS using the data suppression rules listed below. Rate data are calculated by taking the aggregated event data for the county or state and dividing that number by the total number of births for the county or state for the same period. That result is multiplied by 1,000.

The Birth and Infant Mortality Statistics data are aggregated at two levels: state and county. The geography to which a record is assigned is based on the state residence Federal Information Processing Standards (FIPS) code and the county residence FIPS code as identified in the NCHS data. The suppression rules are applied to each data set separately, which means that data suppressed at the county level may not be suppressed at the state level.

Each input data set from NCHS contains data for one full year. There is a three (3) year lag in the release of new data (for example, the 2007 data became available in 2010).

The following U.S. Territories do not report health statistical data to the NCHS so they are not represented in the HRSA Data Warehouse for Birth and Infant Mortality Statistics data:

• Federated States of Micronesia
• Marshall Islands
• Republic of Palau
• U.S. Minor Islands

The Birth and Infant Mortality Statistics data reported in the HRSA Data Warehouse follows the NCHS rules as set forth in the data use agreement between HRSA and NCHS. These rules apply to the data for any single state or county.

Data are not reported when any one of the following conditions is met:

• NCHS data state FIPS code is not a valid state FIPS code
• NCHS data state FIPS code and county FIPS code combination is not a valid state FIPS code and county FIPS code combination

Data are reported as blank (null) when any of the following conditions is met:

• For natality aggregations: where there is not any natality event data for the given geography.
• For mortality aggregations: where there is not any mortality event data for the given geography
• For all rates: where the Total Live Births is null (blank) or zero (0).
• For specific rates: where the numerator is null (blank).

Data are reported as suppressed (-999) when any of the following conditions are met:

• For aggregated values: where the value for the year less than 10 for the given geography.
• For aggregated values: where the data element could be used to calculate a related value that has been suppressed.
• For aggregated values: where any member of a set of related data elements is suppressed (e.g. suppressing any single combination of gender and race causes all combinations of gender and race to be suppressed for a record).
• For all rates: where the Total Live Births is suppressed (-999).
• For specific rates: where the numerator is suppressed (-999).

See the Infant Mortality Suppression Application Table and the Live Birth (Natality) Suppression Application Table for clarification of the rules of data suppression where the data element that is at the same level or a lower level of aggregation than a data element that has been suppressed.

Data are reported as zero (0) when either of the following conditions is met:

• Value is zero (0).
• There are mortality records (all deaths, not limited to infant deaths) for each of the five years but there are no infant mortality events for the given geography.

Citation:

• National Center for Health Statistics. Linked Birth/Infant Death (2005, 2004, 2006), as compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program.
• National Center for Health Statistics. Mortality (1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006), as compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program.
• National Center for Health Statistics. Mortality Geographics Identified (2005), as compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program.
• National Center for Health Statistics. Natality (2006), as compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program.

Infant mortality records are identified from the NCHS mortality source data where the NCHS source data element [Age Recode 12] is equal to 01 (Under 1 year (includes not stated infant ages)).

The Birth and Infant Mortality Statistics data are composed of two types of information: aggregated event data and rate data. The event data are aggregated over a five year period from the Birth and Infant Mortality Statistics data provided by NCHS using the data suppression rules listed below. Rate data are calculated by taking the aggregated event data for the county or state and dividing that number by the total number of births for the county or state for the same five year period. That result is multiplied by 1,000.

The Birth and Infant Mortality Statistics data are aggregated at two levels: state and county. The geography to which a record is assigned is based on the state residence FIPS code and the county residence FIPS code as identified in the NCHS data. The suppression rules are applied to each data set separately. This means that data that is suppressed at the county level may not be suppressed at the state level.

Each input data set from NCHS contains data for one full year. There is a three (3) year lag in the release of new data (e.g. the 2002 data became available in 2005). Data for any specific state or county may not appear in each of the five (5) years of NCHS input data. The HRSA Data Warehouse only reports data for states and counties where the state or county is included in all five years of the input data.

The following U.S. Territories do not report health statistical data to the NCHS so they are not represented in the HRSA Data Warehouse for Birth and Infant Mortality Statistics data:

• Federated States of Micronesia
• Marshall Islands
• Republic of Palau
• U.S. Minor Islands

The Birth and Infant Mortality Statistics data reported in the HRSA Data Warehouse follows the NCHS rules as set forth in the data use agreement between HRSA and NCHS. These rules apply to the data for any single state or county.

Data are not reported when any one of the following conditions is met:

• NCHS data state FIPS code is not a valid state FIPS code
• NCHS data state FIPS code and county FIPS code combination is not a valid state FIPS code and county FIPS code combination

Data are reported as blank (null) when any of the following conditions is met:

• For natality aggregations: where there are not 5 years of natality event data.
• For mortality aggregations: where there are not 5 years of mortality event data.
• For all rates: where the Total Live Births is null (blank) or zero (0).
• For specific rates: where the numerator is null (blank).

Data are reported as suppressed (-999) when any of the following conditions are met:

• For aggregated values: where the value for the 5 year period is one (1) or two (2).
• For aggregated values: where the data element could be used to calculate a related value that has been suppressed.
• For aggregated values: where any member of a set of related data elements is suppressed (e.g. suppressing any single combination of gender and race causes all combinations of gender and race to be suppressed for a record).
• For all rates: where the Total Live Births is suppressed (-999).
• For specific rates: where the numerator is suppressed (-999).

See the Infant Mortality Suppression Application Table and the Live Birth (Natality) Suppression Application Table for clarification of the rules of data suppression where the data element that is at the same level or a lower level of aggregation than a data element that has been suppressed.

Data are reported as zero (0) when either of the following conditions is met:

• Value is zero (0).
• There are mortality records (all deaths, not limited to infant deaths) for each of the five years but there are no infant mortality events.