Birth data, computed from the original files

Interactive analyses of United States national birth and infant death data.

Every figure on this site is computed directly from the primary public-use files published by the National Center for Health Statistics. Nothing is copied from a summary table or a secondary report. Where a number is derived, the arithmetic is shown alongside it, and absolute risks are given wherever a relative risk appears.

Tools

Neonatal Mortality by Birthplace

22,708,247 US term live births at 37 weeks or later, 2017–2023, linked to 45,110 infant deaths. Neonatal mortality by birth setting and by birth attendant, in hospital and out of it, against three hospital comparison groups, with absolute excess, births per extra death, and crude and adjusted relative risks. Filter by low-risk status, by clinical group, by parity, by maternal age, or by number of risk factors present.

Data explorerNCHS linked files

→

How these tools are built

Primary source only

Figures are computed from the CDC public-use data files themselves, parsed from the published record layout. No number is taken from a secondary report.

Validated before publication

Each rebuild is checked against the national totals NCHS publishes. Annual resident live births for 2017 through 2023 reproduce those totals exactly, pooled 25,936,466.

Absolute risk, always

Every relative risk is shown beside the absolute difference it came from: excess per 10,000, excess deaths over the period, and how many births separate one extra death.

Self-contained pages

Each tool is a single HTML file with the data inside it. Nothing is sent anywhere, nothing is tracked, and each page works offline once loaded.

Status. The data files are official and public. The analyses of them presented here are the authors' own. They are not yet published and have not been peer reviewed, and figures may change. Each tool carries its own full reference list and a description of how its population was built.

What these tools are not

These are descriptions of what was recorded on birth and death certificates, not measurements of what any setting, attendant or intervention causes. Women who give birth in different settings, and with different attendants, differ from one another in ways the certificate does not capture. Comparisons between groups are associations. Where adjustment has been applied it is stated, and where it has not, that is stated too.

Nothing here replaces clinical judgment or an individual conversation about a particular pregnancy.