Two certificates for one child, joined into one record

A death certificate says how someone died. For an infant, that is a thin
description of a large question.

So **for every infant under 1 year who dies in the United States, Puerto Rico,
the US Virgin Islands and Guam, the information from the death certificate is
linked to the information from that infant's own birth certificate.**

What the link adds

From the birth certificate:

Age, race and Hispanic origin of the parents · birth weight · **period
of gestation** · plurality · prenatal care usage · **maternal
education** · live birth order · marital status · maternal smoking

From the death certificate:

Age at death · underlying and multiple cause of death

Separately, these are two administrative documents. Joined, they answer
questions neither can: whether the infants who die were born early, born small,
born to mothers who had prenatal care, born to mothers who smoked — and how
each of those interacts with the cause of death.

**The purpose of the linkage is to use the many additional variables available
from the birth certificate to conduct more detailed analyses of infant
mortality patterns.**

Why it is built this way

Nobody surveys the parents of infants who die. The information has to come from
records that already exist, gathered for other reasons, and the birth
certificate is the only one that captures the circumstances of the pregnancy
for every birth.

The alternative — a study recruiting bereaved families — would be slower,
smaller, and biased towards whoever agreed to take part. The linked file has no
such bias because nobody opted in: it is every infant death, joined to every
corresponding birth.

What it has been used for

Published analyses drawing on it include:

• **Behind International Rankings of Infant Mortality: How the United States
  Compares with Europe** — a comparison that turns on how countries record
  births at the edge of viability, which is exactly what the birth certificate
  variables make visible
• Recent Trends in Infant Mortality in the United States
• **Trends in Preterm-Related Infant Mortality by Race and Ethnicity,
  1999–2004**

All three ask a question about why, and all three need what was true at
birth.

*Source: Centers for Disease Control and Prevention, National Center for
Health Statistics.*