IN ITS ideal sense, the epidemiology of infant mortality transcends the mere recording of infant deaths.
Rather, it extends back to the infants' births to determine the medical and social context into which they were born. Only then can health care workers identify weaknesses in maternal and infant care and design strategies to address these problems.
Despite the importance of linking data from death certificates (age, cause of death, state, and race) with data from birth certificates (the infant's birth weight, live-birth order, length of gestation, and type of delivery and the mother's education, age, marital status, and prenatal care history), the years between 1960 and 1980 lapsed without a national linkage study.
"One of the biggest issues of the 1980s was, which babies are dying? How can we characterize the ones that aren't going to do well? We can't answer that question without linking the deaths to the births,"