Emory Medicine  
Fighting Infant Mortality,
One Census Tract at a Time

By Kay Torrance

    Close window

Print page

Every now and then, Al Brann goes to the top floor of Grady Hospital in the late afternoon to watch the planes descend methodically on their final approach to Atlanta's airport. The pediatrics professor likens his calling, and that of his colleagues in obstetrics, to an air traffic controller. All want to ensure a "safe landing" for each pregnancy.
     But that continues to be a challenge in Georgia, where some 2,500 babies are born each year at less than eight months' gestation and weighing less than 3.3 lbs.
     Brann and his colleagues want to know why. Highly specialized obstetric and neonatal care, regionalized under Governor Jimmy Carter in the early 1970s, has led to improved survival for all newborn infants. But while infant mortality has declined 50% statewide over the past 25 years, Georgia still ranks a dismal 40th nationwide in infant mortality.

         And unfortunately, the frequency of very low birth weight (VLBW) births has remained unchanged. As these babies account for half of the nearly 1,200 infant deaths in the state annually, reducing their numbers is crucial to further decreasing infant mortality.
      As obstetricians know, the rate of newborn infant deaths varies according to where the mother lives and other sociodemographic characteristics of her life. In North Georgia, Caucasian women 20 or older and with 13 or more years of education have a feto-infant mortality rate of five per 1,000 births. But for infants born to African American women statewide and to Caucasian women living in central and southern Georgia, the rate is almost three times higher, with 14.35 deaths per 1,000 births.
     It would be easy to assume that factors such as economic status and education level drive the higher rates, but the answer is not so simple. Infants born to educated African American women 20 or older still have a greater risk of infant mortality than those born to Caucasian teenagers. "African American women, compared with Caucasian women, are three to four times more likely to have a very low birth weight infant who is at greater risk of dying," says Brann. So what gives?
     The World Health Organization Collaborating Center in Reproductive Health, a partnership led by Brann with Emory, the CDC, and Georgia's Division of Public Health, is analyzing birth outcomes throughout the state to identify census tracts with a significant concentration of VLBW babies. The project started in Fulton and DeKalb counties, where census tracts were identified with a VLBW infant rate as high as 4%. The statewide average ranges from 0.9% to 1.2%. If doctors understand more about what leads to these poor outcomes, they can predict and prevent bad outcomes in more first-time pregnancies, Brann says. Currently, that's impossible to do.
  We can predict with good accuracy the chances of having a very low birth weight infant in subsequent pregnancies. We want to predict and prevent these bad outcomes in women giving birth for the first time.—Al Brann  
         "The problem is that we haven't had data on the mothers of VLBW infants to know if there are environmental effects on women's health that may lead to bad pregnancy outcomes," says Brann. "Having such data will tell us what we should be recommending. We are using the surveillance system to identify specific risks amenable to intervention within these census tracts."
      Generally, conditions adversely affecting the health of mothers, such as hypertension, sexually transmitted diseases, domestic violence, and depression, account for 60% of Georgia's feto-infant mortality. The collaborating center team will look for demographic and environmental factors common to mothers who deliver VLBW infants and the census tracts in which they live, including the cost of housing, income, education level, race and age of residents in the tract, the types of illness they have, and neighborhood crime and violence. "Once we obtain this information," Brann says, "we can determine not only where but how to intervene."
     A recent study at Grady Hospital led by Brann and Emory physician Anne Lang Dunlop indicates that health care interventions targeting women after the birth of a VLBW baby can influence subsequent reproductive outcomes. Results showed that women who received primary health care, who spaced their children 24 months apart after delivering a VLBW baby, and who received vocational and educational counseling had fewer adverse pregnancy outcomes than expected statistically. Building on that study, Brann is seeking to identify additional demographic factors that can be targeted to improve birth outcomes statewide.
     Ultimately, the lessons learned may lead to fewer infant deaths in Georgia, the nation, and the world.


Copyright © Emory University, 2007. All Rights Reserved