Infant mortality in a rural health district in Georgia, 1974 to 1981.

In 1979 the infant mortality rate (IMR) dropped nearly 50% in a rural health district in southeast Georgia, and this lower rate continued during the next two years. For infants born during 1979 to 1981, the IMR was 12.4 as compared with 21.6 for those born from 1974 to 1978. Using linked birth and infant death records to identify shifts in the components of infant mortality in this district, we found that the change in IMR primarily reflected a decrease in neonatal and postneonatal deaths among infants with birth weight greater than or equal to 2,500 gm, rather than a change in the distribution of birth weights. Although the IMR was approximately twofold higher in white than in black infants, a similar relative decline in mortality was observed in both racial groups. For infants weighing greater than or equal to 2,500 gm, approximately half of the lower death rate was due to fewer deaths caused by infections. The decline in mortality in this district was greater than declines in neighboring districts and accompanied efforts to improve services for medically indigent mothers and infants. The findings suggest that in areas with high infant mortality, initial efforts to lower mortality should focus on primary care programs rather than more specialized interventions.