State-Specific Trends in Preterm Delivery: Are Rates Really Declining Among Non-Hispanic African Americans Across the United States?

Objectives : This study sought to examine state-specific trends in preterm delivery rates among non-Hispanic African Americans and to assess whether these rates are influenced by misclassification of gestational age. Methods : The sample population consisted of singleton non-Hispanic White and non-Hispanic African–American infants born in 1991 and 2001 to U.S. resident mothers. For both time periods, state-specific and national preterm delivery rates were calculated for all infants, stratified by infant race/ethnicity. Next, birth-weight distributions within strata of gestational age were studied to explore possible misclassifications of gestational age. Lastly, state-specific and national preterm delivery rates among infants who weighed less than 2,500 g were separately computed. Results: National analyses showed that the frequency of preterm delivery increased by 15.8% among non-Hispanic Whites but declined by 10.3% among non-Hispanic African Americans over the same period. For both subgroups, a bimodal distribution of birth weights was apparent among preterm births at 28–31 weeks of gestation. The second peak with its cluster of normal-weight infants was more prominent among non-Hispanic African Americans in 1991 than in 2001. After excluding preterm infants who weighed 2,500 g or more, the national trends persisted. State-specific analyses showed that preterm delivery rates increased for both subgroups in 13 states during this period. Of these 13, 6 states had a number of non-Hispanic African–American births classified as preterm that were apparently term births mistakenly assigned short gestational ages. Such misclassification was more frequent in 1991 than in 2001 and inflated 1991 rates. Conclusion : There is heterogeneity in state-specific preterm delivery rates. Such differences are often overlooked when aggregate results are presented.Peer Reviewedhttp://deepblue.lib.umich.edu/bitstream/2027.42/45321/1/10995_2005_Article_32.pd