Background: The changing racial and ethnic diversity of the U.S. population along with delayed childbearing suggest that shifts in the demographic composition of gravidas are likely. It is unclear whether trends in the proportion of births to porous women in the United States have changed over the decades by race and ethnicity, reflecting parallel changes in population demographics. Methods: Singleton deliveries ≥20 weeks of gestation in the United States from 1989 through 2000 were analyzed using data from the "Natality data files" assembled by the National Center for Health Statistics (NCHS). We classified maternal age into three categories; younger mothers (aged <30 years), mature mothers (30-39 years) and older mothers (≥40 years) and maternal race/ethnicity into three groups: blacks (non-Hispanic), Hispanics and whites (non-Hispanic). We computed birth rates by period of delivery across the entire population and repeated the analysis stratified by age and maternal race. Chi-squared statistics for linear trend were utilized to assess linear trend across three four-year phases: 1989-1992, 1993-1996 and 1997-2000. In estimating the association between race/ethnicity and parity status, the direct method of standardization was employed to adjust for maternal age. Results: Over the study period, the total number of births to blacks and whites diminished consistently (p for trend <0.001), whereas among Hispanics a progressive increase in the total number of deliveries was evident (p for trend <0.001). Black and white women experienced a reduction in total deliveries equivalent to 10% and 9.3%, respectively, while Hispanic women showed a substantial increment in total births (25%). Regardless of race or ethnicity, birth rate was associated with increase in maternal age in a dose-effect fashion among the high (5-9 previous live births), very high (10-14 previous live births) and extremely high (≥15 previous live births) parity groups (p for trend <0.001). After maternal age standardization, black and Hispanic women were more likely to have higher parity as compared to whites. Conclusions: Our findings demonstrate substantial variation in parity patterns among the main racial and ethnic populations in the United States. These results may help in formulating strategies that will serve as templates for optimizing resource allocation across the different racial/ethnic subpopulations in the United States.