classification of “levels” into two or even three categories of influence is somewhat crude, because the definition of the higher-level units and the borders between the various levels are frequently unclear.

The overall high rate of preterm birth in the United States and the persistent racial-ethnic gap is one of the most significant public health problems today. Despite many years of observational and clinical research, the exposures that place women at risk are not well understood. The substantial intergroup as well as intragroup variabilities in the risk of preterm birth have been shown to be related related to socioeconomic condition, nativity, acculturation, or other maternal characteristics. Although future research should continue to focus on factors that contribute to the high rates of preterm births among African American infants, much can be learned from examinations of racial-ethnic disparities outside the context of the disparities between African American versus white women, as well as disparities within a particular racial or ethnic group. Common measures of socioeconomic status (e.g., income and education) and other potential mediators may not fully capture the magnitude of group differences. For example, the median income of African American families is about 64 percent of the median income of white families, but the median net worth of African American families is only 12 percent of that of white families (Mishel and Bernstein, 2003).

Similarly, a list of stressful life events cannot adequately measure the multiple dimensions of stress, including acute and chronic stressors, stress appraisal, and the environmental (including social and cultural) contexts of stress. For example, racial discrimination disproportionately affects women of color and is associated with preterm birth, yet it often goes unmeasured in studies linking stress to preterm birth. Thus, better measures (both for “exposures,” such as socioeconomic condition and race, and for potential “mediators,” such as stress) are needed in research on these disparities.

The paradox of favorable birth outcomes, despite social disadvantages, among some immigrant groups and the increased rates of preterm birth with increasing length of residence in the United States have been attributed, in part, to the loss of resiliency factors with increasing acculturation. This suggests that research on disparities in the rates of preterm birth needs to pay more attention to protective factors (which include such factors as personal resources, social support, and spirituality).

Current research into the causes of the disparities commonly attempts to isolate the effect of a single risk factor, without accounting for the cooccurrence and potential interactions among multiple protective and risk factors (e.g., age and race or education and race) operating at multiple lev-