Background:
Most efforts to reduce adverse birth outcomes have focused on prenatal care. Although such care is critical, recent data suggest that it may be insufficient to prevent a substantial portion of adverse outcomes, particularly extreme prematurity. Therefore, new approaches will be necessary to improve birth outcomes and reduce disparities in infant mortality.

Objectives:
To critically assess the epidemiology of adverse birth outcomes in the United States to identify arenas of intervention most likely to reduce disparate infant mortality.

Methods:
Analysis of birth and infant death files for different social groups. Of special interest are indicators of risk and their amenability to efficacious interventions

Results:
The major contributor to disparate infant mortality rates is extreme prematurity. An isolated focus on interventions initiated only after pregnancy has begun is not likely to be effective. Focused preconceptual interventions are similarly not likely to be highly effective. Broader efforts focused on reducing risks in women of reproductive age regardless of pregnancy status or plans are likely to be most important in reducing adverse birth outcomes.

Conclusion and implications for practice:
The reduction of disparate infant mortality rates will need to move beyond prenatal or preconceptual care. Rather, policies and programs will need to be transformed to embrace a comprehensive commitment to women's health.