tionships across racial and ethnic groups provide a substantial challenge to researchers.

The possibility that behavior has different effects for different groups needs broader investigation. Work in this area has focused mainly on obesity among blacks and whites, leading so far to inconclusive findings. If different effects can be confirmed, more work would be needed to investigate the possible reasons. Attention is also needed to other ethnic groups, such as Hispanics, for whom obesity may still be risky but possibly proportionally less of a risk: if this is confirmed, it might shed further light on the mechanisms involved.

Variation in behavior risk factors across racial and ethnic groups cannot be accounted for solely by socioeconomic factors. While other factors have been suggested, especially acculturation, these are probably insufficient to account, for instance, for the relatively unhealthy behaviors among American Indians and Alaska Natives and blacks. Such problematic conditions as obesity among black women need to be better understood. The reasons for the high rates, and the reasons that obesity sometimes appears to have proportionally less effect than in other groups, need to be clarified. Cutler et al. (2003) argue that overweight and obesity have increased most among those, often women, who have benefited most from technological innovations that have made foods more convenient and reduced food preparation effort. Whether this argument has anything to do with the obesity problems among black women deserves exploration.

The role that social context—families, peer groups, neighborhoods, communities—plays in such contrasts, and generally in deterring healthy behavior and facilitating risky behavior, deserves further study. For instance, how are neighborhood factors related to racial and ethnic composition? Are neighborhood factors simply socioeconomic in origin, or are there modifiable features that could be made more favorable to healthy behavior even in the absence of reductions in poverty? Systematic data on neighborhoods could be collected (Morenoff and Lynch, 2004; Raudenbush and Sampson, 1999), and geographical analysis of national data sets could focus on “very small areas” (Bond Huie et al., 2002; see also Chandra and Skinner, 2004).

The focus of research in this area has been largely on blacks and whites, and, to a lesser extent, on Hispanics. Work is desirable to clarify behavioral differences for the other major groups—American Indians (particularly on reservations) and Asians, as well as subgroups of these groups. The behavior of these two groups appears to provide clear contrasts with whites, but the analysis necessary to determine whether these contrasts are at root socioeconomic has not been done. Does acculturation affect the behavior of these groups in the same way it affects Hispanics? If this could also be determined, one might have further clues to how the negative behavioral effects of acculturation might be arrested. Within the major racial and



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement