the study of collective properties of populations and local physical surroundings as they influence individual and overall population-level health. At least two options are available for developing core samples to facilitate the kinds of studies prioritized in this report. One option is to start new birth cohorts and community studies that are responsive to the integrated biopsychosocial lines of inquiry. The second is to take existing longitudinal and community studies that have strengths in particular areas (e.g., psychosocial factors) and add biomedical factors to them. In the case of the Framingham Study, for example, psychosocial data would be needed to augment the superb biomedical assessments that have been the basis for this study from its inception.

As indicated in Chapter 2 and Chapter 3, there are few extant longitudinal samples that could be, even in principle, the basis for empirical characterizations of pathways over major portions of people's lives. The 1946 and 1958 British birth cohorts (Power and Matthews, 1998; Wadsworth and Kuh, 1997), the Wisconsin Longitudinal Survey (WLS; Hauser et al., 1993), the Alameda County Study (Berkman and Breslow, 1983), the Baltimore Longitudinal Study of Aging (National Institute on Aging, 1991), and the Harvard Mastery of Stress Study (Russek and Schwarz, 1997) are prominent examples of long-term studies that could be included in the core investigations of biopsychosocial pathways. Furthermore, it would be useful to add biomarker and other health-related data to the National Longitudinal Survey (NLS), the Panel Study of Income Dynamics (PSID), the Health and Retirement Survey (HRS), the National Survey of Families and Households (NSFH), and the Survey of Income and Program Participation (SIPP). 2

Correlatively, it would be useful to add psychosocial and socioeconomic data to the Framingham Study and to the longitudinal components of the National Health Interview Survey (NHIS) and the National Health and Nutrition Examination Survey (NHANES). 3 The common feature of all these samples is that they are rich in either psychosocial or biomedical information but not both. A variety of possibilities need to be considered in


The data are available electronically: National Longitudinal Survey:; Panel Study of Income Dynamics:; Health and Retirement Survey:; National Survey of Families and Households: ; Survey of Income and Program Participation:


The data are available electronically: Framingham Study:; National Health Interview Survey: ; National Health and Nutrition Examination Survey:

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