evidence for specification of pathways that combine sources of cumulative adversity and advantage over time. Both data sets show evidence of negative mental health consequences in adolescence and adulthood that accrue from adverse experience in early childhood as well as the influence of cumulative experience from early childhood on self-reported midlife physical health. Until recently, direct physiological assessments of these populations had not been made. With planned future longitudinal follow-up on both psychosocial and physiological indicators, the British birth cohorts represent promising sources of data for constructing integrative life history accounts of cross-time associations between socioeconomic standing and health.
In the United States, the Wisconsin Longitudinal Study (WLS; Hauser et al., 1993), which is focused on the birth cohort of 1939, is the closest to the British birth cohorts in richness of psychosocial information but goes well beyond the British studies with its in-depth assessments of educational attainment and occupational experience as well as accompanying data from siblings, spouses, and parents. WLS is, however, a random sample of the high school graduating class of 1957 in the state of Wisconsin and lacks the in-depth psychosocial assessments of infancy and early childhood found in the British birth cohorts. WLS also does not include the lowest level of the education hierarchy. Nonetheless, the age 52-53 round of WLS data collection initiated extensive inquiry on physical health and psychological well-being, and a subsample (N = 115) of respondents was resurveyed at age 59, at which time extensive physiological assessments were also obtained. These data have afforded an evidential basis for linking cumulative economic and relational adversity to high allostatic load (Singer and Ryff, 1999). WLS and the British birth cohorts represent the best of what is currently available on psychosocial experience across multiple life domains through time. Other important longitudinal resources, although more restricted in length of follow-up and/or coverage of psychosocial experience, are the Alameda County Study (Berkman and Syme, 1979), the Panel Study of Income Dynamics (McDonough et al., 1997), the National Long Term Care Survey (Corder et al., 1993), the Whitehall II Study of the British Civil Service (Marmot et al., 1991), and the Malaysian Life History Survey (Da Vanzo, 1984; Haaga et al., 1994).
Shifting to an emphasis on direct physiological assessment, the Framingham Study of heart disease (Allaire et al., 1999; Dawber, 1980) is an invaluable longitudinal data source with biennial physical health assessments running over 45 years. Regrettably, this rich study of physical health is nearly devoid of parallel psychosocial assessments. The implication for NIH funding priorities is that there is need to develop and maintain longitudinal studies of new birth cohorts that contain the necessary psychosocial and physiological assessments in parallel. Equally important is the need to