and heart disease as the child ages. These nutritional deficiencies were increasingly alleviated in the post-1840s birth cohorts.
Chronic disease risks were further altered in the early part of the twentieth century. For example, in the 1920s and 1930s changes in food preservation, thermal preparation, and storage affecting microbial food contaminants likely combined with changes in salt intake and water quality (e.g., reducing the prevalence of H. pylori infections) to alter the incidence of a wide range of chronic diseases, including stroke, hypertension, gastric and other cancers, and peptic ulcers (Fogel, 1999; Fogel and Costa, 1997). A similar pattern was found in Great Britain. Up to the 1940s the British centenarian population grew 1 percent per year (Perutz, 1998). After the 1940s (i.e., centenarians born after the 1840s) the growth rate was nearer 6 percent. Thus, in both Britain and the United States the major socioeconomic and nutritional changes appear to have affected the health and survival of post-1840 birth cohorts.
The national trends described above are accompanied by enormous variation within countries. County-specific analyses of historical trends in the adoption of primary prevention strategies and shifts in average socioeconomic status levels relative to those for a given state, or for the country at large, could provide a useful baseline for the formulation and targeting of future health promotion and disease prevention strategies.
Several influences on declining rates of disability among the elderly have been proposed. One suggestion is that these health improvements result from changes in the nature of work. Work has become less manually intensive and more cognitive over time, potentially delaying the onset of a range of adverse conditions, including musculoskeletal disorders and cardiovascular complications. In addition, exposure to dust and hazardous chemicals has declined. Preliminary evidence suggests that these changes may explain up to one-quarter of improvements in health for the elderly since the turn of the century (Costa, 1998), although no evidence exists for recent years.
The nature of work may matter in other ways as well. Work that is mentally stressful or not mentally challenging enough may lead to psychological stress that is manifest in physical disorder. For example, musculoskeletal disorders are more common in people with low job satisfaction, elevated psychophysiological stress reactions, and lack of opportunity to unwind, all of which are characteristic of repetitious work with short time cycles (Melin and Lundberg, 1997). Such findings are also reported in the Whitehall studies (Marmot et al., 1991; Marmot, 1994).
A second possibility is that health improvements result from improved