Cohort Differences in Disability and Disease Presence
Citation: Reynolds, et al. (1998) Cohort Differences in Disability and Disease Presence.
Introduction: "Most studies of health trends emphasize a period approach toward examining trends . . . A cohort approach provides a different picture of health change in a society, one in which the relative health of successive cohorts, rather than of successive years, is the focus. Period trends in health are potentially confounded by historical events and conditions that birth cohorts experience at different points in their respective life courses. . . Those for whom childhood was a period was a period of deprivation because of depression or war may experience lasting effects from these events (Barker 1994). Alternatively, cohort differences in early-life exposure to disease and pathogens may act to produce differences in later-life health" (p. 578). Because existing evidence regarding changes in health across cohorts is mixed, this article attempts to "provide more evidence on the levels of disability and the prevalence of major diseases among a set of cohorts that span a wide range of birth years and adult ages in the United States" (p. 579).
Methods: The National Health Interview Survey (NHIS), which is a ongoing household survey of the noninstitutionalized population of all ages in the US, is the source of data for this study. 599,141 adults aged 30-69 were included in the sample. Fifteen separate cohorts-each a compilation of 3 single-year cohorts-were analyzed for the years 1916 (i.e., 1915-17) to 1958 (i.e., 1957-59). Each cohort was then followed from 1982 to 1993. Long-term disability due to chronic disease or impairment was measured by questions on limitation in major activity and ability to work. Measures of diseases and other chronic conditions were used to determine cohort patterns of arthritis, asthma, bronchitis, cardiovascular diseases, diabetes, emphysema, mental disorders, musculoskeletal conditions, and orthopedic impairments; some infectious diseases were also included. "Some of the analysis is presented graphically because strong cohort patterns are readily apparent in the graphs. The statistical significance of cohort differences in health indicators is determined using logistic regressions" (p. 580).
Results: Among men, disability is generally higher for early-born cohorts. For instance, "The cohorts of 1919 and 1916 are about 34% more likely to be limited that those born in 1937" (p. 581). However, later-born cohorts (i.e., born after 1946) report somewhat more limitations. In terms of chronic diseases among men, earlier-born cohorts have higher levels of arthritis, cardiovascular diseases and emphysema, but lower levels of diabetes and orthopedic impairments. Among women, later-born cohorts report disability levels that are slightly higher than early-born cohorts, but they also report less inability to work. With the exception of arthritis, which is higher in early-born cohorts, women do not report significantly different levels of chronic diseases or infections across cohorts. When education is taken into account (by separating each cohort into a high and low education group), expected differentials emerge. Also, later-born cohorts with high education generally report fewer disabilities and chronic conditions than early-born cohorts with high education. Interestingly, however, later-born cohorts with low education actually fare worse in some respects (e.g., orthopaedic impairments) than their early-born counterparts.
Conclusions: Patterns of disability and disease change are mixed. Disability appears to have declined among cohorts born in the initial decades of the 20th century, but there is some indication of increasing disability among cohorts born in the 1950s. Similarly, while cardiovascular diseases (men) and arthritis (women) have shown steady improvement across cohorts, later-born cohorts are more likely to suffer from asthma (especially women) and orthopaedic impairments.