Results Acknowledgements

5. Summary and Discussion

The reported prevalence of specific lethal diseases has generally risen in the older noninstitutionalized population in recent years and it appears to have risen more for men than for women. This increase is not surprising given the increased survival from many mortal conditions and the fact that there is little evidence of reductions in disease incidence. Mortality improvement during this period was faster among men than among women. The largest increases in disease prevalence have been in heart disease and cancer, two major causes of old age mortality. Declines in mortality from heart disease have continued from the late 1960s through the present, but decreased mortality from all cancers combined has only recently been a major feature of mortality decline. The increase in reported prevalence of heart disease and cancer most likely results from mortality declines and longer survival for people with these diseases.

There have been some increases in the reported prevalence of non-lethal conditions and impairments, particularly arthritis, osteoporosis and visual conditions. The increase in arthritis occurred among both men and women; the prevalence of osteoporosis more than doubled among women perhaps representing increased interest and diagnosis in this condition.

The increased prevalence of diseases has resulted in a decrease in the proportion of the older population with no disease and an increase in the number of diseases comorbid within individuals. This is to be expected where people survive one disease and live long enough to acquire others. If one looks at change in the presence of disease as an indictor of health, the conclusion is that morbidity has been expanding in recent years.

In spite of this increase in the prevalence of disease and comorbidity, functioning and IADL disability improved among older women. This improved functioning among older women appears among women who have most of the diseases investigated and could represent a decrease in the severity of prevalent disease among women. It could result from improvement or expansion of treatment of diseases. Functioning improvement is not reported for women who have no diseases. For older men functioning deteriorates among those with disease. In general, men have more disease in 1994 with similar disability levels to those of 1984; women have more disease at the later date but less disability. As noted above, men experienced both more decline in mortality during this period and a greater increase in disease prevalence and comorbidity. For these reasons men may not have had the same improvement in functioning and disability as women.

While reduction in mortality without evidence of a reduction in disease incidence would lead us to expect the empirical evidence to reflect an increase in prevalence of disease between 1984 and 1994, disease prevalence can clearly be influenced by changes in factors other than the physical condition of the population. Peoples' awareness of disease states can change with improvements in diagnosis. In addition, cohorts who use more health care services are likely to be more knowledgeable about disease states because of their contact with the medical establishment. It is also possible that people have medicalized some conditions that were once regarded as "aging" and not disease. For instance, people may now be more likely than in the past to declare aches and pains to be arthritis and osteoporosis. There has been a small decrease in the percent of the older population institutionalized during this time (from 6.3 to 5.2 percent) [Manton, Stallard, and Corder 1997]. While this small change is not enough to account for the changes observed here, it does increase the proportion of the population with health problems.

The results presented in this paper are actually fairly consistent with the existing literature. Results are quite similar to those for France which are also based on the comparison of two cross-sectional surveys in a similar period [Robine, Mormiche, Sermet 1998]. Both studies show some improvement in disability accompanied by increases in disease prevalence. Studies using different data sets, time periods, and sometimes different age groups for the United States have also resulted in generally similar findings on changes in disease prevalence as well as disability and functioning changes.

Findings for the older U.S. population using annual cross-sectional surveys [Crimmins, Saito, and Reynolds 1997] and two longitudinal surveys [Crimmins, Saito, and Reynolds 1997, Manton, Corder, and Stallard 1993, Manton, Stallard, and Corder 1995, Manton, Stallard, and Corder 1997] have shown a statistically significant but small decrease in the overall level of IADL and ADL disability, primarily due to decreased IADL disability with mixed results on the trends in ADL disability. On the other hand, studies including younger age groups and less severe disability have tended to find improvement in functioning and less disability and some reduction in the prevalence of major diseases including arthritis and heart disease [Crimmins, Saito, and Ingegneri 1997, Crimmins, Reynolds, and Saito 1999, Freedman and Martin 1998, Reynolds, Crimmins, Saito 1998].

Results from the National Long Term Care Survey on changes in disease prevalence are also fairly similar. Manton et al. [Manton, Stallard, and Corder 1995] present the prevalence of a number of diseases in the 65 + noninstitutionalized population for two dates, 1982 and 1989. Estimates of disease prevalence without controls for disability, produce findings similar to our results. They report increases in prevalence for diabetes, cancer, and arthritis; no change in stroke prevalence, which we find to be true only for women; and a small increase in hypertension, which we find only for men. We differ in our findings for heart disease. Where the study of the 65+ population finds no change in prevalence, looking at the 70+ population we find increase. This difference may be due to the age composition of the samples because in earlier studies of the 60-69 year old population we find decreases in cardiovascular conditions in a similar time period [Crimmins, Reynolds, and Saito 1999]. This pattern of improvement in disease prevalence among younger persons and increase among older persons was also found in the French study [Robine, Mormiche, Sermet 1998].

The fact that more older people now report that they have the major mortal diseases of old age means that more people are likely to be undergoing treatment or monitoring of their conditions. More people are likely to be undergoing treatment for multiple conditions. One of the effects of the success at reducing mortality may be a higher prevalence of disease and greater use of medical services.


Results Acknowledgements

Change in the Prevalence of Diseases among Older Americans: 1984-1994
Eileen M. Crimmins, Yasuhiko Saito
© 2000 Max-Planck-Gesellschaft ISSN 1435-9871