In this prospective cohort study of Korean elderly subjects, low life satisfaction was associated with long-term mortality. It was also associated with cardiovascular disease-related mortality. These data are largely consistent with the findings of previous studies, such as a Finnish prospective study [4
], a Finnish study of elderly people in their 80s [6
], a Berlin aging study [7
], and a Taiwanese prospective study [8
]. The Finnish prospective study showed that low life satisfaction was associated with the increased risk of total death in men, but only of injury death in women [4
]. In contrast, the present study shows that the relationship between low life satisfaction and mortality risk was significant in both men and women. Many previous studies of LSI and mortality categorized deaths into total death, disease death, and injury death, whereas the present study evaluated the relationship according to the specific diseases that caused deaths. The present study shows that LSI was associated with a risk of cardiovascular disease, but not of cancer. The outcome suggests a strong relationship between life satisfaction and cardiovascular disease.
The present study used the LSI-A, a 20-item questionnaire, as an assessment tool of quality of life satisfaction. The Taiwanese study also used the LSI-A, and it is known to be useful in assessing overall quality of life satisfaction [8
]. Previous studies found that self-reported life satisfaction was associated with cognitive function, formal support, social interaction, social relationships, socioeconomic status, age, marital status, health status, religion, economic status, and/or living arrangements [19
]. The present study also found that life satisfaction was associated with chronic disease conditions, education level, cognitive function, and disability in ADL and IADL in Korean elderly men.
This study has several limitations. First, we adjusted BMI as a covariate. However, other classical biological risk factors for cardiovascular-related death, such as hypertension, diabetes, and dyslipidemia, were not included in this study. Second, non-inclusion of variables, such as depressive symptoms, sense of well-being, and social support, may be a limitation, although well-being is a dimension measured by the LSI-A in a comprehensive sense. Third, deaths from diseases other than cardiovascular disease, cancer, and all causes could not be analyzed due to an insufficient number of cases. More causes of death need to be analyzed in further studies. Fourth, selection bias is often an issue in studies of life satisfaction. The subjects in this study were an elderly group residing in the same administrative and geographical community, and approximately 85% of them engaged in agriculture, thus there was a great social and cultural similarity among them.
In conclusion, our findings indicate an association between low life satisfaction and long-term mortality among elderly people, in particular cardiovascular disease mortality. Further studies are necessary to describe the relationship between life satisfaction and mortality risk for a wider range of diseases.