Life satisfaction in late life: Markers and predictors of level and change among 80+ year olds

Abstract: The overall aim of the thesis was to examine within-individual and between-individual changes in life satisfaction in the oldest-old using different time metrics and to study life satisfaction within the context of psychosocial and health-related variables. Data were obtained from the Swedish OCTO-Twin Study of individuals aged 80 and older who were able to complete the Life Satisfaction Index-Z (LSI-Z). In Study I the association between life satisfaction and scales and questions regarding demographics, self-rated overall health and medically based health, functional capacity (instrumental and personal activities of daily living), cognitive function, depression, locus of control, and social network was investigated. Analyses indicated that social network quality, self-rated overall health, sense of being in control of one’s life, widowhood, and depressive symptoms were associated with life satisfaction. A gender-specific pattern was found; self-rated overall health and depressive symptoms were related to life satisfaction in women, whereas widowhood was significantly associated with lower satisfaction among men. In Study II the associates identified in Study I; perceived quality of social network, self-rated overall health, depressive symptoms, locus of control, and widowhood, in addition to financial satisfaction and the personality traits neuroticism and extraversion, were investigated as predictors of change in life satisfaction across four measurements over a 6-year period. Growth curve analysis showed a fairly consistent significant linear decline in life satisfaction. Certain markers predicted decrease in life satisfaction; the loss of spouse, particularly in men, and higher levels of depressive symptoms. Results suggest that life satisfaction is influenced by changes in psychosocial variables although there is an overall stability in level of life satisfaction in the oldest old. In Study III different time metrics were examined in the study of late life changes in life satisfaction. Findings of age-graded stability of life satisfaction, despite health-related losses distinctive of the oldest old, suggest that mortality-related processes could be more influential than chronological age. The study investigated changes in life satisfaction at 4 measurement occasions over a 6-year period using two competing parameterizations of time, chronological age and time-to-death. Growth curve analyses showed a linear decrease in both time-structures, but the time-to-death metric revealed a significantly better model fit. Notably, age, gender, SES, years to death, level or change in overall load of disease and self-rated health did not predict time-to-death related changes. Lower overall disease load was, however, related to higher levels of life satisfaction. In individuals with higher disease load, an external locus of control was related to lower satisfaction with life. Among those who rated their health as poor, a higher level of neuroticism was related to lower life satisfaction. The results suggest that a time-to-death metric was superior to chronological age to predict change in life satisfaction. In Study IV the relationships between life satisfaction and 25 specific chronic diagnoses were investigated. Problems with sleep, urinary incontinence and stroke were significantly related to life satisfaction in both men and women. Among men, angina pectoris and eczema were related to lower life satisfaction, whereas among women peptic ulcer was related to lower life satisfaction. The results confirm previous findings of a weak relationship between medically based measures of health and life satisfaction. However, health care and future studies of health and life satisfaction need to recognize and address that the meaning and consequences of various diseases may differ among individuals and that gender differences should be considered in this context. Overall findings from the thesis demonstrate a homogenous decline in life satisfaction in the oldest-old. Despite health-related losses, social network and personal resources accounted for substantial inter-individual differences in life satisfaction. The thesis demonstrated the need to analyse associates of life satisfaction at the intra-individual level and within a broader context of psychosocial and health-related variables also in late life.