Levels of and changes in life satisfaction predict mortality hazards: Disentangling the role of physical health, perceived control, and social orientation

It is well-documented that well-being typically evinces precipitous decrements at the end of life. However, research has primarily taken a postdictive approach by knowing the outcome (date of death) and aligning in retrospect how well-being has changed for people with documented death events. In the present study, we made use of a predictive approach by examining whether and how levels of and changes in life satisfaction prospectively predict mortality hazards and delineate the role of contributing factors, including health, perceived control, and social orientation. To do so, we applied shared parameter growth-survival models to 20-year longitudinal data from 10,597 participants (n = 1,560 or 15% deceased; age at baseline: M = 44 years, SD = 17, range: 18-98 years) from the national German Socio-Economic Panel Study (SOEP). Our findings showed that lower levels and steeper declines of life satisfaction each uniquely predicted higher mortality risks. Results also reveal moderating effects of age and perceived control: Life satisfaction levels and changes had stronger predictive effects for mortality hazards among older adults. Perceived control is associated with lower mortality hazards; however, this effect is diminished for those who experience accelerated life satisfaction decline. Variance decomposition suggests that predictive effects of life satisfaction trajectories were partially unique (3-6%) and partially shared with physical health, perceived control, and social orientation (16-19 %). Our discussion focuses on the strengths and challenges of a predictive approach to link developmental changes (in life satisfaction) to mortality hazards and considers implications of our findings for healthy aging.