OBJECTIVE: Subjective well-being (SWB) is an important predictor of mortality. To date, surprisingly little is known about whether cognitive and emotional facets of SWB independently predict mortality, and whether such effects vary by age. METHOD: This study examined differential effects of life satisfaction (LS), positive affect (PA), and negative affect (NA) on mortality in a conjoint analysis using data of German adults between the ages of 40 to 85 years (N = 3,124). Effects of SWB facets were analyzed using Cox Proportional Hazards Models. RESULTS: LS and PA predicted mortality over and above sociodemographic factors and physical health (HRLS = .89, 95% CI = .79-1.00, p < .05, HRPA = .81, 95% CI = .70-.93, p < .05). However, this effect diminished when including self-rated health and physical activity. NA was not associated with mortality. Age group comparative analyses revealed that PA predicted mortality in older adults (65+) even after controlling for self-rated health and physical activity (HRPA = .82, 95% CI = .70-.97, p < .05). In middle-aged participants, no SWB indicator predicted mortality when controlling for covariates. CONCLUSIONS: These findings suggest a differential impact of cognitive and emotional well-being on mortality and suggest that in middle-aged adults the effects of SWB on mortality are attenuated by self-rated health and physical activity. The study underscores the importance of SWB for health and longevity, particularly for older adults. PsycINFO Database Record (c) 2011 APA, all rights reserved.
OBJECTIVE: Subjective well-being (SWB) is an important predictor of mortality. To date, surprisingly little is known about whether cognitive and emotional facets of SWB independently predict mortality, and whether such effects vary by age. METHOD: This study examined differential effects of life satisfaction (LS), positive affect (PA), and negative affect (NA) on mortality in a conjoint analysis using data of German adults between the ages of 40 to 85 years (N = 3,124). Effects of SWB facets were analyzed using Cox Proportional Hazards Models. RESULTS: LS and PA predicted mortality over and above sociodemographic factors and physical health (HRLS = .89, 95% CI = .79-1.00, p < .05, HRPA = .81, 95% CI = .70-.93, p < .05). However, this effect diminished when including self-rated health and physical activity. NA was not associated with mortality. Age group comparative analyses revealed that PA predicted mortality in older adults (65+) even after controlling for self-rated health and physical activity (HRPA = .82, 95% CI = .70-.97, p < .05). In middle-aged participants, no SWB indicator predicted mortality when controlling for covariates. CONCLUSIONS: These findings suggest a differential impact of cognitive and emotional well-being on mortality and suggest that in middle-aged adults the effects of SWB on mortality are attenuated by self-rated health and physical activity. The study underscores the importance of SWB for health and longevity, particularly for older adults. PsycINFO Database Record (c) 2011 APA, all rights reserved.