How Long Would You Like to Live? A 25-year Prospective Observation of the Association Between Desired Longevity and Mortality

Background Desired longevity represents how strongly people esteem possible extensions of their own lifetime. The association between desired longevity and mortality risk has been reported in only one prospective study, which examined a small sample of older participants. We aimed to examine the hypothesis that desired longevity at middle-age predicted long-term survival. Methods In the prospective cohort study, residents aged 40–64 years were asked how long they would like to live and asked to choose one from three options: longer than, as long as, or shorter than the life expectancy. We used Cox proportional hazards model to calculate the hazard ratios (HRs) and 95% confidence intervals (CIs) for all-cause and cause-specific mortality according to the three groups for desired longevity, treating the “longer than” group as the reference. We conducted mediation analysis to investigate the mechanism for the association between desired longevity and mortality. Results We recruited 39,902 residents to the study. Risk of all-cause mortality was significantly higher in the “shorter than” group (HR 1.12; 95% CI, 1.04–1.21). The association was independent of sex, age, marital status, education, medical history, and health status. Regarding cause of death, mortality risk of cancer (HR 1.14; 95% CI, 1.00–1.29) and suicide (HR 2.15; 95% CI, 1.37–3.38) were also higher in the “shorter than” group. The unhealthy lifestyle mediated this association with all-cause mortality by 30.4%. Conclusion Shorter desired longevity was significantly associated with an increased risk of all-cause mortality, and mortality from cancer and suicide. Lifestyle behaviors particularly mediated this association.

eTable 2. Demographic characteristics of sample according to the desired longevity after including those with history of cancer, stroke, and myocardial infarction (n=41,758) a Dummy variables for body mass index: <18.5, 18.5-24.9,≥25.0 kg/m 2 and missing.
b Dummy variables for smoking status: never smoking, ever smoking, and missing.
c Dummy variables for drinking status: never drinking, ever drinking, and missing.
e Dummy variables for time spent walking: <1 h/day, ≥1 h/day, and missing.
f Dummy variables for eating breakfast: Yes, No, and missing.
g Combined a to f .* P<0.005.

eFigure 1 .
Lifestyle behavior models of the association between desired longevity and all -cause mortality using Sobel test Lifestyle behavior models of the association between desired longevity and all-cause mortality after including those with history of cancer, stroke, and myocardial infarction using Sobel test Lifestyle behavior models of the association between desired longevity and mortality of cancer using Sobel test Lifestyle behavior models of the association between desired longevity and mortality of suicide using Sobel test Baseline lifestyle characteristics according to the desired longevity stratified by sex after including those with history of cancer, stroke, and myocardial infarction (n=41,758) Hazard ratios and 95% confidence intervals for all-cause mortality according to desired longevity after including those with history of cancer, stroke, and myocardial infarction ( n=41,758) Base model hazard ratios and 95% confidence intervals for cause-specific mortality according to desired longevity including those with history of cancer, stroke, and myocardial infarction (n=41,758) Hazard ratios and 95% confidence intervals for all-cause mortality according to desired longevity after excluding all deaths within the first two years (n=39,727) Base model hazard ratios and 95% confidence intervals for all -cause mortality according to desired longevity stratified by past medical history Mediation effects of lifestyle behaviors on the association between desired longevity and mortality including those with history of cancer, stroke, and myocardial infarction (n=41,758) Mediation effects of lifestyle behaviors on the association between desired longevity and mortality of cancer and suicide (n=39,902)Base model HR: adjusted for age (continuous variable), sex (men or women), marital status (married, divorced/widowed, or single), education ( in school until age ≤15 years, age 16-18 years, or age ≥19 years).