Low life satisfaction predicts mortality in a cohort study of very old men-the Manitoba follow up study

(1) To determine the life satisfaction (LS) of members of a cohort study of aging men; (2) to determine if LS predicts death at various ages amongst members of a cohort study.

We analyzed a prospective cohort study of men who qualified for air crew training in the Second World War. In 2005, 860 participants with a mean age of 85, were alive and responded to the annual questionnaire - which is sent to participants living in the community. Since 2005, LS was measured annually. Kaplan-Meier plots were constructed, and median survival times were calculated. Cox proportional hazards models were constructed comparing those with 'Poor/Bad'; 'Fair'; 'Good; and 'Very Good' to 'Excellent' LS.

Throughout the study period from 2005 to 2024, most men were satisfied with life. Lower LS predicted mortality throughout the course of the study, even at very old ages and even after accounting for the effect of chronic diseases.

LS appears relatively stable in community living old men, and predicts mortality even into advanced age.