Life-course pathways to exceptional longevity: Evidence from the Lothian Birth Cohort of 1921

Abstract

Background Longevity, a hallmark of successful ageing, is a multifactorial trait with influences from birth onwards. However, limited evidence exists on the pathways linking diverse life-course exposures to longevity, especially within a single cohort. Methods We investigated associations between life-course factors and longevity among community-dwelling adults aged 79 (N=547) from the Lothian Birth Cohort 1921 with a mortality follow-up of 24 years. Cox proportional hazards and structural equation (path) models were used to explore how factors from early-life (social class, childhood IQ, education), mid-life (social class), later-life (health, lifestyle, psychosocial well-being), as well as sex, personality and APOE e4 status, influence survival time in days. Results During follow-up (1999-2023), 538 participants (98%) died (mean age of death=89.3 years) and 9 survived (mean age=101.6 years). Factors associated with lower mortality risk in the multivariable Cox model were higher cognitive function (HR=0.72; 95% CI:0.59-0.88), better physical function (HR=0.61; 95% CI:0.44-0.85), and greater physical activity (HR=0.81; 95% CI 0.71-0.92), while history of cancer was associated with higher mortality risk (HR=1.84; 95% CI:1.22-2.77). The life-course path model identified the same direct predictors, with additional contributions from female sex and non-smoking status, to greater longevity. Early- and mid-life factors (IQ, education, social class), and emotional stability, conscientiousness, and female sex, were indirectly and positively associated with survival trajectories via multiple dimensions of adult health. Conclusions In understanding why people live to very old ages it is necessary to consider factors from throughout the life course, and to include demographic, psychosocial, and health variables.