OBJECTIVE: Investigations of cohort differences in relationships between education and health tend to focus on mortality or self-reported health. We report one of the first analyses of cohort differences in relationships between education and objective measures of functional abilities across the lifespan.

METHOD: Up to 26 years of follow-up data were available from 859 adults from the Swedish Adoption/Twin Study of Aging. The sample was divided into two cohorts by birth year: 1900-1924 and 1925-1948. Latent growth curve models (LGCM) were compared across cohort and educational levels.

RESULTS: LGCM indicated divergence between adults with lower and higher educational attainment in longitudinal trajectories of change with age in the Balance and Flexibility factors for the later born cohort only.

DISCUSSION: Results support the cumulative advantage theory and suggest that education-health disparities are increasing in recent cohorts, even in counties with national health care systems and strong support of education.