Objectives: To evaluate temporal dynamics between loneliness and both objective and subjective health (i.e. functional impairment and self-rated health) in mid- to late-adulthood.

Method: We applied bivariate dual-change-score models to longitudinal data from 3 Swedish twin studies (N = 1,939) to explore dynamic associations between loneliness and health across 3 age ranges (50–69, 70–81, and 82+ years) to investigate whether associations between loneliness and health change with age due to increasing incidence of chronic health conditions and bereavement.

Results: Results showed bidirectional associations between loneliness and both objective and subjective health, with adverse impacts of loneliness observed on subsequent subjective and objective health beginning at age 70. Associations between health and subsequent loneliness were observed after age 82 and varied for subjective and objective health, with subjective health associated with less loneliness and objective health associated with greater loneliness.

Conclusions: Our results indicate dynamic associations between loneliness and health with age in mid- to late-adulthood, with earlier impacts of loneliness on health and later impacts of health on loneliness that vary for objective and subjective measures of health. These findings suggest impacts of health on loneliness may arise later in life when worsening health or mobility interfere with social interaction.