Temporal trends in late-life depressive symptom trajectories: a longitudinal population-based cohort study from...

While studies of depression prevalence in Sweden suggest improving trends over time among older adults, longitudinal approaches can provide a more nuanced perspective on the changing burden of late-life depression. This study mapped 19-year trajectories of depressive symptoms among cohorts of older adults born between 1904 and 1951 residing in Stockholm, Sweden.

We examined 2960 individuals aged 60-99 years from the ongoing longitudinal population-based Swedish National Study on Aging and Care in Kungsholmen (SNAC-K) representing eight partially overlapping birth cohorts followed up between 2001 and 2020. Depressive symptoms were measured using the Montgomery-Åsberg Depression Rating Scale. Linear mixed-effect models with cubic splines were used to derive depressive symptom trajectories across age and birth cohort. Baseline education, functional impairment and loneliness were considered as potential explanatory factors for any temporal patterns.

Depressive symptoms increased with age in all birth cohorts. Age-related symptom increase accelerated in each later-born cohort (eg., at age 84, predicted MADRS was 2.9 (95% CI 2.6 to 3.2) in the birth cohort 1920-1924, 3.2 (95% CI 3.0 to 3.5) in the birth cohort 1929-1933 and 4.0 (95% CI 3.8 to 4.3) in the birth cohort 1935-1939). Among similarly aged individuals separated in time, a worsening burden over time was noted among the 60-year-olds but not the 81-year-olds. These patterns persisted after considering baseline education, functional impairment and loneliness.

Despite some suggestions of declining prevalence, when considered longitudinally, late-life depressive burden appears to be increasing more recently in this Stockholm-based population. Given the growing number of older adults in Sweden, these trends carry important public health implications for prevention and management of depression in late life.