24-hour movement behaviour patterns differ across depressive symptom trajectories in community-dwelling older adults: a group-based trajectory model and compositional data analysis

Abstract

Depressive symptoms in older adults change over time and do not follow a single uniform pattern. Whether 24-hour movement behaviour composition differs across depressive symptom trajectories remains unclear. This longitudinal secondary analysis used data from the Survey of Health, Ageing and Retirement in Europe (SHARE). Depressive symptoms across five waves were analysed using group-based trajectory modelling. Accelerometer-derived sleep, sedentary behaviour, light physical activity, and moderate-to-vigorous physical activity were treated as compositional data. Isometric log-ratio transformation and compositional regression were used to examine differences in 24-hour movement behaviour composition across trajectory groups, adjusting for sociodemographic and health covariates. Five depressive symptom trajectories were identified: persistently very low, stable low, rapidly decreasing, increasing, and persistently high. Overall, trajectory membership is associated with 24-hour movement behaviour composition (p = 0.003). The persistently high depression group retained a lower relative light physical activity component and a higher relative sleep component than the stable low group; differences in moderate-to-vigorous physical activity and sedentary behaviour were attenuated. Long-term depressive symptom trajectories are associated with distinct 24-hour movement behaviour compositions in older adults. Persistently high depressive symptoms are associated with distinct daily time-use patterns, highlighting the importance of integrated behavioural strategies across the whole 24-hour day.