Background: The retirement transition provides a window of opportunity for the optimization of activity, sedentary and sleep behaviors. Identifying groups at risk for non-favorable changes is important in this matter. There are indications that lower socio-economic position (SEP) adults might be more prone to non-favorable changes. However, previous studies mainly used self-reported behaviors and only one indicator of SEP. The purpose of the present study was therefore to examine the association of SEP with changes in device-measured physical activity, SB and sleep during the retirement transition in adults in Flanders (Belgium) in a longitudinal study.

Methods: The behaviors were measured pre-retirement and at three, six and twelve months post-retirement with a wrist-worn accelerometer (n = 96). The raw acceleration data were processed using the R package GGIR and analysed using compositional data analysis in linear mixed models including the SEP indicators education, occupation and income. Additionally, changes in intensity gradient and average acceleration were examined.

Results: The results showed that on average, physical activity was stable, sleep increased (+ 18 min) and SB decreased (- 15 min). The intensity gradient and average acceleration did not change significantly. The higher income group had significantly more favorable changes in movement behaviors compared to the lower income group. More specifically, they increased physical activity and shifted towards more intense physical activity, while the lower income group did not. The higher education and occupation groups showed a non-significant trend towards more favorable changes. The changes occurred mainly between pre-retirement and three months post-retirement and were relatively stable afterwards.

Conclusion: The behaviors shifted towards more healthy behaviors in general, with stable physical activity, a decrease in SB and an increase in sleep. The changes in the behaviors were more favorable for retirees with higher SEP compared to retirees with lower lower socio-economic position. There seems to be a need for strategies to improve 24-h movement behaviors of lower SEP adults during the retirement transition. However, our results are based on a small sample and should be validated in larger studies.