Objectives This study aimed to investigate how daily use of time changes across the retirement transition and how these changes may differ according to socio-demographic characteristics.

Study design This longitudinal cohort study was based on interviews with 124 people at pre-retirement and at three, six and 12 months after retirement.

Main outcome measures The Multimedia Activity Recall for Children and Adults (MARCA), a computer-assisted telephone interview, measured use of time. Random effects mixed modelling (REMM) was used to examine time use changes across retirement, and ANCOVA to investigate the extent of the change by sex, education and health.

Results Apart from the decrease in time spent in work across retirement (−122 min/day, p < 0.001), less time was also spent in both active and passive transport (−26 min/day, p < 0.001). There were significant increases in time spent on Chores (+55 min/day, p < 0.001), Screen time (+32 min/day, p < 0.001), Sleep (+32 min/day, p < 0.001), Quiet time (+17 min/day, p = 0.02), Self-care (+8 min/day, p = 0.01) and Physical activity (+7 min/day, p = 0.01). There was little variation in change between sex, education and health sub-groups. Most changes occurred at retirement, with time use patterns fairly stable between three and 12 months post-retirement.

Conclusions The small amount of time spent in physical activity, combined with less time spent in active transport and increased time spent watching television, indicate the need for a more active lifestyle.