Depressive symptoms and objectively measured physical activity and sedentary behaviour throughout adolescence: a prospective cohort study

BACKGROUND: Identifying modifiable risk factors is essential to reduce the prevalence adolescent depression. Self-report data suggest that physical activity and sedentary behaviour might be associated with depressive symptoms in adolescents. We examined associations between depressive symptoms and objectively measured physical activity and sedentary behaviour in adolescents. METHODS: From a population-based cohort of adolescents whose mothers were invited to participate in the Avon Longitudinal Study of Parents and Children (ALSPAC) study, we included participants with at least one accelerometer recording and a Clinical Interview Schedule-Revised (CIS-R) depression score at age 17·8 years (reported as age 18 years hereafter). Amounts of time spent in sedentary behaviour and physical activity (light or moderate-to-vigorous) were measured with accelerometers at around 12 years, 14 years, and 16 years of age. Total physical activity was also recorded as count per minute (CPM), with raw accelerometer counts averaged over 60 s epochs. Associations between the physical activity and sedentary behaviour variables and depression (CIS-R) scores at age 18 years were analysed with regression and group-based trajectory modelling. FINDINGS: 4257 adolescents from the 14 901 enrolled in the ALSPAC study had a CIS-R depression score at age 18 years. Longitudinal analyses included 2486 participants at age 12 years, 1938 at age 14 years, and 1220 at age 16 years. Total follow-up time was 6 years. Total physical activity decreased between 12 years and 16 years of age, driven by decreasing durations of light activity (mean 325·66 min/day [SD 58·09] at 12 years; 244·94 min/day [55·08] at 16 years) and increasing sedentary behaviour (430·99 min/day [65·80]; 523·02 min/day [65·25]). Higher depression scores at 18 years were associated with a 60 min/day increase in sedentary behaviour at 12 years (incidence rate ratio [IRR] 1·111 [95% CI 1·051–1·176]), 14 years (1·080 [1·012–1·152]), and 16 years of age (1·107 [1·015–1·208]). Depression scores at 18 years were lower for every additional 60 min/day of light activity at 12 years (0·904 [0·850–0·961]), 14 years (0·922 [0·857–0·992]), and 16 years of age (0·889 [0·809–0·974]). Group-based trajectory modelling across 12–16 years of age identified three latent subgroups of sedentary behaviour and activity levels. Depression scores were higher in those with persistently high (IRR 1·282 [95% CI 1·061–1·548]) and persistently average (1·249 [1·078–1·446]) sedentary behaviour compared with those with persistently low sedentary behaviour, and were lower in those with persistently high levels of light activity (0·804 [0·652–0·990]) compared with those with persistently low levels of light activity. Moderate-to-vigorous physical activity (per 15 min/day increase) at age 12 years (0·910 [0·857–0·966]) and total physical activity (per 100 CPM increase) at ages 12 years (0·941 [0·910–0·972]) and 14 years (0·965 [0·932–0·999]), were negatively associated with depressive symptoms. INTERPRETATION: Sedentary behaviour displaces light activity throughout adolescence, and is associated with a greater risk of depressive symptoms at 18 years of age. Increasing light activity and decreasing sedentary behaviour during adolescence could be an important target for public health interventions aimed at reducing the prevalence of depression. FUNDING: Details of funding are provided in the Acknowledgments.