Changes in physical activity over time in young children: a longitudinal study using accelerometers.

Previous research has suggested that marked declines in physical activity occur during the preschool years, and across the transition into school. However, longitudinal studies using objective measures of activity have been limited by sample size and length of follow-up. The aims of this study were to determine how overall activity and time in different intensities of activity change in children followed from 3 to 7 years. Children (n = 242) wore Actical accelerometers at 3, 4, 5, 5.5, 6.5 and 7 years of age during all waking and sleeping hours for a minimum of 5 days. Time in sedentary (S), light (L), moderate (M), and vigorous (V) physical activity was determined using available cut points. Data were analyzed using a mixed model and expressed as counts per minute (cpm, overall activity) and the ratio of active time to sedentary time (LMV:S), adjusted for multiple confounders including sex, age, time worn, and weather. At 5 years, physical activity had declined substantially to around half that observed at 3 years. Although starting school was associated with a further short-term (6-month) decline in activity (cpm) in both boys (difference; 95% CI: -98; -149, -46) and girls (-124; -174, -74, both P<0.001), this proved to be relatively transient; activity levels were similar at 6-7 years as they were just prior to starting school. Boys were more physically active than girls as indicated by an overall 12% (95% CI: 2, 22%) higher ratio of active to sedentary time (P = 0.014), but the pattern of this difference did not change from 3 to 7 years. Time worn and weather variables were significant predictors of activity. In conclusion, both boys and girls show a marked decline in activity from 3 to 4 years of age, a decrease that is essentially maintained through to 7 years of age. Factors driving this marked decrease need to be determined to enable the development of targeted interventions.