School's Out: What Are Urban Children Doing? The Summer Activity Study of Somerville Youth (SASSY)

Background: Research indicates that in the United States, children experience healthier BMI and fitness levels during school vs. summer, but research is limited. The primary goal of this pilot study was to assess where children spend their time during the months that school is not in session and to learn about the different types of activities they engage in within different care settings. A secondary goal of this pilot study was to learn what children eat during the summer months.

Methods: A nine-week summer study of 57 parents of second and third grade students was conducted in an economically, racial/ethnically and linguistically diverse US urban city. Weekly telephone interviews queried time and activities spent on/in 1) the main caregiver's care 2) someone else's care 3) vacation 4) and camp. Activities were categorised as sedentary, light, moderate, or vigorous (0-3 scale). For each child, a mean activity level was calculated and weighted for proportion of time spent in each care situation, yielding a weighted activity index. On the last phone call, parents answered questions about their child's diet over the summer. Two post-study focus groups were conducted to help interpret findings from the weekly activity interviews.

Results: The mean activity index was 1.05 +/- 0.32 and differed between gender (p = 0.07), education (p = 0.08) and primary language spoken in the household (p = 0.01). Children who spent a greater percentage of time in parent care had on average a lower activity index (beta = -0.004, p = 0.01) while children who spent a greater percentage of time in camp had a higher activity index (beta = 0.004, p = 0.03). When stratified into type of camp, percentage of time spent in active camp was also positively associated with mean activity index (beta = 0.005, p =< 0.001). With regards to diet, after adjusting for maternal education, children who attended less than five weeks of camp were four times more likely to eat their meals in front of the TV often/almost all of the time (OR = 4.0, 95%CI 1.0-16.2, p < 0.06).

Conclusions: Summer activities and some dietary behaviours are influenced by situation of care and socio-demographic characteristics. In particular, children who spend a greater proportion of time in structured environments appear to be more active. We believe that this pilot study is an important first step in our understanding of what children do during the summer months.