Childhood obesity is a significant health concern, influenced by modifiable factors like physical activity (PA) and sleep. Structured routines, such as those provided by school and extracurricular activities, are thought to support consistent engagement in healthy behaviors. The "structured days hypothesis" suggests that structured school days offer opportunities for PA, reduced screen time, and regulated sleep schedules, compared to unstructured days. Studies have shown increased BMI during summer breaks, further supporting the importance of structured routines. The COVID-19 pandemic caused widespread school and activity closures, potentially disrupting children's PA and sleep patterns. While some studies have examined these effects, many relied on subjective data or short-term accelerometer data, leading to inconsistent findings. This study aimed to objectively assess the impact of COVID-19-related school closures on PA and sleep in obese children, a population already at higher risk for unhealthy behaviors.

Existing literature indicates that COVID-19 school closures globally led to decreased PA, increased sedentary time, and disrupted sleep in children. Studies using subjective data reported PA reductions ranging from 10–90 min/day and sleep changes including delayed bedtimes, shifted sleep midpoints, reduced sleep efficiency, and increased sleep disturbances. These varied findings may be due to differences in data collection methods, durations, and timing within the pandemic. Importantly, the rate of BMI increase in children doubled during the pandemic, particularly among those with pre-existing obesity. While some studies have addressed PA and sleep changes in obese children during the pandemic, most used subjective measures. Objective data from wearable sensors, offering more accurate and detailed information on PA and sleep, are needed to better understand these effects.

This secondary data analysis used data from the Hearts & Parks (H&P) randomized controlled trial (RCT), which enrolled 260 obese children (5-17 years) into an intervention or waitlist control group. Data collection began March 1, 2019, using Garmin Vivofit 3 wristbands for continuous, 24-hour PA and sleep monitoring. The study defined three periods: pre-closure (March 1, 2019 – March 14, 2020), during-closure (March 15, 2020 – March 31, 2021), and post-closure (April 1, 2021 – June 30, 2021). PA was measured as daily step counts, while sleep was assessed using Garmin's sleep detection algorithm. Data were included if participants had ≥60 days of valid data (≥41% wear time per day) during each period. The Mann-Whitney U test compared step counts and sleep duration between pre- and during-closure periods. Data were analyzed using Python in a protected computing environment. The Garmin Vivofit 3's accuracy in measuring step counts in children has been previously validated.

Of the 94 participants included (55% female, median age 9.7 years), pre-closure average daily step counts ranged from 8239 to 9521 steps (mean 8810 ± 453). During closure, this significantly decreased to 7155 ± 669 steps (p<0.05), a 1655-step decline. Step counts increased post-closure (8763 ± 325). Analysis of step counts during typical school hours (7 am-4 pm, weekdays, excluding summer) showed a 1973-step decline during closure (5441 ± 229 pre-closure vs. 3468 ± 542 during-closure, p<0.05). Summer months showed a similar trend. Sleep onset was delayed by an average of 1 h 45 min during closure compared to pre-closure, and wake time was delayed by a similar amount. Overall sleep duration decreased by 12 min (8.1 ± 0.17 h pre-closure vs. 7.9 ± 0.22 h during-closure, p=0.01).

This study provides objective evidence of significant reductions in PA and disruptions in sleep patterns among obese children during COVID-19 school closures. The observed changes strongly correlate with the timing of closures and reopenings, suggesting a causal link between disrupted routines and these health behaviors. The decreases in PA, particularly during school hours, likely reflect the loss of structured activity opportunities. The findings underscore the importance of structured school and extracurricular programs in promoting healthy PA and sleep habits, especially for obese children who are already at increased risk for these issues. These results highlight the need for strategies to maintain or improve PA and sleep during periods of disruption.

This study demonstrates the significant negative impact of COVID-19-related school closures on objectively measured PA and sleep in obese children. The findings support the crucial role of structured daily routines in promoting healthy lifestyle behaviors. Future research could explore interventions to mitigate the detrimental effects of routine disruptions on children's health, particularly during emergencies or periods of social distancing. Longitudinal studies tracking long-term effects are needed.

This study's findings are limited by its sample size, geographic location (Durham, NC), and focus on obese children. The use of Garmin Vivofit 3 data might not capture all types of physical activity. The study's reliance on a secondary data analysis from an RCT limits the generalizability of the results to other populations and contexts. While the study controlled for some confounders, residual confounding might exist due to the observational nature of some aspects of the data analysis.