Development and application of a scoring and visualization approach for 24-hour movement behaviours: an example based on social-emotional development in early years children

Abstract

Abstract Background Current methods for assessing the healthfulness of 24-hour movement behaviours (sleep, sedentary time, light physical activity, moderate-to-vigorous physical activity) use binary classifications that fail to capture their continuous and compositional nature. This study introduces a percentile-based scoring and visualization approach to evaluate the healthfulness of movement behaviour time-use compositions, using social-emotional development in early childhood as an example. Methods This cross-sectional study includes 560 children aged 1.2–2.9 years and 1,500 children aged 3.0-4.9 years from Sleep and Activity Database for the Early Years (SADEY), an international accelerometer repository of young children’s movement behaviours. Sedentary time, light physical activity, and moderate-to-vigorous physical activity were measured using accelerometers. Sleep duration was parent-reported. Social-emotional development was assessed using age- and sex-normalized scores from the Strengths and Difficulties Questionnaire. Linear regression models with compositional covariates were used to model associations between movement behaviours and Strengths and Difficulties Questionnaire scores. Representative grids containing all possible time-use compositions (in 5 min/d increments) of sleep, sedentary time, light physical activity, and moderate-to-vigorous physical activity were developed. The regression models were applied to each time-use composition in the grid, and the predicted scores were ranked to create percentile scores for different movement behaviour time-use compositions. Results The 24-hour movement behaviour composition was associated with all five Strengths and Difficulties Questionnaire scores in both age groups (p ≤ 0.01). The grids contained 17,577 and 16,535 possible time-use compositions for 1–2 and 3–4-year-olds, respectively. Time-use compositions ranked at the 0th percentile had the least sleep and highest sedentary time, while those ranked at the 100th percentile had the most sleep and least sedentary time. Across the central range of the percentile score distribution (e.g., rankings between the 25th to 75th percentiles), some very different time-use compositions had the same percentile score. Interactive visualization tools were presented to enable real-time exploration of percentile scores for various movement behaviour time-use compositions. Conclusions This study introduces a novel approach to evaluate the health benefits of movement behaviours. This approach moves beyond traditional binary cutoffs to recognize the gradual improvements in health that occur with small changes in behaviours, and that there are multiple pathways to achieving the same health benefits.