21st Dec 2023

Introduction: Inconsistencies in estimated levels of physical activity (PA) and sedentary time (ST) in young children have partly been attributed to the measurement of the behaviours, including the various cut-points applied when processing accelerometer data. This systematic review and meta-analysis aimed to compare estimated levels of PA and ST of young children (3-5 years) based on various age-specific accelerometer cut-points. Methods: A systematic review was conducted to identify studies assessing accelerometer estimated levels of PA (light PA [LPA], moderate-to-vigorous PA [MVPA]) or ST using multiple accelerometer cut-points in a sample of preschool-aged children. Data were extracted and risk of bias assessed for all included studies. Random effects meta-analysis was used to estimate pooled effects for unique combinations of accelerometer cut-points for each movement behaviour outcome. Results: Twenty-four studies, reporting on 18 unique samples, were included. Results demonstrated substantial variability in estimates of PA and ST across different cut-points. Of the 17 assessed cut-points, few showed similarity on estimates of the behaviours; Evenson (2008) and Pate (2006) were most similar for the assessment of ST and LPA. Pate (2006) cut-points consistently demonstrated the highest levels of MVPA. Conclusions: This review illustrates the limitations of having numerous accelerometer cut-points to determine PA and ST of young children, highlighting the substantial differences in estimates produced. Whilst similar estimates were identified based on a few cut-points; accelerometer cut-points were generally not comparable. This has implications for movement behaviour research, which impacts policy and recommendations. Research identifying an optimal approach to estimating movement behaviour outcomes in young children is required, while considering congruence with past and future research.