Abstract / Summary
This systematic review aimed to evaluate the validity and reliability of accelerometers for measuring sedentary behavior (SB) and physical activity (PA) in preschool-aged children (up to 6 years) during free-living conditions. A systematic search was conducted in PubMed, ProQuest, SCOPUS, Web of Science, and SPORTDiscus for articles published up to October 29, 2025, following PRISMA guidelines. Methodological quality was evaluated using the Methodological Index for Non-Randomized Studies (MINORS) scale. Twenty-seven studies met the inclusion criteria, with the majority classified as high quality. Accelerometers demonstrated moderate-to-high validity for classifying SB and moderate-to-vigorous physical activity (MVPA), with ROC-AUC values ranging from 0.56 to 0.99 for SB and 0.57 to 0.96 for MVPA, Cohen’s kappa up to 0.78, and sensitivity and specificity ranging from 8.9–100% and 17.2–99.4%, respectively, depending on intensity and cut-point used. These results were consistent across ActiGraph and non-ActiGraph devices. However, precision for quantifying light-intensity PA (LPA) was limited, with weaker correlations (ρ = 0.21–0.63) and lower classification accuracy than for SB and MVPA. Reliability for assessing habitual SB and PA improved with monitoring duration, reaching high stability (ICC = 0.93) after 7 days. Accelerometry is a valid method for assessing SB and MVPA in free-living preschool children, though it shows limitations for LPA. To maximize accuracy and reliability, researchers should implement protocols of at least 7 consecutive days using short epoch lengths (5–15 sec) and carefully select age-appropriate cut-points. Standardization of data processing methods is needed to improve comparability across studies.