Effect of accelerometer assessment methods on the evaluation results of sedentary behaviour and physical activity in school-aged children
Objective To explore the effects of different positions,epoch lengths and cut points of accelerometers on the meas-urement of sedentary behaviour(SB),light physical activity(LPA)and moderate-vigorous physical activity(MVP A)in school-aged children,so as to provide a reference for optimizing the criteria of accelerometers to measure SB and physical activity in school-aged children.Methods From May 2021 to September 2022,110 school-aged children aged 8-12 years old were conven-ient selected from a primary school in Hangzhou,and the accelerometers were worn on the waist,back,thigh,and wrist during the school time,and the collected data were transformed into six datasets with different epoch lengths(1,5,10,15,30,and 60 s)and compared using 4 common cut-points developed by Puyau,Freedson,Evenson,and Pulsford(a total of 96 combinations)for classifying the percentages of time spent in SB,LPA,and MVPA.The effects of different positions,epoch lengths and cut points and their interactions on the assessment results were analyzed using a three-way repeated measures ANOVA,and the variability of multiple comparisons between groups was analyzed using a Bonferroni post-hoc test.Results Except for the interaction of cut point x position x epoch on the percentage of time spent in SB and the main effect of epoch on he percentage of time spent in MVPA were not statistically significant(F=0.66,1.18,P＞0.05),the remaining main effects and interactions of the three factors on the as-sessment results were all statistically significant(F=6.28-11 662.28,P＜0.01).The differences between the results of different po-sitions were statistically significant(F=90.98-308.79,P＜0.01).The percentages of time spent in SB and MVPA decreased with the increase of epoch lengths,while the percentage of time spent in LPA increased with the increase of epoch lengths(F=16.54-676.35,P＜0.01).The difference between the results using different cut points measurements was statistically significant(F=98.14-976.40,P＜0.01).Conclusions Accelerometer methodological factors including positions,epoch lengths and cut points will affect the estimates of SB and physical activity in school-aged children.Therefore,it is recommended that future studies need to optimize the applicability of accelerometer analysis methods in school-aged children,and the comparisons of results between studies need to be treated with caution.
Sedentary lifestyleMotor activityAnalysis of varianceChild
NETL
NSTL
万方数据
