RESUMO
Objective To analyze the feasibility of using walking indicators to estimate other physical activity data,so as to provide evidences for the application of using walking monitoring tools to comprehensively assess the daily physical activity level.Methods Eighty participants were assessed 7 kinds of physical activities through wearing Cosmed K4b2 portable indirect calorimetry system and Actigraph GT3X accelerometer simultaneously.The indirect calorimetry was used as the criterion for measuring energy expenditure to develop individual equations using the vector magnitude (VM) as the independent variable.Then all participants were randomly divided into an experimental group(n=60) and a validation group (n=20).Omron HJ-113 pedometer and GT3X were then simultaneously worn to monitor their daily physical activities,with the wearing time of 3～4 days for the experimental group and 7 days for the validation group.Physical activity data were calculated using the individual equations.The correlation between walking indicators and the other physical activity data was analyzed in the experimental group,and the indicators of high correlation and application value were selected to develop equations to predict physical activity indicators.Then the data of the validation group were used to validate these equations.Results The correlation coefficients between daily steps and the moderate-to-vigorous physical activity(MVPA) time,MVPA volume,MVPA10 time and MVPA10 volume were 0.723,0.730,0.681 and 0.677 respectively(P＜0.01 for all).The correlation coefficient of the daily aerobic walking time to MVPA10 time and MVPA10 volume were 0.752 and 0.759 respectively(P＜0.01 for both).Six equations were developed based on these correlations(r2=0.55～0.63),and paired t test showed that there were no significant differences between the physical activity data predicted using these equations and the data measured by GT3X(P＞0.05).Moreover,Bland-Altman plot showed there was little predicted error for the 6 equations.Conclusion Walking indicators can indirectly reflect the overall physical activity level,and the equations developed in this study can be used to monitor the physical activity of larger samples.