Predicting Oxygen Uptake From Accelerometer Counts During Free-living Activities In Children

Abstract

Accelerometry is a valid method of objectively assessing physical activity (PA). To be able to further explore the association between accelerometer counts and health outcomes, it is important to know how accelerometer counts (measuring movement) relate to oxygen uptake (VO2) (a measure of energy expenditure). Most previous studies have examined the ability of accelerometers to estimate VO2 in controlled laboratory settings rather than in individuals performing free-living activities. PURPOSE To predict VO2 uptake from accelerometer counts during free-living activities. METHODS Children aged 11 to 12 were recruited from a large birth cohort- the Avon Longitudinal Study of Parents and Children (ALSPAC). VO2 and accelerometry data were collected during the following activities, each of which lasted 5 minutes: lying, sitting playing a handheld computer game, walking slowly, walking briskly, jogging and hopscotch. The mean values from minutes 3 1/2 to 4 1/2 were used in the analysis. Multi-level modelling was used to derive regression co-efficients with accelerometer counts per minute being used to predict exercise VO2 (mL·kg−1·min−1). RESULTS Data on 24 boys and 26 girls (mean age 12.27 years) were obtained. Table 1 shows mean (SD) accelerometer counts for each activity. Accelerometer counts were zero for 90% of subjects for lying and 62% for sitting, therefore these data were not included in the models.TableA random intercepts model was fitted to account for within-subject multiple observations. The resulting regression equation was VO2 = 6.018 + 0.00461 × cpm. Models were similar for males, females, and also for those above and below the median level of body mass index. CONCLUSION VO2 is associated with accelerometer counts in 11 to 12 year old children, participating in free-living activities of varying intensities. Supported by NIH Grant R01 HL071248-01A1