Validity and Reliability of the activPAL4TM for Measurement of Body Postures and Stepping Activity in 6–12-Year-Old Children

Abstract

A link between inappropriate physical behaviour patterns (low physical activity and high sedentary behaviour) and poor health outcomes has been observed. To provide evidence to quantify this link, it is important to have valid and reliable assessment tools. This study aimed to assess the validity and reliability of the activPAL4TM monitor for distinguishing postures and measuring stepping activity of 6–12-year-old children. Thirteen children (8.5 ± 1.8 years) engaged in pre-determined standardised (12 min) and non-standardised (6 min) activities. Agreement, specificity and positive predictive value were assessed between the activPAL4TM and direct observation (DO) (nearest 0.1 s). Between-activPAL4TM (inter-device) and between-observer (inter-rater) reliability were determined. Detection of sitting and stepping time and forward purposeful step count were all within 5% of DO. Standing time was slightly overestimated (+10%) and fast walking/jogging steps underestimated (−20%). For non-standardised activities, activPAL4TM step count matched most closely to combined backward and forward purposeful steps; however, agreement varied widely. The activPAL4TM demonstrated high levels of reliability (ICC(1, 1) > 0.976), which were higher in some instances than could be achieved through direct observation (ICC(2, 1) > 0.851 for non-standardised activities). Overall, the activPAL4TM recorded standardised activities well. However, further work is required to establish the exact nature of steps counted by the activPAL4TM.

Agreement of Two Physical Behaviour Monitors for Characterising Posture and Stepping in Children Aged 6–12 Years

“…While only a small number of activPAL4 TM and activPAL3 TM monitors were used in the current study, their inter-device reliability has previously been reported as excellent [6,7], suggesting that the number of monitors was sufficient to avoid reliability issues having an impact on outcomes.…”

“…When comparing the outcomes derived from different generations of devices, it is important to understand the agreement between device outcomes to inform a comparison of values. The validity and reliability of the activPAL TM family of devices has been reviewed [2,3], demonstrating high levels of validity (in relation to direct observation) under standardised conditions (original activPAL TM [4]), but lower levels of validity in detecting slow stepping (original activPAL TM [5]) and jogging (original activPAL TM [6]) or in more challenging movement protocols (activPAL4 TM [7]). Also, its reliability has been shown to be high (inter-device ICC(1,1) > 0.976 for activPAL4 TM [7]); however, betweendevice agreement has not been studied extensively.…”

“…The validity and reliability of the activPAL TM family of devices has been reviewed [2,3], demonstrating high levels of validity (in relation to direct observation) under standardised conditions (original activPAL TM [4]), but lower levels of validity in detecting slow stepping (original activPAL TM [5]) and jogging (original activPAL TM [6]) or in more challenging movement protocols (activPAL4 TM [7]). Also, its reliability has been shown to be high (inter-device ICC(1,1) > 0.976 for activPAL4 TM [7]); however, betweendevice agreement has not been studied extensively. Agreement between the original activPAL TM and the activPAL3 TM suggested high levels of agreement for standardised activity protocols [8,9] but lower levels for more complex activities, e.g., involving more complex full-body movements or small intermittent stepping activity [8].…”