A systematic review in device-measured physical activity during active commuting to/from school: practical considerations to assess when, where, and how much it occurs

Contribution of active commuting to and from school to device‐measured physical activity levels in young people: A systematic review and meta‐analysis

“…The use of GPS is convenient, but if it is not possible to use it, an activity diary could also be a good option. If neither of these two instruments are available, a time interval per participant should be adapted according to the distance they live from the school 13 ; (2) for the quantification of ACS‐PA, it is necessary to report the active mode of commuting to/from school used, since walking and cycling are both active modes of commuting, but they provide different PA intensities and benefits. For the assessment of PA, the most recommended device is the accelerometer, although in case the study population predominantly walks, pedometers might be used 13 .…”

“…In the current meta‐analysis, considering that the methodology used by each study may influence the results, meet the 48% of the PA recommendations could be equivalent to actively commuting both trip directions, with the distance commuted between the two trips (i.e., home‐school and school‐home) being around two kilometers 70 ; (2) no gray literature and studies that were not in English or Spanish were not included, and although the literature search was conducted in five databases, some eligible studies from other databases may have been missed; (3) in line with the meta‐analysis conducted by Martin et al, 7 it was not possible to examine the contribution of cycling to/from school to LPA and MVPA levels. Cycling is often underestimated due to the processing of accelerometry data and/or the placement of devices, among other factors; (4) many studies did not differentiate the mode of active commuting (i.e., walking or cycling) and, therefore, it is not possible to know which mode the results refer to; (5) dividing the studies by mode of active commuting and by trip direction did not allow to divide the results by age group (i.e., children and adolescents), gender, or country/geographic location; (6) the fact that there is no consensus on how to measure ACS‐PA using devices can lead to limitations in interpreting the results, as in the case of MVPA and LPA 13 …”

“…Following previous systematic reviews with meta‐analysis 30–32 and the results by Campos‐Garzón et al 13 which pointed out the existence of different methodologies that may influence the results of ACS‐PA, a random‐effects model was used to pool the PA results of the included studies (DerSimonian and Laird method) 33 using Stata (version 17.0; StataCorpo.) and the admetan procedure 34 .…”

“…Meta‐analysis results were reported differentiating by PA intensity (LPA or MVPA), differentiating by trip direction (home‐school trips, school‐home trips, or unspecified), and within each trip direction differentiating by mode of commuting (ACS, walking, cycling, or unspecified). Furthermore, the same analyses were replicated by including only those studies that used GPS or used a time interval up to 60 minutes, as these are the most used methodologies in the scientific literature for the trip identification 13 (see Figures S1 and S2 in Additional File 2).…”

“…Cycling is often underestimated due to the processing of accelerometry data and/or the placement of devices, among other factors; (4) many studies did not differentiate the mode of active commuting (i.e., walking or cycling) and, therefore, it is not possible to know which mode the results refer to; (5) dividing the studies by mode of active commuting and by trip direction did not allow to divide the results by age group (i.e., children and adolescents), gender, or country/geographic location; (6) the fact that there is no consensus on how to measure ACS-PA using devices can lead to limitations in interpreting the results, as in the case of MVPA and LPA. 13 The present study also shows several strengths: (1) last PRISMA guidelines were followed to ensure that the systematic review was conducted with adequate rigor. In addition, it was registered in PROSPERO and the systematic review protocol was also published 24 ; (2) the quality and risk of bias of the included studies were assessed, as they are different concepts, following the recommendations of Gunnell et al 23 Most studies had a low risk of bias (see Figures S3 and S4 in Additional File 3) and medium quality (see Tables S3 and S4 in Additional File 3); (3) a meta-analysis was conducted to contribute to the scientific literature on how much device-measured LPA and MVPA is related to ACS; (4) to best of our knowledge, it is the first systematic review and meta-analysis that separates the home-school and school-home trips, suggesting that they should be evaluated separately.…”