Measuring physical activity during US Army Basic Combat Training: a comparison of 3 methods

Background An understanding of the demands of physical activity (PA) during US Army Basic Combat Training (BCT) is necessary to support Soldier readiness and resilience. The purpose of this study was to determine the agreement among 3 different PA measurement instruments in the BCT environment.

Methods Twenty-four recruits from each of 11 companies wore an ActiGraph accelerometer (Actigraph, LLC, Pensacola, FL) and completed a daily PA log during 8 weeks of BCT at 2 different training sites. The PA of one recruit from each company was recorded using PAtracker, an Army-developed direct observation tool. Information obtained from the accelerometer, PA log, and PAtracker included time spent in various types of PA, body positions, PA intensities, and external loads carried. Pearson product moment correlations were run to determine the strength of association between the ActiGraph and PAtracker for measures of PA intensity and between the PAtracker and daily PA log for measures of body position and PA type. The Bland-Altman method was used to assess the limits of agreement (LoA) between the measurement instruments.

Results Weak correlations (r=-0.052 to r=0.302) were found between the ActiGraph and PAtracker for PA intensity. Weak but positive correlations (r=0.033 to r=0.268) were found between the PAtracker and daily PA log for body position and type of PA. The 95% LoA for the ActiGraph and PAtracker for PA intensity were in disagreement. The 95% LoA for the PAtracker and daily PA log for standing and running and all PA types were in disagreement; sitting and walking were in agreement.

Conclusions The ActiGraph accelerometer provided the best measure of the recruits’ PA intensity while the PAtracker and daily PA log were best for capturing body position and type of PA in the BCT environment. The use of multiple PA measurement instruments in this study was necessary to best characterize the physical demands of BCT.