Abstract presented at the Annual Meeting of the ACSM 2007, New Orleans, LA. USA

Miguel A. Calabro, Gregory J. Welk, FACSM, Joey C. Eisenmann. Iowa State University, Ames, IA.

Although accelerometry-based activity monitors are widely used, they have been shown to have limitations for assessing physical activity (PA) and energy expenditure (EE) during free-living conditions. A new generation of “pattern recognition” monitors may be able to overcome some of the limitations associated with traditional accelerometry-based approaches.

Purpose: The purpose of this study was to evaluate the agreement of two pattern-recognition monitors (the Intelligent Device for Estimating Energy Expenditure and Activity - IDEEA and the Sensewear Pro II armband - SP2) during free-living activity.

Methods: Twenty participants (10 males, 10 females, mean age 29.9±5.7 yrs) wore both activity monitors during a whole day, in two different trials. Participants were asked to include at least 30 minutes of moderate to vigorous PA (MVPA) during their active day and to complete an interview in order to identify possible estimation differences due to the activities performed.

Results: The two monitors yielded similar estimates of total EE (1990±557 and 1928±539 kcal/day, respectively) but estimates of MVPA were more variable (149.9±78.5 and 170.3±74.8 minutes, respectively). Bland Altmann plots showed no tendency for systematic bias between the estimates from the two monitors. Correlations between the two instruments were high for total EE (r=0.81 to r=0.89), but lower for MVPA (r=0.38 to r=0.60). Individual minute-by-minute correlations for EE ranged from 0.45 to 0.96 (mean 0.76). The large majority (70%) of the trials yielded correlation coefficients higher than 0.70, but four participants had lower correlations (ranging from 0.45 to 0.69) for both of their trials. The unique characteristics of the participants and the specific activities performed may account for the variability in agreement.

Conclusions: The results of this study show good overall agreement between the IDEEA and the SP2 for assessing total EE and PA. The higher correlations observed for estimates of EE are likely due to the fact that body mass is included in the EE estimates from both devices. Additional work with a criterion measure of EE and PA is needed to better understand sources of variability between the monitors.

Publication: Abstract presented at the Annual Meeting of the American College of Sports Medicine 2007, New Orleans, LA. USA