Accuracy of the Actiheart for the assessment of energy expenditure in adults

Background/Objective: The Actiheart (Mini Mitter, Sunriver, OR, USA) uses heart rate (HR) and activity data to predict activity energy expenditure (AEE). Currently, the Actiheart has only been tested during laboratory conditions. Therefore, the objective of this study was to validate the Actiheart prediction method against indirect calorimetry during a wide range of activities in a field setting. Subjects/Methods: Forty-eight participants (age: 35 +/- 11.4 years) were recruited for the study. Eighteen activities were split into three routines of six activities and each routine was performed by 20 participants. During each routine, the participants wore an Actiheart and simultaneously, AEE was measured with a Cosmed K4b(2) portable metabolic system. The manufacturer's HR algorithm, activity algorithm, and combined activity and HR algorithm were used to estimate AEE. Results: The mean error (and 95% prediction intervals) for the combined activity and HR algorithm, HR algorithm, and activity algorithm versus the Cosmed K4b(2) were 0.02 kJ kg(-1) min(-1) (-0.17, 0.22 kJ kg(-1) min(-1)), -0.03 kJ kg(-1) min(-1) (-0.24, 0.18 kJ kg(-1) min(-1)), and 0.14 kJ kg(-1) min(-1) (-0.12, 0.40 kJ kg(-1) min(-1)), respectively. Conclusion: The Actiheart combined activity and HR algorithm and HR algorithm provide similar estimates of AEE on both a group and individual basis.