Development and validation of methods to determine energy expenditure in humans

Obesity has increased dramatically in the last 40 years in America and most industrialized nations. The primary cause of obesity is an imbalance between energy intake and energy expenditure. Assessing quantitatively the energy expenditure (EE) on a daily basis is of particular importance in addressing this rising health problem, particularly physical activity related to energy expenditure EE_act.; In the first part of this study,{09}(n = 60) healthy normal female volunteers were used to improve the prediction of a nonlinear model previously developed using a portable triaxial accelerometer monitor (Tritrac-R3D). To improve the existing model, a wrist-worn accelerometer (ActiWatch) was worn to measure upper limb movements to compensate for periods underestimated by the Tritrac-R3D. In a subgroup of 12 volunteers, a second 24-hour measurement period was used to validate the model. The model was able to accurately predict activity related EE and total EE compared to the measured EE whole-room indirect calorimeter (r = 0.92).; In the second part of this study, we analyzed the ability of a newly designed insole measurement system to measure the reproducibility of ground reaction forces (GRF) during walking at different speeds and grade. Participants (n = 25) conducted a protocol consisting of five successive 10-minute walking segments at different speeds and degree of incline on a treadmill. Walking speed did not affect the reproducibility of the measured GRF's. Measured forces were accurate when walking at different speeds with or without additional weight carried in a backpack.{09}We concluded that our newly designed insole system is a reliable tool for measuring ground forces during walking at different speed and grade, independent of subject's weight.; In the third part of this study, (n = 37) healthy adult females were used to establish a model that fit ground reaction forces (GRF) and step-frequency to predict EE_walking. GRF and EE were measured simultaneously during treadmill walking at different speeds and degrees of incline while in a whole-room indirect calorimeter. Two variables, toe push-off force and step frequency, were used in a four-parameter power model to estimate EE_walking on a minute-by-minute basis. Individual models were developed and the estimated EE_walking were correlated with the measured EE_walking (r = 0.96) and standard error of estimate (SEE = 0.45 kcal/min). The forces measured using the insole device are useful in predicting EE_walking in energy balance studies.; Quantitative assessment of physical activity and energy expenditure was conducted in these studies. It was concluded that existing physical activity monitoring devices prediction of energy expenditure proved to be useful in limited situations (r = 0.94). To address these limitations a newly developed physical activity monitoring device was developed. An insole force measurement was developed and employed in monitoring physical activity. The force measurement system improved prediction of energy expenditure versus the existing techniques.