Energy expenditure determined by self-reported physical activity is related to body fatness

Objective: Activity self-reports are a commonly used tool in assessing dail y physical activity (PA) and associated energy expenditure (EE). This study examined the effect of relative body fatness (%BF) on differences between self-reported and measured duration and associated EE in healthy adults. Research Methods and Procedures: Men and women (n = 115, age 38+/-9 years), ranging in %BF from 7.9% to 58.9%, spent two separate days (normal and exe rcise) in a whole-room indirect calorimeter where EE was measured. While in the room calorimeter, subjects reported the type, intensity, and duration of each performed PA. The Compendium of Physical Activity was used to calcu late the energy cost of each reported activity. The EE of all self-reported activities (EEr) was categorized into four intensity levels, synchronized, and compared with EE from the room calorimeter (EEm). Results: With increasing %BF, subjects significantly overestimated duration of more strenuous activities (greater than or equal to 4.5), while underes timating moderate activities (2.5 to 4.4 metabolic equivalents (METs)). Mis reporting of duration and/or intensity caused an overestimation or underest imation of PA-associated EE at these levels. Reported EE sleep was lower th an measured EE sleep, although both had similar durations. As a result, tot al EEr was similar to EEm. Discussion: Individual variability of daily total PA and associated EE gene rated from self-reports in adults is high. Persons with a higher %BF report duration and/or intensity of moderate to high levels of PA with lower accu racy than leaner individuals. We conclude using the Compendium of Physical Activity is not suitable for the accurate estimation of self-reported EE of AA in adults with a higher %BF.