Physical Activity Energy Expenditure by Activity Domains in US Populations

Estimating energy expenditure (EE) for reported physical activities (PA) is important for assessing population PA status.

PURPOSE: To estimate EE of activities in different PA domains from the 2003-2004 NHNAES PA questionnaire data.

METHODS: A sub-set of adult (N = 3,083, males = 1,511 and females = 1,572) data from the "NHANES 2003-2004" study was employed. EE (in MET-Minutes/day) of occupational, transportation, domestic, leisure-time, and sedentary activities was calculated based on the frequency, duration and their MET values by overall and different demographic characteristics. SAS survey procedures were used for data analyses to account for the complex NHANES sampling design.

RESULTS: American adults spent 726.85 (±SEM: 1.28) MET-Minutes/day in occupational activity, 23.35 (±5.2) in transportation activity, 91.75 (±8.65) in domestic activity, 69.77 (±2.37) in leisure-time moderate and 61.19 (±3.64) vigorous activities, 134.14 (±2.87) watching TV, and 64.14 (±1.61) using a computer. Males reported higher EE in occupational (p<.001), domestic (p<0.01) and leisure-time PA (p<.001) than females. Non-Hispanic Whites (NHW) showed higher EE in domestic activity than non-Hispanic Blacks (NHB; p<.01) and Hispanics (p<.01), and higher EE in using a computer than Hispanics (p<.001). NHB reported higher MET-Minutes/day scores in watching TV than NHW (p<.001) and Hispanics (p<.001). Hispanics spent more energy on occupational activity than NHW and NHB (p<.01). Respondents with a higher education level showed higher EE in leisure-time activities (p<.01) and using a computer (p<.001) than those with a lower education level, while the latter reported higher EE in watching TV (p<.001). Higher income respondents reported higher EE in domestic (p<.01) and moderate activities (p<.01) and lower EE in transportation activity (p<.05) and watching TV (p<.001) than lower income respondents. Older adults reported higher EE in domestic activity (p<.05) and watching TV (p<.001) while younger adults spent more energy in vigorous activity (p<.001) and using a computer (p<.001).

CONCLUSIONS: EE of reported PA across different PA domains varies among selected subpopulations. Such information is helpful in allowing a greater understanding of a subpopulation's PA patterns.