Measurement and prediction of METs during household activities in 35- to 45-year-old females.

This study determined whether four self-paced household tasks, conducted in the subjects' homes and a standardised laboratory environment, were performed at a moderate intensity [3-6 metabolic equivalents (METs)] in a representative sample of thirty-six 35- to 45-year-old females. Energy expenditure was also predicted via indirect methods. Self-paced energy expenditure during sweeping, window cleaning, vacuuming and mowing was measured using the Douglas bag technique. Heart rate, respiratory frequency, Computer Science Applications (CSA) movement counts (hip and wrist), Borg rating of perceived exertion and Quetelet's index were also recorded as potential predictors of energy expenditure. While the four activities were performed at mean intensities >or=3.0 METs in both the home and laboratory, all comparisons between these two environments were statistically significant ( P<0.001). The 95% confidence intervals (CIs) for the home and laboratory prediction equations were +/-1.1 METs and +/-1.0 MET, respectively. These data suggest that the aforementioned household chores can contribute to the 30 min x day(-1) of moderate-intensity activity required to confer health benefits. However, the substantial between-subject variability in energy expenditure resulted in some persons performing these tasks at a light intensity (<3.0 METs). The significant MET differences between the home and laboratory emphasise the effects of 'environment and terrain' and the 'mental approach to a task' on self-paced energy expenditure. Considering the means for the five activities ranged from 3.1 METs to 6.0 METs, the 95% CIs for the regression equations lack predictive precision.