Eating in the laboratory: is it representative?

The degree to which caloric intake of food consumed in a laboratory setting can approximate caloric intake measured under free-living conditions was assessed in the present study. Four men and four women of normal body weight weighed and recorded their food intake for four days (period 1) during which they were eating in their normal home environment. On the following week (period 2) which also lasted 4 days, the subjects consumed all their food in a laboratory setting. Based upon the findings that no significant change in body weight occurred during either period of the study, that no significant difference in paired caloric intake was observed between period 1 and period 2, and that a highly significant intraclass correlation coefficient, ri = .80 (p less than .005) existed between period 1 and period 2, it was concluded that food consumed in a laboratory setting is a reasonable approximation of caloric intake as measured under free-living conditions.