The reproducibility of reported height and body weight in repeated questionnaire surveys

Reproducibility of reported height and weight were studied via repeated questionnaire surveys distributed to subjects within a year. Effects of the factors which might influence the reproducibility of these values were investigated. Although only about half of the subjects reported the same height and 30% the same weight on the second questionnaire, no change was found in average height and weight after a one year interval for both male and female subjects. The reproducibility decreased with age for both height and weight. Smokers tended to have lower reproducibility for height than non-smokers. The shorter subjects had lower height reproducibility than their taller counterparts, while heavier subjects had lower weight reproducibility than light subjects. It was observed that the lower the weight reproducibility, the lower the height reproducibility and vice versa. Reported height or weight figures ending in zero or five were more frequent than the expected 10% with low reproducibility.