Reliability and validity of a questionnaire for assessment of energy expenditure and physical activity in epidemiological studies

Abstract

A self-administered physical activity questionnaire (PA-questionnaire) was developed to assess daily energy expenditure and weekly physical activity in epidemiological studies. The Calorie Counter method (CC-method) was administered to 49 male and 32 female volunteers aged 18-64 years, on 7 consecutive days; after the measurement, the subjects were asked to complete the PA-questionnaire (validation study). The PA-questionnaire was completed by 95 males and 119 females (aged 35-73 years) twice with one-year interval (reliability study). The validation study showed that the mean daily energy expenditure estimated by the PA-questionnaire was slightly and significantly (4.5%) lower than that determined by the CC-method for males (p < 0.05); while no significant difference was observed for females (p > 0.05), mean weekly physical activities were similar between the PA-questionnaire and CC-method in males and females. Daily energy expenditures by the two methods strongly correlated with each other: r = 0.56 (p < 0.001) in males and r = 0.67 (p < 0.001) in females. Weekly physical activities by the two methods also strongly correlated with each other: r = 0.68 (p < 0.001) in males and r = 0.69 (p < 0.001) in females. The reliability study indicated that the Pearson's correlation coefficients between two assessments of daily energy expenditure and weekly physical activity over one-year ranged from 0.37 to 0.62 (p < 0.001). These data indicate that the PA-questionnaire has adequate levels of validity and test-retest reliability in assessment of daily energy expenditure and weekly physical activity in epidemiological studies.