Background: Physical activity is an important lifestyle which is often poorly assessed in epidemiological studies. The European Prospective Investigation into Cancer Study-Norfolk cohort (EPIC-Norfolk), a large population-based cohort study, has developed a comprehensive questionnaire to assess activity in different domains of life aimed at assessing total energy expenditure. We report the repeatability of this instrument and its validity against repeated objective measures of fitness and energy expenditure undertaken throughout the time frame of reference of the questionnaire.
Methods: The validity of the instrument was measured in 173 individuals randomly selected from a continuing population-based cohort study. Energy expenditure was assessed by four separate episodes of 4-day heart-rate monitoring, a method previously validated against whole body calorimetry and doubly-labelled water. Cardio-respiratory fitness was assessed by four repeated measures of sub-maximum oxygen uptake. At the end of the 12-month period, participants completed the physical activity questionnaire that assesses past-year activity at home, work and in recreation. Repeatability was assessed in a separate group of 399 randomly selected participants in EPIC who completed the physical activity questionnaire twice with a 3-month interval.
Results: The age- and sex-adjusted correlation between the objective measure of daytime energy expenditure and the sum of recreational and occupational reported physical activity (in MET h per week) was 0.28 (P < 0.001). The reported time spent in vigorous activity was correlated with cardio-respiratory fitness (0.16, P < 0.05) and with the proportion of time when energy expenditure was more than five times basal (0.17, P < 0.05). The repeatability of the sum of recreational and occupational reported activity was high, r = 0.73.
Conclusions: The indices of physical activity derived from this questionnaire have levels of validity and repeatability comparable to other physical activity instruments that are used in large epidemiological studies and which have undergone such intense development and testing.