A physical activity questionnaire(PAQ) has been the practical and widely used approach for assessing PA in epidemiological study. The Japan Arteriosclerosis Longitudinal Study (JALS) has started to investigate relationship between several risk factors and CVD among multi-center integrated cohort populations (> 100,000 people) in Japan.
To investigate the validity of PAQ for JALS by comparering with accelerometer method.
A self-administered PAQ was developed to assess energy expenditure. This PAQ, we can get some information about behavioral situation, and intensity, frequency and duration of exercise. A sample of 42 middle-aged people (aged 57.5±7.7 yr) completed PAQ and wore the accelerometer (Lifecorder (LC), Suzuken Co, Japan) for 12 to 14 consecutive days.
Total Energy Expenditure (TEE) was computed from PAQ as the sum of the energy expended in sleep, work-related activity, commuting, household, leisure time activity and sedentary activities. Mean TEE estimated by PAQ and LC was 36.5±5.8 and 30.5±2.5 (kcal/kg/day), respectively. TEE by the two methods moderately correlated with each other: r=0.44(p = 0.004). Spearman's correlation coefficients among the EE (excluding basal metabolism) and time (hour/day) of PAQ activities, commuting, sports were 0.36(p = 0.019) and 0.48(p = 0.001), respectively. We found that the difference of EE by the PAQ and LC in this study. Previous studies reported that the LC underestimated EE compared that of DLW methods, we considered that the limitations of the LC to assess EE include its inability to measure low intensity PA. In order to obtain the precise absolute EE by the PAQ, we should calibrate EE equation by another valid methods.
The EE estimated by the PAQ correlated with that determined by the LC moderately, these data indicate that PAQ has adequate levels of validity in epidemiological study.