Background
: Physical activity is highly associated with health, especiall reduces the risk coronary heart diseases and strengthens cardiovascular system, and then reduces morbidity and mortality. One of the most reliable and valid measures of physical activity is direct observation but it is not practical due to the most costly and burdensome. There this study compares two methods, Stanford 7-day Physical activity Recall questionnaire and electronic motion sensor monitor, and evaluates usefulness of these two methods in the field of primary care.

Methods : Fifty-one young male-adults in the second and third decades were recruited, 11 from medical school students, 15 from college students and 25 from residents and interns of university hospital doctors. Each subject was required to complete the Stanford 7-day Physical Activity Recall questionnaire and wear electronic motion sensor one day a week. Two weeks later, the second test was perfomed as the same way.

Results : In questionnaire. total physical activity was classified as sleep, light, moderate, hard and very hard activities. Mean hours spent for each activity for a week were 45.02±7.43, 99.11±13.19, 19.72±9.84, 2.90±3.04, 0.94±1.78 hours and reliabilities measured by test-retest were 0.72, 0.64, 0.70, 0.52, 0.55(p<0.01). The mean of daily energy expenditure was 42.74±5.14 kcal/kg/day and the reliability was 0.65(p<0.01).
Means of daily walking steps, distances and energy consumption for walking measured by electronic motion sensor were 13,195±5,036 steps, 6.58±2.55 km, 578.5±257.2 kcal and each correlation coefficient between the first and second tests was 0.65, 0.65, 0.75(p<0.01).
In comparison the energy expenditures resulted form Stanford questionnaire with the walking energy consumption from electronic motion sensor, the correaltion coefficients were -0.18(p=0.22) for sleep, -0.39(0=0.06) for light, 0.56(p<0.01) for moderate, -0.01(p=96) for hard, -0.02(p=0.91) for very hard activities in the first test. The moderate activity showed the highest correlation in both first and second tests(p<0.01).

Conclusion : The results of the Standard 7-day Physical Activity Recall questionnaire showed high reliabilities and relatively high correlation coefficieny with electronic motion sensor, especially in moderate activity, therefore it seems that the concurrent use of both methods for physical activity measure is more valuable than single use in the field of primary care and epidemiologic study.