[A physical performance battery assessing low/high extremity functional fitness in older Japanese women]

The aim of this study was to establish a physical performance test battery to assess the wide variation of functional fitness in older Japanese women. The criteria for sampling low/high extremity functional fitness were the amount of physical activities that older women perform on a regular basis and the distance they could move by foot. Seventeen tests related to the activities parallel to daily living (APDL) were completed for 178 women, aged 60 to 91 years. Principal component analysis of the 17 performance tests in the reference group (n = 140) yielded 5 components accounting for 63.5% of the total variance. Fourteen test items were heavily loaded on the 1st principal component, so that 31.4% of the total variance was accounted for by this component. Considering these results as well as test-retest reliability, kurtosis, and skewness of each item, the following four items were selected as a combination of test battery: (X1) repetition of the bicipital flexion/extension, (X2) walking around two cones and sitting on a chair, (X3) moving beans with chopsticks, and (X4) functional reach. Principal component analysis was again applied to these four variables so as to obtain the first principal component score of each person. As a result, the following equation was drawn: the first principal component score = 0.063X1 - 0.055X2 + 0.098X3 + 0.042X4 - 2.65. The scores averaged 0.68 +/- 0.27 for the exercise group (n = 19) (those who played croquet or other similar sports activities consistently twice or more a week); and -0.76 +/- 0.55 for the sedentary group (n = 19) (those who did no exercise and confined their activities to the home). Biserial correlation coefficient as an index of criterion-related validity was 0.67 (P < 0.05) between the scores in the exercise and sedentary groups. These results indicate that a wide range of functional fitness among older Japanese women can be assessed by the four tests.