ABSTRACT
Chinese Medicine Questionnaire (CCMQ-J) consists of sixty independent questionnaires and 9 physical constitutions called subscales. One type is balanced constitution (i.e., gentleness), and the following eight types represent unbalanced constitution: Qi-deficiency constitution, Yang-deficiency constitution, Yin-deficiency constitution, Phlegm-dampness constitution, Damp-heat constitution, Stagnant Blood constitution, Stagnant Qi constitution, and Inherited Special constitution. In this study, we proposed to determine optimal number of groups in 851 participants recruited from crowdsourcing answered CCMQ-J questionnaire consisting of 60 questions. In the present study, we applied k-means clustering with gap statistics to the questionnaire data and the number of optimal groups was estimated by five. The five groups are mainly characterized by 3 subscales in CCMQ-J, i.e. (i) two subscales corresponding to Yang-deficiency and Qi-depress, (ii) three subscales corresponding to gentleness, Yang-deficiency and Qi-depress (iii) Yang-deficiency, (iv) gentleness, and (v) Qi-depress. In the crowdsourcing survey, two subscales, Yang-deficient and Qi-depress are the most frequently occurred in current Japanese people.
ABSTRACT
Japanese version of the Constitution in Chinese Medicine Questionnaire (CCMQ-J) in Chinese consists of 60 items for 9 sub-scales called Gentleness, Qi-deficiency, Yang-deficiency, Yin-deficiency, Phlegm-wetness, Wet-heat, Blood-stasis, Qi-depression, and Special diathesis. Each question is answered by choosing one form 5 grades of a Likert scale where the grades from 1 to 5 corresponding to Never, Rarely, Sometimes, Often and Always, respectively. In the present study, we examined the relationship of scores based on 597 respondents to 60 items. We tentatively classified 60 items into 12 groups by utilizing the Ward’s hierarchical clustering method and discussed similarity of items and 9 body conditions. Body Mass Index (BMI) and age can be explained by the score of 60 items by partial least square model. Significantly high correlation coefficients between real and estimated values were obtained for BMI (0.81 for male and 0.82 for female) and age (0.82 for male and 0.83 for female). Those results indicate that the answers to the 60 items can reflect aging and BMI properties and CCMQ-J can be used to assess the situation of health for evaluating the actual aging conditions in human.