The simplified Japanese version of the Self-Perception Scale for Adolescent (SJS-PSA) was used to measure self-perception. Reliability and validity were examined. The selfconcept of the adolescents who had participated in the longitudinal study and its related variables were studied by this scale (SJSPSA). As the results, (1) The developmental status in infant and preschool years (the result of the JDDST), personal relationship in school years (loneliness), health condition, the attitude towards the future (level of education aspired to in the future) and present support from friends were correlated significantly with the self-perception of the adolescent. Also, from the multivariate analysis, a combination of five items such as task consciousness in adolescence, the results of the JDDST, level of life satisfaction in adolescence, mother's educational level and father's selfconceptcontributed to the development of the self-concept of children. (2) The self-perception score could be used in identifying adolescents at risk and for intervention.
Obesity is the most prevalent, chronic condition in the Japanese population, and is directly or indirectly associated with a wide variety of diseases . Although the term "obesity" generally refers to an excess of body fat, it is difficult to determine the definition and classification of obesity. Therefore, we must rely upon only a few arbitrary standards . In order to resolve this health need, it is necessary to establish reasonable obesity indexes, which should be free from the aesthetic standard of "looking good". For this purpose, the authors studied ultrasonic measurement of the thickness of subcutaneous adiposal tissue layer, and conducted medical examinations of serum lipid twice, in 1988 and in 1990, in a rural population living in Hase, Kami-Ina, Nagano prefecture. The points of measurement were biceps, triceps, subscapular, suprailiac and thigh anterior. Using the obtained data, the relationships between the thickness of the subcutaneous adiposal tissue layer and data of medical examinations, including blood pressure, total cholesterol, triglyceride and HDL-cholesterol, were clarified. The results obtained were as follow: 1. There were no consistent relationships found between thickness of subcutaneous fat layer and the results of each medical examination by year, sex and age in this follow-up study, although the authors reported positive correlation coefficients between them in a cross-sectional study, 1988. 2. In analyses using the multiple logistic model, there were positive, but weak, correlations between increase of fat thickness and Hata's Severity Grading of Hyperlipidemia, especially between the increase of fat thickness at both suprailiac and thigh anterior and increase of severity grade of TC, TG, and HDLC in males . Positive correlations were also found between the increase of fat thickness at both triceps and suprailiac and increase of the severity grade of TG. These results support the need to continue such follow-up studies. 3.In cross-sectional studies, such as etiological and/or health administrative studies, reproducibility and reliability in the relationship between medical examination and health-related phenomenon are difficult to obtain. For this reason, studies on etiology should be carefully conducted and followed up repeatedly.
We conducted an epidemiological study on liver cancer in the former coalmining area of Chikuho where mortality from liver cancer was high, using the data on liver cancer mortality in lizuka Health Center in Fukuoka Prefecture. Using the Japanese mortality from liver cancer as standard, we found that the ratios of observed to expected deaths from liver cancer in the Iizuka Health Center District between 1986 and 1988 were 213 for males and 198 for females. Of those deceased from liver cancer, 17.4% were HBs antigen positive. A high positive rate of HBs antigen was particularly observed in the 30-39 age group (positive rate=100%) and the 40-49 age group (positive rate=42.9%) among the younger age groups. To clarify the relationship of liver cancer to HBs antigen, blood transfusion, drinking habits and smoking habits, we also conducted a case-control study. We analyzed 133 cases who had died of liver cancer and 132 controls who were randomly selected from inpatients in the three large hospitals in the Iizuka Health Center District. The relative risks were 10.23 for HBs antigen positive (p<0.001) and 2.33 for blood transfusion (p<0.01), which were statistically significant, but 1.20 for drinking and 1.59 for smoking, which were not statistically significant. HBs antigen positive and blood transfusion were significantly associated with liver cancer.