In developed and developing countries, morbidity and mortality rates are increasing in individuals classified as being obese [1-6]. The higher morbidity and mortality rates of obese people are due to the increased incidence of obesity-related (lifestyle-related) diseases. It is now recognized that different criteria of obesity by BMI are necessary in different ethnic groups and populations [7,8]. In addition to the degree of obesity, other factors which increase the morbidity rate of obese people include abnormal fat distribution such as upper body obesity and visceral obesity [9,10]. Recently, the definition of 'pathological obesity' has been made in Japan [11]. Definition of Obesity and Previous Criteria of ObesityObesity is defined as excessive fat accumulation but not over-weightedness. The average human body usually consists of 82% lean body mass, which is essential for sustaining daily life and physical activities, and 18% body fat which, in essence, is energy store for emergency situations [12]. Thus, obesity can be defined as 'overstorage of body fat beyond 18%'. Usually, body fat above 25% in men and 30% in women is considered to be obese. According to this definition, obesity should be determined by measuring body fat. Although there are presently many methods for measuring body fat, no methods can be conducted easily, accurately and inexpensively.

The aim of this study was to describe the 20-year changes in body mass index (BMI; kgm(-2)) and the prevalence of overweight in Japanese adults. Cross-sectional annual nationwide surveys (National Nutrition Survey, Japan) were carried out with a large probability sample of the Japanese population. Data sets of the 1976-95 surveys, comprising 91983 men and 120822 women (> or =20 years of age), were used. The analyses were carried out in age and gender groups, and by residential area according to the size of the municipality (metropolitan areas, cities and small towns). The mean BMI increaseed in men with an increment of +0.44 kg m(-2) 10 years(-1) and slightly decreased in women, by -0.09 kg m(-2) 10 years(-1), after adjustment for age. A decreasing trend of mean BMI was most significant in the female 20-29 years age-group (-0.38 kg m(-2) 10 years(-1)), in contrast to an increasing trend observed in elderly women (60-69 and 70+ years age-groups). The prevalence of preobese (BMI: 25-29.9kg m(-2)) and obese (BMI > or = 30 kg m(-2)) men increased from 14.5% and 0.8%, respectively, in the time-period 1976-80 to 20.5% and 2.01% during 1991-95. The increasing trend was most evident in the youngest age-group (20-29 years) and in those from small towns. The overall prevalence of preobese and obese women did not change during the 20-year study period. The prevalence in younger women decreased, this trend being more prominent in metropolitan areas. Although the prevalence of overweight (BMI > or =25 kg m(-2)) in Japanese men and elderly women has increased in the last 20 years, the mean BMI in younger women, especially those in metropolitan areas, has decreased. A population approach to control obesity and prevent obesity-associated diseases should be mainly focused on men and women older than 40 years of age.

Usual sleep duration is closely associated with serum lipid and lipoprotein levels.

OBJECTIVE: To determine the prevalence of obesity and reference percentile curves of body mass index (BMI; kg/m 2 ) in Japanese adults. DESIGN: Cross-sectional nationwide surveys (The National Nutrition Survey, Japan) carried out in 1990 ± 1994. A sample of 23 556 males and 28 751 females aged 15 ± 84 y was used for this analysis. RESULTS: The prevalence of grades 1 (BMI: 25.0 ± 29.9), 2 (30.0 ± 39.9) and 3 ( ! 40.0) overweight, was estimated to be 24.3%, 1.84% and 0.019% in males, respectively, and 20.2%, 2.87% and 0.032% in females, respectively, subjects aged 35 ± 64 y after standardization to the age-distribution of the standard world population (WHO, 1993). The agestandardized prevalence of obesity (for BMI ! 26.4, a criterion of the Japan Society for the Study of Obesity; and ! 30.0 was 13.6% and 1.86% in males, respectively, and 13.5% and 2.90% in females, respectively, subjects aged 35 ± 64 y. CONCLUSIONS: The standardized prevalence of obesity (BMI ! 30.0) in Japanese adults was quite low compared with the data in western populations.

MATSUSHITA, YUMI, NOBUO YOSHIIKE, FUMI KANEDA, KATSUSHI YOSHITA, AND HIDEMI TAKIMOTO. Trends in childhood obesity in Japan over the last 25 years from the National Nutrition Survey. Obes Res. 2004;12:205-214. Objective: To describe the 25-year changes in BMI (measured in kilograms per meters squared) and the prevalence of obesity in Japanese children with special reference to urban-rural differences. Research Methods and Procedures:We used the data sets from the cross-sectional annual nationwide surveys (National Nutrition Survey, Japan) conducted from 1976 to 2000 and comprising 29,052 boys and 27,552 girls between 6 and 14 years of age. We carried out the trend analyses with the data on sex and age groups and on residential areas according to the size of the municipality (metropolitan areas, cities, and small towns). Results: The mean (age-adjusted) BMI increased by ϩ0.32 kg/m 2 per 10 years in boys and by ϩ0.24 kg/m 2 per 10 years in girls, increases that were remarkable in small towns. The prevalence of obese boys and girls increased from 6.1% and 7.1%, respectively, in the time-period 1976 to 1980, to 11.1% and 10.2% in 1996 to 2000. The increasing trend was most evident in 9-to 11-year-old children of both sexes living in small towns, whereas no changes were observed in girls in metropolitan areas. Discussion: Our data clearly show increasing trends in obesity prevalence in Japanese school children. Degrees of the increasing trends, however, differed across sex and age groups and residential areas, demonstrating a particular phenomenon that girls in metropolitan areas were unlikely to become obese. These epidemiological aspects indicate the priorities for intervention in population strategies to control obesity in children.