Background: Although the metabolic syndrome (MetS) is considered to be caused primarily by visceral fat accumulation, epidemiological evidence is lacking as to whether or not obesity is an essential element in the syndrome. Methods and Results: Between 1990 and 2005, the Japan Public Health Center-based Prospective (JPHC) Study conducted baseline measurements of metabolic risk factors in 12,412 men and 21,639 women, aged 40-69 years, with no history of cardiovascular disease (CVD) or cancer. To clarify the role of obesity, which the definition of MetS in Japan has adopted as an essential criterion, clustering of risk factors in data grouped according to overweight condition was examined. During a 12.3-year follow-up there were 2,040 deaths, including 947 from cancers and 304 from CVD. MetS significantly increased the hazard ratios for all-cause mortality in women and CVD mortality in men. Non-overweight with ≥2 risk factors had a similar impact on all-cause and CVD mortality. Clustering of metabolic factors caused a linear increase in the hazard ratios for mortality. Conclusions: MetS caused moderate increases in all-cause and CVD mortality. However, the MetS definition requiring obesity may not necessarily identify non-overweight individuals who have a high mortality risk and are more prevalent than subjects with MetS. (Circ J 2009; 73: 878 - 884)