O besity has reached epidemic proportions in the United States. Currently, one third of US adults are classified as obese and another one third as overweight. Obesity is also a growing epidemic in other parts of the world. The French, who have enjoyed relatively low rates of obesity and cardiovascular disease (CVD), are also experiencing expanding waistlines. 1 Although the health hazards of obesity are widely recognized, the relationship between the middle range of body mass index (BMI) values and mortality is less clear because epidemiologic studies have found linear and curve-linear relationships between BMI and mortality. However, epidemiologic studies of BMI and mortality have been fraught with methodologic problems, including failure to control for cigarette smoking; inappropriate control of biologic effects of obesity, such as hypertension and hyperglycemia; and failure to consider weight loss attributable to subclinical diseases. 2 These biases have led to the typical J-or U-shaped relationship between BMI and mortality seen in many epidemiologic studies and to a systematic underestimate of the impact of obesity on premature mortality. For example, in a recent analysis of NHANES (National Health and Nutrition Examination Survey) data sets, Flegal et al 3 found that excess mortality attributable to obesity was much lower than reported previously, and that being overweight was associated with a lower mortality compared with normal weight. However, these findings are most likely the result of artifact attributable to the methodologic flaws mentioned above. In particular, the study failed to exclude persons from the analyses who had chronic diseases at baseline and did not adequately control for smoking, leading to artificially elevated mortality among the lean subjects. Nonetheless, these findings have caused a great deal of confusion among the general public, and commercial interest groups have attempted to exploit the study and dismiss obesity as a public health issue. 4 In the current issue of the Hypertension, Thomas et al 5 evaluated the impact of overweight on cardiovascular mortality in a large French population (139 562 men and 104 236 women). During an average of 14 years of follow-up, they documented 11 688 deaths (2949 from CVD) among men and 4188 deaths (929 from CVD) among women. This is a self-selected population who came for a health checkup, with a mean age of 43 years for men and 41 years for women and mean BMI values of 24.6 for men and 22.8 for women. The prevalence of overweight and obesity combined was 42% in men and 21% in women. The prevalence of obesity was quite low (Ϸ5% in both genders). This large and relatively young population provides a unique opportunity to examine the impact of overweight on cardiovascular mortality.Two major findings emerged from these analyses. First, in age-adjusted analyses, increasing BMI was strongly associated with increased cardiovascular morality, but the association became nonsignificant after adjusting for cardiovascular risk factors such as hypert...