Conflicting results have been reported concerning the association between body weight and longevity. The shape of the curve relating weight to all-cause mortality has been variously described as linear, J-shaped, and even U-shaped. To assess the validity of the evidence for optimal weight recommendations, we examined the 25 major prospective studies on the subject. Each study had at least one of three major biases: (1) failure to control for cigarette smoking, (2) inappropriate control of biologic effects of obesity, such as hypertension and hyperglycemia, and (3) failure to control for weight loss due to subclinical disease. The presence of these biases leads to a systematic underestimate of the impact of obesity on premature mortality. Although these biases preclude a valid assessment of optimal weight from existing data, available evidence suggests that minimum mortality occurs at relative weights at least 10% below the US average.