There are indications that the treatment costs arising from obesity are significant. However, the cost-of-illness studies undertaken to date also highlight two other important points. First, the cost of treatment is sensitive to the body mass index (BMI) cut-off used. Given that there is no definitive definition of obesity as based on BMI, a range of costs reflecting differing BMI cut-offs may be more appropriate than the use of a single figure. Second, the costs are, not surprisingly, also sensitive to the defined associated diseases. Again there is little agreement on these and also little information on the relative risks of attributable diseases arising from obesity. The calculations of the cost-of-illness arising from the treatment of obesity, and its associated conditions must, therefore, remain indicative rather than authoritative.