This paper considers alternate measures of overweight in the U.S. that are sensitive to changes in the body-mass index (BMI) distribution, more robust to measurement error and continuous in the body-mass index (BMI) at the overweight threshold. The measures suggest that standard prevalence rates may be understating the severity of the problem. Since 1971, overweight prevalence has increased by 40% while the distribution-sensitive measure has increased by 174%. They also provide some useful insight into socioeconomic differences in overweight. For example, overweight prevalence rates for the poor and the rich have been very similar over the last 30 years, with the rich have a slightly higher rate in the most recent 2001- 2002 data. In contrast, the distribution sensitive measures reveal that overweight rich people exceed the overweight threshold by 23% while the overweight poor exceed the threshold by a much greater amount (31%) on average.