Boston, MA – The prevalence of obesity in the U.S. states has been greatly underestimated. Researchers at the Harvard School of Public Health (HSPH) analyzed data from health surveys, which are used to estimate obesity levels in states. Because people tend to provide incorrect information about their weight and height, especially in telephone surveys, the researchers concluded that estimates of obesity in individual states have been too low, by more than 50%. Their study, which corrects for misreporting in those surveys, appears in the May 2006 issue of the Journal of the Royal Society of Medicine.

Obesity is one of the leading causes of morbidity and mortality, causing some 2.6 million deaths worldwide each year. In the U.S., survey data on obesity on a national and state level is obtained using information gathered by the Behavioral Risk Factor Surveillance System (BRFSS), which uses telephone interviews; national data is also collected using the National Health and Nutrition Examination Survey (NHANES), which does in-person interviews and follow-up height and weight measurements on people who agree to a clinical exam. Lead author Majid Ezzati , Associate Professor of International Health at HSPH, and his colleagues analyzed and compared the data from the two surveys in order to quantify the level of bias when people self-report their height and weight, especially in a telephone interview.

Based on this new understanding of the survey data, the authors found that, on average, women tend to underestimate their weight while men do not. When it comes to height, young and middle-aged men tend to overestimate their height more than women in the same age group. In 2002, the corrected prevalence of obesity in the U.S. population was 28.7% for adult men and 34.5% for adult women, more than 50% higher than previously estimated.

The research, which presents the first-ever corrected estimates of obesity for individual states, found that Southern states have the highest levels of obesity in the country. Using the authors’ corrected data for 2000, the highest obesity levels for men were found in Texas (31%) and Mississippi (30%). For women, Texas (37%), Louisiana (37%), Mississippi (37%), District of Columbia (37%), Alabama (37%) and South Carolina (36%) led the pack. States with the lowest prevalence of corrected obesity for men in 2000 were Colorado (18%), District of Columbia (21%) and Montana (21%); for women Colorado (24%), Montana (25%) and Massachusetts (27%).

“The results show that there are clear geographical patterns of obesity,” said Ezzati. “Using this data, national and state public health institutions can better target resources and programs toward combating the growing epidemic.”

To view a copy of the study, go to http://www.rsm.ac.uk/media/pdf/j06-05obesity.pdf

This work was supported by a cooperative agreement from the Centers for Disease Control and Prevention through the Association of Schools of Public Health and by the National Institute on Aging.