The obesity problem in individual US states appears much greater than official health surveys have indicated, according to a new study.

The problem, researchers found, is that the telephone surveys used to estimate states' obesity rates are prone to error namely, that female respondents underestimate their weight, on average, while men often add inches to their actual height.

National estimates of the obesity epidemic, however, are more on target, because the methodology used is different.

It's estimated that one third of US adults are obese, and that national statistic is probably a close reflection of reality, according to Dr Majid Ezzati, an associate professor of international health at the Harvard School of Public Health in Boston.

But, Ezzati said his team's findings indicate that individual states' obesity levels are "grossly underestimated."

The disparity between national- and state-level obesity estimates stems from differences in how the data are collected by the Centres for Disease Control and Prevention. Nationwide data are gathered through the National Health and Nutrition Examination Survey (NHANES), where respondents are interviewed in person about their height and weight, and then asked to have a physical exam.

State statistics are collected via the Behavioral Risk Factor Surveillance System (BFRSS), which asks respondents to give their height and weight over the phone.

By comparing information from this survey with data from the NHANES project, Ezzati and his colleagues found that every US state's obesity burden was underestimated.

The study found that young and middle-aged women generally shaved pounds off of their weight when asked over the phone, while younger adult particularly men tended to overestimate their height.

Source: China Daily