The definitions or measurement strategies obesity and overweight have changed over time, from one study to another study, and from one country to another. The varied definitions affect statistics related prevalence of overweight and make it very difficult to take for comparison data from various studies. Prevalence refers to the total number of existing persons of an illness or problem in the study population at a given time. Some correlation to prevalence rates of obesity and overweight are expressed as very crude or unadjusted estimates, while others are age-adjusted estimates of the problem. Unadjusted prevalence estimates are used to express a meaningful cross-sectional data for the study population groups at a given point or time period being studied. For age-adjusted rates, statistical procedures are utilized to eliminate the effect caused from age differences in populations that are compared over different time periods. Unadjusted estimates and age-adjusted estimates will provide slightly different set of values.

Prior studies in U.S. have utilized the desirable weight-for-height tables developed by Metropolitan Life Insurance as the reference for overweight. More recently, a number of Governmental agencies and organizations and scientific health groups have estimated overweight and obesity using data from a series of cross-sectional surveys, which are known as the National Health Examination Surveys and the National Health and Nutrition Examination Surveys.

A number of reports reported in the past utilize a statistically obtained definition of overweight from National Health and Nutrition Examination Surveys (NHANES) II (1976-1980). This definition, which is based 85th percentile of gender-specific values for body mass index (BMI) for persons belonging to the 20 to 29 year old population, is a BMI equal to or greater than 27.3 for women and 27.8 for men. NHANES II further defines "severe overweight" based on 95th percentile values as body mass index (BMI) of more than 31.1 for men and BMI greater than 32.2 for women. Some studies rounded off BMI values to the nearest whole number, which affected the calculated prevalence. In 1995, the World Health Organization suggested a stratification for three levels of overweight using BMI cutoff points of 25, 30, and 40 respectively. The International Obesity Task Force suggested an additional cutoff point of 35 and used somewhat different terminology.

The expert panel convened by NHLBI and NIDDK released a report in 1998, which provided definitions for obesity and overweight similar to those suggested by the World Health Organization. The panel identified overweight as a body mass index (BMI) greater than 25 upto a value of less than 30, at which point it would become obesity (BMI more 30 or more). These definitions, widely used by the Federal government agencies and more and more by the general medical and scientific communities, are also based on evidence that health risks increase more steeply in individuals with a body mass index (BMI) more than twenty five. BMI cutoff points are nothing more than a guide for definitions of obesity and overweight and are very useful for comparative purposes across various populations and over time; however, the health risks linked to overweight and obesity should be expected to be on and generally do not correspond to rigid cutoff points. For example, an overweight individual with a BMI of 29 does not incur significantly much more health consequences that are commonly enumerated with obesity simply by crossing the BMI threshold of more than 30. However, health risks generally go up with higher and higher BMI even though the increased risk linked to a change from 29 to 30 may not be very significant.

Posted by: Emily

---