Mistakes were made: misperception as a barrier to reducing overweight.

OBJECTIVE: To examine the agreement between individuals' weight status as measured by their body mass index (BMI) and their perceptions of their weight status in the US population.

DESIGN: Data from the third National Health and Nutrition Examination Survey, 1988-94 (NHANES III), were used to cross-tabulate actual weight status and self-assessed weight status, identifying population proportions that underassessed or overassessed weight status. The study accounts for gender and socioeconomic and demographic variables to identify subpopulations in which relatively large numbers of individuals misperceive their weight status.

MEASUREMENTS: Survey data included clinically measured height and weight. BMI was categorized (overweight, healthy weight, or underweight) following conventional cutpoints. In addition, each surveyed individual was asked to assess their own (categorical) weight status.

SUBJECTS: NHANES III is representative of the US population. The sample included 7758 males and 8451 females aged 20 y or above after excluding women who were pregnant or breastfeeding.

RESULTS: We show that large segments of the US population misperceive their weight status. The mix of misperceptions differs by gender, with men who are obese or overweight more likely than obese or overweight women to underassess their weight status. Women who are healthy weight/underweight are more likely than men to believe they are overweight. In addition to the gender differences, underassessors are more frequently found among those aged 65 y and over, individuals with relatively low education levels, lower income levels, and among non-Hispanic black subjects. Overassessors are more frequently found among women less than 65 y old (between the ages of 35 and 64 y for men), individuals with higher education levels, higher income levels, and among non-Hispanic white subjects.

CONCLUSIONS: Information programs linking overweight and obesity with health risks might fail to induce diet and lifestyle changes if individuals fail to recognize they are overweight or obese. While there are large number of individuals who fail to recognize their overweight or obese status, there are indicator variables that can help identify this subpopulation. Thus, it is possible to first target a message that would attempt to correct misperceptions. If successful, the size of the population susceptible to a weight-health risk information program could increase.