Search tips
Search criteria 


Logo of jrsocmedLink to Publisher's site
J R Soc Med. 2006 May; 99(5): 215–217.
PMCID: PMC1457753

Self-image in obesity: clinical and public health implications

Andrew J Krentz, Consultant in Diabetes & Endocrinology

`I know that I am more than my personality, my body, and my body image'. Oprah Winfrey

No-one can be in any doubt that a global epidemic of obesity in adults1 and children2 is well and truly upon us. According to current estimates more than 1 billion adults are overweight, with over 3 million who are obese.1 These figures are based on body mass index (BMI), a relatively imprecise measure of body fat, calculated as weight divided by the square of height, measured in population samples. The World Health Organization [] classifies overweight as a BMI > 25 kg/m2 with obesity being a BMI > 30 kg/m2. This classification reflects a dose-effect relationship between increasing adiposity and adverse clinical outcomes.

The threats to health associated with obesity stem principally from the development of adverse metabolic profiles and an excess of certain cancers. In fact, the risks of metabolic complications including diabetes, dyslipidaemia and hypertension start to increase at lower levels of BMI. This risk varies with ethnicity and, for this reason a lower level of BMI, 23 kg/m2 is now regarded as more appropriate threshold for South and East Asian populations. Having said this, there is a growing view that waist circumference, a surrogate of visceral adiposity, may be a more sensitive indicator of some forms of risk.1 For Asians, levels of waist girth denoting an increased risk of diabetes and cardiovascular disease are lower than for White Europeans; this view is incorporated into the new International Diabetes Federation [] definition of the metabolic syndrome. Inconsistencies between different studies and variations in methodology provide continuing uncertainty about the best means for quantifying health risks associated with overweight and obesity. For example, in a recent large international case-control study, waist-to-hip ratio was a better predictor than BMI of myocardial infarction.3

Assessments of the societal burden of disease attributable to obesity relying on data from population samples carry a potential for error. This may be compartmentalized into random error and, in studies based on self-reported information, systematic reporting bias. It has long been recognized that people tend to think that they are taller than they really are (men especially) and somewhat slimmer than the bathroom scales would tell them. This, of course, is no surprise and world-weary clinicians tend to take a somewhat sceptical view of self-reported smoking habits, alcohol consumption, and in the obese, daily food intake. Moreover, it is predictable that, in general, self-reports tend to be less accurate when the chance of being challenged with compelling evidence to the contrary is low. On an individual level, the reliability of self-reported weight decreases with the magnitude of obesity.4 Race, age and digit-preference are ancillary predictors of error.4 Among American women, those who are unemployed, retired or disabled are more likely to under-report their BMI.5 Data from Australia6 and Germany7 suggest that these issues are not confined to North America. For instance, among German adults, estimates of obesity by other members of a household are more accurate than self-reported figures.7 Obesity in younger people is a particular concern, since most will carry their adiposity into adulthood.2 The accuracy of BMI estimated by parents is better than those derived from self-reports by teenagers.8 In a community-based study from Manitoba, overweight and obese secondary school children showed greater bias to under-estimate their adiposity and more variability in self-reported BMI;9 as for adults, personal body image dissatisfaction predicted this bias. The health implications of obesity in childhood are not only physical; an association between depressive symptoms and BMI has been demonstrated even among pre-adolescent girls.10

Weight-stigmatization is a common experience for obese subjects seeking weight loss treatment; this is associated with poor psychological adjustment that may hinder successful weight-reduction. Long-term weight problems have an adverse impact on self-esteem. Could it be that healthcare professionals, perhaps unwittingly, contribute to such negative attitudes? A provocative study of health professionals, clinicians and researchers attending an obesity conference, using a self-reported questionnaire provides proverbial food for thought: among these professionals, self-selected, of course, for their interest in helping people with obesity-related problems, a significant `anti-fat, pro-thin' bias was evident; implicit stereotypes of `lazy, stupid' and even `worthless',11 were endorsed. Since a non-judgmental approach is regarded as crucial in the management of obesity1 this suggests a need for some self-reflection.

Do the preceding considerations have implications for public health statistics? The short answer seems to be affirmative. In the current issue of the JRSM, Ezzati and colleagues from the Harvard School of Public Health provide new data on recent obesity trends in the USA.12 The authors present what they claim are the first unbiased estimates of national and state levels and trends in obesity. National population-level comparison of self-reported and measured weight and height were compared to the gold standard of a measured health examination survey. On average, women underreported their weight, whereas men did not—with younger and middle-aged adult men over-reporting their height more than women of the same age. At older ages, over-reporting of height was similar in men and women. Population-level bias in self-reported weight was larger in telephone interviews without a subsequent measurement than in in-person interviews. Except in older adults, height was over-reported with larger bias in telephone interviews than in person-to-person interviews with a follow-up examination. This study differs in an important way from previous studies that have addressed the issue of bias, which have relied on individual-level, as distinct from population-level, data. Earlier studies were unable to examine two factors important for population-level monitoring: first, individual-level misreporting caused by absence of measurement of weight and height either before or after interview and by the mode of self-report; second, differential levels of participation based on the survey mode. Ezzati et al. were able to apply a correction to existing estimates of obesity in various states in the USA. The unwelcome message is that previous reports on state-level obesity levels and trends in the USA—based on telephone surveys—have led to significant underestimations of the true size of the problem. Using corrected weight and height values applied to individual states, in the year 2000, Mississippi and Texas had the highest prevalence of obesity for men (31% and 30%, respectively); both states also being in the top six for women (37% for both). The authors suggest that their estimates are likely to be the best currently available option for population-level monitoring. These figures should sound yet more public health alarm bells in the USA.

While the Health Survey for England [] is based on measured height and weight, trends in the UK are no less worrying. The proportion of adults categorized as obese increased from 13.2% of men in 1993 to 23.6% in 2004, and from 16.4% of women in 1993 to 23.8% in 2004 []. These figures do not tell the whole story of what has been described as the developed world's fastest growing rate of obesity. This unenviable position is accompanied by a coronary heart disease mortality rate that is one of the highest in Western Europe []. We can point to socio-economic and ethnic differences that might protect us to some degree from following in the USA's footsteps. However, current evidence cautions against sitting back and viewing the revised transatlantic figures with complacency.


1. Haslam DW, James WP. Obesity. Lancet 2005;366: 1197-209 [PubMed]
2. Laron Z. Increasing incidence of childhood obesity. Pediatr Endocrinol Rev 2004;1(Suppl 3): 443-7 [PubMed]
3. Yusuf S, Hawken S, Ounpuu S, et al. Obesity and the risk of myocardial infarction in 27,000 participants from 52 countries: a case-control study. Lancet 2005;366: 1640-9 [PubMed]
4. Rowland ML. Self-reported weight and height. Am J Clin Nutr 1990;52: 1125-33 [PubMed]
5. Nawaz H, Chan W, Abdulrahman M, Larson D, Katz DL. Self-reported weight and height: implications for obesity research. Am J Prev Med 2001;20: 294-8 [PubMed]
6. Flood V, Webb K, Lazarus R, Pang G. Use of self-report to monitor overweight and obesity in populations: some issues for consideration. Aust N Z J Public Health 2000;24: 96-9 [PubMed]
7. John U, Hanke M, Grothues J, Thyrian JR. Validity of overweight and obesity in a nation based on self-report versus measurement device data. Eur J Clin Nutr 2006;60: 372-7 [PubMed]
8. Goodman E, Hinden BR, Khandelwal S. Accuracy of teen and parental reports of obesity and body mass index. Pediatrics 2000;106(1 Pt 1): 52-8 [PubMed]
9. Elgar FJ, Roberts C, Tudor-Smith C, Moore L. Validity of self-reported height and weight and predictors of bias in adolescents. J Adolesc Health 2005;37(5): 371-5 [PubMed]
10. Erickson SJ, Robinson TN, Haydel KF, Killen JD. Are overweight children unhappy? Body mass index, depressive symptoms, and overweight concerns in elementary school children. Arch Pediatr Adolesc Med 2000;154: 931-5 [PubMed]
11. Schwartz MB, Chambliss HO, Brownell KD, Blair SN, Billington C. Weight bias among health professionals specializing in obesity. Obes Res 2003;11: 1033-9 [PubMed]
12. Ezzati M, Martin H, Skjold S, Vander Hoor S, Murrary CJL. Trends in national and state-level obesity in the United States. J R Soc Med 2006:99: 250-257 [PMC free article] [PubMed]

Articles from Journal of the Royal Society of Medicine are provided here courtesy of Royal Society of Medicine Press