Pediatricians Own Weight: Selfperception ... - Wiley Online Library

Pediatricians’ Own Weight: Self-perception, Misclassification, and Ease of Counseling Eliana M. Perrin,*† Kori B. Flower,*† and Alice S. Ammerman‡

Abstract PERRIN, ELIANA M., KORI B. FLOWER, AND ALICE S. AMMERMAN. Pediatricians’ own weight: selfperception, misclassification, and ease of counseling. Obes Res. 2005;13:326 –332. Objectives: Pediatricians underdiagnose overweight and feel ineffective at counseling. Given the relationship between physicians’ health and health habits and counseling behaviors, we sought to determine the 1) percentage of pediatricians who are overweight; 2) accuracy of pediatricians’ own weight status classification; and 3) relationship between weight self-perception and perceived ease of obesity counseling. Research Methods and Procedures: This study was a cross-sectional, mail survey of North Carolina pediatricians that queried about their weight status and ease of counseling. Accuracy of pediatricians’ self-classification of weight status was compared with BMIs derived from self-reported height and weight. Using logistic regression, controlling for potential confounding variables, we examined the association between weight perception and ease of counseling. Results: The unadjusted response rate was 62%, and the adjusted response rate was 71% (n ⫽ 355). Nearly one-half (49%) of overweight pediatricians did not identify themselves as such. Men had greater adjusted odds of misclassifying overweight than women [odds ratio (OR), 3.61; 95% confidence interval (CI) ⫽ 1.81, 7.21]. Self-classified “thin” pediatricians had nearly six times the odds of reporting more counseling difficulty as a result of their weight than

Received for review April 15, 2004. Accepted in final form December 14, 2004. The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. *Division of General Pediatrics and Adolescent Medicine, Department of Pediatrics, †Robert Wood Johnson Clinical Scholars Program, and ‡Department of Nutrition and Center for Health Promotion and Disease Prevention, University of North Carolina at Chapel Hill Schools of Medicine and Public Health, Chapel Hill, North Carolina. Address correspondence to Eliana M. Perrin, Division of General Pediatrics and Adolescent Medicine, Department of Pediatrics, University of North Carolina at Chapel Hill, CB #7220, 130 Mason Farm Road, Fifth Floor, Chapel Hill, NC 27599-7220. E-mail: [email protected] Copyright © 2005 NAASO

326

OBESITY RESEARCH Vol. 13 No. 2 February 2005

“average” weight pediatricians (OR ⫽ 5.69; 95% CI ⫽ 2.30, 14.1), and self-identified “overweight” pediatricians reported nearly four times as great counseling difficulty as “average” weight physicians (OR ⫽ 3.84; 95% CI ⫽ 1.11, 13.3), after adjustment for self-reported BMI weight status and other potential confounders. Discussion: The roles that physician weight misclassification and self-perception potentially play in influencing rates of obesity counseling warrant further research. Key words: weight status, gender, classification, selfperception, self-report

Introduction Primary care physicians have been called on to combat the epidemic of overweight (1– 4), and pediatricians, especially called to the front lines, must make the critical first steps of screening and counseling their young patients regarding overweight (5,6). However, pediatricians often under-detect overweight (7,8) and either under-counsel (7,9) or feel ineffective doing so (10,11). Because physicians’ personal health and health habits help predict their professional practice patterns and frequency of counseling in diverse areas (10,12–18), one determinant of obesity practice patterns may be pediatricians’ own weight and feelings about weight. Although the literature on the relationship between physicians’ personal health habits and counseling largely confirms the findings of Cornuz et al. (17), that physicians who do not pay attention to their own health also are inattentive to their patients’ preventative health care, studies that have examined the impact of physicians’ own actual weight status on their counseling practices or feelings about counseling have found conflicting results (9 –11,13,15,16). We hypothesize that perceived weight status might be a greater counseling practice determinant than actual weight status. To our knowledge, the relationship between perceived weight status and ease of counseling has not been well researched. One study, which included only the responses of female physicians, found that there was a trend toward

Pediatricians’ Own Weight, Perrin, Flower, and Ammerman

physicians who reported “feeling overweight” being more likely to counsel every visit or at least once a year on issues regarding weight (15). To our knowledge, no studies have examined the impact of perceived thinness on counseling tendencies or patterns. Furthermore, whereas many studies have found that actual weight status is discordant with perceived weight status in the lay population (19 –22), no studies have examined discordance in physicians. While physicians may be more accurate in their classifications because of their professional training, any misclassification may be of interest for its potential relationship to underdetection of patient weight problems. Using self-reported heights and weights and self-reported perceived weight status, we explored the proportion of primary care pediatricians who are themselves overweight, how often they appropriately classify their own weight status, what demographic variables affect that accuracy of classification, and whether weight self-perception influences reported ease of counseling.

Research Methods and Procedures Data Source North Carolina pediatricians who were both members of the North Carolina Pediatrics Society (NCPS)1 and Fellows of the American Academy of Pediatrics (AAP) (n ⫽ 824) served as our initial study sampling frame. However, pediatricians were excluded (n ⫽ 86) if review of their addresses or state licensure files indicated a subspecialty affiliation or if data revealed that they were retired or deceased. In June 2002, the eligible pediatricians (n ⫽ 738) were mailed a 54-item previously described (23) survey on counseling about overweight for self-administration. A reminder postcard was sent 1 month after the initial mailing, and a second questionnaire was sent to non-respondents 6 weeks after the initial mailing. To include only pediatricians whose responsibilities included monitoring weight, they were asked one screening item: “Do you provide routine health care for children?” If not, they were asked to return their survey with the rest of the items blank. The Institutional Review Board at the University of North Carolina at Chapel Hill approved the study. Measures As part of the survey, respondents were asked which sentence described them most accurately: “I am thin,” “I am average weight,” or “I am overweight” and whether their weight made it “more difficult,” “easier,” or “no difference” for them to counsel about issues related to obesity. Later, in the survey’s demographic query, pediatricians self-reported

1 Nonstandard abbreviations: NCPS, North Carolina Pediatrics Society; AAP, American Academy of Pediatrics; OR, odds ratio; CI, confidence interval.

their height (inches) and weight (pounds), as well as their gender, age, ethnicity, practice type (academic vs. nonacademic), practice community type (urban, suburban, or rural), and percentage of low-income patients served. Data Analysis We calculated our survey’s unadjusted and adjusted response rates. Adjusted response rates were calculated using the Council of American Survey Research Organizations method, which adjusts for the expected number of ineligible nonrespondents based on the number of ineligible respondents (24). BMI was calculated based on self-reported weight and height (25–29) and was used to classify respondents as not overweight (BMI ⬍ 25 kg/m2) or overweight (ⱖ25 kg/m2) based on national and international guidelines (30,31). Respondents’ weight self-perceptions of “thin” and “average” were collapsed into one “non-overweight” category for analysis purposes to distinguish from “overweight.” Weight self-perceptions were classified as accurate when respondents self-identified as “non-overweight” and BMI was ⬍25 kg/m2 or self-identified as “overweight” and BMI was ⱖ25 kg/m2; other combinations constituted misclassification. To analyze the ease of counseling variable, we dichotomized respondents’ answers as follows: “easier” and “no difference” vs. “more difficult.” Using logistic regression, we examined the association between weight self-perception misclassification and other potential covariates. Gender, age, BMI, ethnicity (white compared with all others), practice type (nonacademic vs. academic), practice location (rural and urban each compared with the referent suburban), and percentage of low-income patients served were examined as potential covariates in the regression model. Using logistic regression, we also examined the unadjusted relationship between ease of counseling and self perceptions of “thin” and “overweight” (each vs. the referent of average). We used logistic regression to examine this relationship adjusted for potential confounders of the relationship. These potential confounders were self-reported BMI weight status itself (overweight vs. nonoverweight), gender, age, ethnicity (white compared with all others), practice type (nonacademic vs. academic), practice location (rural and urban each compared with the referent suburban), and percentage of low-income patients served. Finally, to discern whether it was truly self-perception or actual weight status that determined difficulty of counseling for individuals, we created four classifications from our data to use in a regression model. These classifications were as follows: 1) those perceiving themselves as “thin” or “average” and were not overweight according to self-reported BMI (“correctly classified, not overweight”); 2) those perceiving themselves as “thin” or “average” but were overweight according to self-reported BMI (“incorrectly classiOBESITY RESEARCH Vol. 13 No. 2 February 2005

327


Table 1. Gender-specific accuracy of pediatrician self-classification of weight status based on self-reported height and weight All respondents (N ⴝ 355)

Not overweight* Overweight*

n

Percent

Percent classifying correctly

213 142

60% 40%

98% 51%

Men (n ⴝ 191)

n 92 99

Women (n ⴝ 164)

Percent


n

Percent


48% 52%

99% 43%

121 43

74% 26%

98% 67%

* Standard of overweight used was BMI ⱖ 25 kg/m2 from national and international guidelines for definition of overweight/obese (30,31).

fied, not overweight”); 3) those perceiving themselves as “overweight” and were actually overweight according to self-reported BMI (“correctly classified, overweight”); and 4) those perceiving themselves as “overweight” but were not overweight according to self-reported BMI (“incorrectly classified, overweight”). We used logistic regression to examine the association between ease of counseling and perceived weight status, first independently and then adjusted for potential confounders. Although our sample size at the lower end of the spectrum did not permit us to use underweight (defined as BMI ⬍ 18.5 kg/m2), and there is no consensus BMI standard cut-off for “thin,” we did adjust for overweight vs. not overweight as described above.

Results Response Rate and Demographics Of the 738 pediatricians who originally seemed to meet eligibility requirements, 524 returned the questionnaire, but 168 responded that they do not practice “routine health care.” We, therefore, analyzed results based on 355 respondents. Our unadjusted response rate was, therefore, 62%, and our adjusted response rate was 71% (24). Of these 355 respondents, 46% were women, and the mean age was 48 years. The majority was white (89%). Only 19% were affiliated with an academic institution. More practiced in suburban (44%) than rural (27%) or urban (29%) communities, and the mean proportion of low-income patients served by their practices was 43%. Misclassification of Overweight Of the respondents, 142 (40%) were overweight using national and international guidelines (30,31). More men (n ⫽ 99, 52%) than women (n ⫽ 43, 26%) were overweight. Nearly all classifying themselves as “overweight” (n ⫽ 75, 98%) were accurate in that assessment. However, nearly one-half (49%) of the overweight physicians misclassified their weight status as “non-overweight” (n for “thin” ⫽ 49; 328


n for “average” ⫽ 230). Overweight men were more likely than women to misclassify (57% vs. 43%, respectively; p ⬍ 0.001). Greater physician age was associated with higher BMI (r ⫽ 0.19; p ⫽ 0.0004) but not weight status misclassification [odds ratio (OR) ⫽ 0.98; 95% confidence interval (CI) ⫽ 0.95, 1.02]. Of covariates examined, only gender and BMI itself influenced the odds of misclassifying overweight. The adjusted odds of misclassification were 3.61 times greater for men than women (95% CI ⫽ 1.81, 7.21), whereas BMI had an OR of 1.11 (95% CI ⫽ 1.03, 1.19). We repeated the analysis adjusting for respondent age, which yielded similar (⬍1.4% difference) results, and we, therefore, do not report age-adjusted results. Table 1 reflects the gender-specific percentages of overweight and nonoverweight and rates of misclassification. Self-perception of Weight and Ease of Counseling Assessing the independent impact on counseling, those who perceived themselves as “thin” had over five times the unadjusted odds of finding obesity counseling “more difficult” compared with “average” weight physicians, and those who perceived themselves as “overweight” had over two times the odds of finding counseling “more difficult,” although this was not statistically significant. When adjusted for potential covariates including BMI status (overweight vs. not overweight based on self-reported height and weight), the relationship between self-identified “thinness” and counseling difficulty increased (OR ⫽ 5.69; 95% CI ⫽ 2.30, 14.1) as did the relationship between self-identified overweight and difficulty counseling (OR ⫽ 3.84; 95% CI ⫽ 1.11, 13.3) (Table 2). To further tease apart whether it was self-perception of weight or actual weight that determined ease of counseling (at least at the upper end of the BMI spectrum), we examined the effects of various combinations of correctly and incorrectly classified weight status on ease of counseling. Only one such classification combination was significantly related to ease of counseling. Those who inappropriately


Table 2. Perceived effect of weight self-perception status on counseling ability (N ⫽ 355)

Number of pediatricians

Unadjusted OR and 95% CI for making counseling “more difficult”*

Adjusted OR† and 95% CI for making counseling “more difficult”

49 230 76

5.13 (2.25,11.7) 1.00 2.18 (0.94,5.10)

5.69 (2.30,14.1) 1.00 3.84 (1.11,13.3)

142 All 191 285 All 97 101 320

0.67 (0.33,1.34) 1.01 (0.97,1.05) 0.66 (0.34,1.30) 0.83 (0.36,1.91) 1.00 (0.99,1.01) 1.08 (0.51,2.26) 0.76 (0.34,1.66) 0.93 (0.31,2.81)

0.62 (0.18,2.09) 0.99 (0.94,1.04) 0.81 (0.34,1.94) 0.61 (0.23,1.59) 0.99 (0.98,1.01) 1.59 (0.59,4.26) 0.80 (0.31,2.08) 1.39 (0.38,5.13)

Weight self-perception “Thin” “Average” “Overweight” Potential confounders Overweight Age (continuous) Male Non-academic (vs. academic) Percent low-income patients (continuous) Rural (compared with referent of suburban) Urban (compared with referent of suburban) White (compared with all others)

* “Easier” and “no difference” collapsed to contrast with “more difficult” for the analysis that yielded these results. † Adjusted for BMI category based on self-reported weight and height (overweight vs. not overweight), age, gender, practice type (nonacademic vs. academic), percent of low-income patients, community type (rural and urban compared with suburban referent), and ethnicity (white compared with all others)

classified themselves as “thin” or “average” when they were actually overweight reported lower odds of difficulty counseling about overweight than their appropriately classified, non-overweight colleagues (unadjusted OR ⫽ 0.21; 95% CI ⫽ 0.05, 0.90; adjusted OR ⫽ 0.24; 95% CI ⫽ 0.05, 1.10).

Discussion This study had two interesting findings worthy of discussion. First, nearly one-half (49%) of overweight pediatricians in our sample did not classify themselves as such. Second, pediatricians identifying themselves as “thin” and also those identifying themselves as “overweight” reported more difficulty counseling about weight than did those identifying themselves as average. The misperception rate of overweight that we found in our pediatrician sample was worse than in United States’ nationally representative samples (19,20). Even in a 1991 nationally representative, cross-sectional survey of the U.S. population that also used self-reported weight and height to calculate respondents’ BMI, only 27.4% of overweight respondents thought that their weight was appropriate (20). Although highly educated and overwhelmingly of white ethnicity, both of which are associated with self-perceived overweight (32), pediatricians in our sample may have more

frequently misclassified themselves as average because they were not as overweight as other samples. Other possibilities for personal misclassification may be physicians’ biases about overweight (33,34), desire to appear as role models (35,36), and confusion about or disagreement with definitions. Regardless, further study should investigate the effects of such misclassification, such as whether physicians’ weight self-perception misclassification is related to their underdetection of patient overweight. In our study, men less accurately perceived their overweight than women, consistent with the literature on gender-related differences in weight self-perception (19,20,32,37– 40). In previous studies, women have more accurately perceived true overweight than men have (19,20,32), and women have reported goal BMIs more frequently in the normal weight range than men (37). Some previous research has found that pediatricians’ BMI itself is not associated with differences in attitudes, behaviors, or skills of obesity management or rates of healthy weight counseling (9 –11). Our examination of weight self-perception and its effects on ease of counseling yielded interesting and novel findings. Even once adjusted for weight status and other potential confounders, selfidentified “thin” pediatricians had nearly six times the odds as “average” weight people of reporting that their weight OBESITY RESEARCH Vol. 13 No. 2 February 2005

329


made healthy weight counseling more difficult, and selfidentified “overweight” pediatricians had nearly four times the adjusted odds of reporting increased counseling difficulties. We hypothesize that, while “overweight” physicians may worry that their counseling would seem hypocritical, “thin” physicians may worry that patients would perceive them as insufficiently empathetic or even insensitive to body image issues in a weight-conscious society. The finding that overweight physicians who do not perceive themselves as such report less difficulty counseling than their average weight and appropriately self-classifying colleagues is more difficult to explain. While this finding is consistent with our other findings, suggesting that perceptions may be greater counseling determinants than actualities, clearly further research with larger sample sizes of pediatricians and in-depth qualitative research is necessary to examine the association between weight self-perceptions and ease of counseling. Forty percent of pediatricians surveyed were overweight, less than the United States population’s average of 64%, recently determined by Flegal et al. (41), yet consistent with other studies of pediatricians using self-reported data. Rattay et al. (9) found a prevalence of overweight of 39.6%, while Kolagotla and Adams (10) found a prevalence of 43% among pediatricians. The differences between pediatricians’ and the general population’s levels of overweight may be explained by the differences between these groups. They may be caused by differences in ethnicity and/or socioeconomic status, which are known to affect weight status (3,41,42), the profession’s effect on personal health status, and/or social desirability response bias. The fact that our study used self-reported heights and weights, whereas Flegal et al. used measured values, calls response bias into question. Other reports have found an underestimation of the prevalence of obesity based on self-reported weight and height data (43,44). Our study’s major limitation is its reliance on self-reported, rather than measured, height and weight to derive BMI and weight status classifications. Because we compared weight status classifications to self-perception, errors in BMI would result in a flawed analysis. However, many studies in adults have found that self-reported height and weight are fairly accurate (25–28). Physicians may have particularly accurate BMI reporting (45). However, if selfreport overestimated height and/or underestimated weight, as in some previous reports (26,39,46,47), the resultant systematically underestimated BMI—what some have termed an “amplification of bias” (43)— could mean that prevalence and misclassification of overweight are even greater than described here. Women, restrained eaters, and those who are overweight are at greater risk for underreporting their weight (43,48 –51). While self-reported height and weight were more practical to collect for this exploratory study and enabled us to infer conclusions from and 330


make comparisons to other studies that also used selfreported data (10,20,32), further research using measured height and weight data is essential to confirm our unexpected findings. Our study’s generalizability is limited to practicing pediatricians in North Carolina. Furthermore, because only those pediatricians who were members of the state’s pediatrics professional society and the AAP were sent surveys, some of the state’s pediatricians may have been missed, leading to a survey bias. However, the NCPS is an active state chapter of the AAP and boasts membership of the majority of the state’s board-certified pediatricians (NCPS, personal communication). Although we do not have information on nonrespondents, even our unadjusted response rate of 62% was higher than national averages for mailed physician questionnaires (52). Another limitation is that overweight respondents may have identified themselves as “average” in relation to their perception of population norms, thus making our misclassification percentage artificially high. However, given the fact that “overweight” was a listed response option, we believe that most knowingly overweight respondents would choose such an option. Finally, only face validity confirms the weight self-perception question. However, because we were interested in people’s perceptions of their own weight status, such perceptions need to be considered valid in their own right, and their relationship to actual weight status may be imperfect, yet still interesting. Our work extends a growing body of knowledge about barriers to obesity management and suggests that, especially for self-identified thin physicians, weight self-perceptions could be one barrier. The extent of misclassification of weight status that we found, coupled with the strength of the association between weight self-perception and ease of counseling, may help to explain why looking at BMI alone as a predictor of counseling patterns has yielded conflicting results (9 –11,13,15,16). Rather than actual BMI, it may be the perception of weight that more influences counseling. Although likely difficult, helping pediatricians overcome personal weight-related obstacles may enable them to become better counselors for achieving healthy weight in their patients.

Acknowledgments This work was made possible by the following sources: the Robert Wood Johnson Foundation; the Institute of Nutrition and Centers for Disease Control Grant H75/ CCH420060 – 01; and the NIH-funded K12 UNC BIRCWH Career Development Program (HD01441). We thank Halle Amick, who provided administrative assistance and tireless data entry, Desmond Runyan, Norman Miller, and Andrew Perrin for careful manuscript review, and William Miller for methodological assistance.


References 1. Rippe JM. The case for medical management of obesity: a call for increased physician involvement. Obes Res. 1998; 6(Suppl 1):23S–33S. 2. Rippe JM, McInnis KJ, Melanson KJ. Physician involvement in the management of obesity as a primary medical condition. Obes Res. 2001;9(Suppl 4):302S–11S. 3. U.S. Department of Health and Human Services. Healthy People 2010: Understanding and Improving Health. 2nd ed. Washington, DC: U.S. Government Printing Office; 2000. 4. U.S. Department of Health and Human Services. The Surgeon General’s Call to Action to Prevent and Decrease Overweight and Obesity 2001. Rockville, MD: U.S. Department of Health and Human Services, Public Health Service, Office of the Surgeon General; 2001. 5. American Academy of Pediatrics; Committee on Psychosocial Aspects of Child and Family Health. Guidelines for Health Supervision III. Elk Grove Village, IL: American Academy of Pediatrics; 2002. 6. American Medical Association. Guidelines for Adolescent Preventive Services (GAPS): Recommendations Monograph. 2nd ed. Chicago, IL: American Medical Association; 1996. 7. Hamilton JL, James FW, Bazargan M. Provider practice, overweight and associated risk variables among children from a multi-ethnic underserved community. J Natl Med Assoc. 2003;95:441– 8. 8. Denen M, Hennessey J, Markert R. Outpatient evaluation of obesity in adults and children: a review of the performance of internal medicine/pediatrics residents. J Gen Intern Med. 1993;8:268 –70. 9. Rattay KT, Fulton JE, Galuska DA. Weight counseling patterns of U.S. Pediatricians. Obes Res. 2004;12:161–9. 10. Kolagotla L, Adams W. Ambulatory management of childhood obesity. Obes Res. 2004;12:275– 83. 11. Story MT, Neumark-Stzainer DR, Sherwood NE, et al. Management of child and adolescent obesity: attitudes, barriers, skills, and training needs among health care professionals. Pediatrics. 2002;110:210 – 4. 12. Lewis CE, Clancy C, Leake B, Schwartz JS. The counseling practices of internists. Ann Intern Med. 1991;114:54 – 8. 13. Lewis CE, Wells KB, Ware J. A model for predicting the counseling practices of physicians. J Gen Intern Med. 1986; 1:14 –9. 14. Geller AC, Robinson J, Silverman S, Wyatt SA, Shifrin D, Koh HK. Do pediatricians counsel families about sun protection: a Massachusetts survey. Arch Pediatr Adolesc Med. 1998;152:372– 6. 15. Frank E, Wright EH, Serdula MK, Elon LK, Baldwin G. Personal and professional nutrition-related practices of US female physicians. Am J Clin Nutr. 2002;75:326 –32. 16. Wells KB, Lewis CE, Leake B, Ware JE Jr. Do physicians preach what they practice? A study of physicians’ health habits and counseling practices. JAMA. 1984;252:2846 – 8. 17. Cornuz J, Ghali WA, Di Carlantonio D, Pecoud A, Paccaud F. Physicians’ attitudes towards prevention: importance of intervention-specific barriers and physicians’ health habits. Fam Pract. 2000;17:535– 40.

18. Freed GL, Clark SJ, Lohr JA, Sorenson JR. Pediatrician involvement in breast-feeding promotion: a national study of residents and practitioners. Pediatrics. 1995;96:490 – 4. 19. Chang VW, Christakis NA. Self-perception of weight appropriateness in the United States. Am J Prev Med. 2003;24:332–9. 20. Chang VW, Christakis NA. Extent and determinants of discrepancy between self-evaluations of weight status and clinical standards. J Gen Intern Med. 2001;16:538 – 43. 21. Rand CSW, Kuldau JM. The epidemiology of obesity and self-defined weight problem in the general population: gender, race, age, and social class. Int J Eat Disord. 1990;9:329 – 43. 22. Crawford D, Campbell K. Lay definitions of ideal weight and overweight. Int J Obes Relat Metab Disord. 1999;23:738 – 45. 23. Perrin EM, Flower KB, Ammerman AS. Body mass index charts: useful yet underused. J Pediatr. 2004;144:455– 60. 24. Frankel L. The report of the CASRO task force on response rates. In: Wiseman F, ed. Improving Data Quality in a Sample Survey. Cambridge, MA: Marketing Science Institute; 1983, pp. 1–11. 25. Stewart AL. The reliability and validity of self-reported weight and height. J Chronic Dis. 1982;35:295–309. 26. Stewart AW, Jackson RT, Ford MA, Beaglehole R. Underestimation of relative weight by use of self-reported height and weight. Am J Epidemiol. 1987;125:122– 6. 27. Stunkard AJ, Albaum JM. The accuracy of self-reported weights. Am J Clin Nutr. 1981;34:1593–9. 28. Jeffery RW. Bias in reported body weight as a function of education, occupation, health and weight concern. Addict Behav. 1996;21:217–22. 29. Cameron R, Evers SE. Self report issues in obesity and weight management. Behav Assess. 1990;12:91–106. 30. National Institutes of Health; National Heart Lung and Blood Institute. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report. Bethesda, MD: National Institutes of Health; National Heart, Lung, and Blood Institute; 1998. 31. World Health Organization. Prevention and Management of the Global Epidemic of Obesity. Report of the WHO Consultation on Obesity. Geneva, Switzerland: World Health Organization; 1997. 32. Paeratakul S, White MA, Williamson DA, Ryan DH, Bray GA. Sex, race/ethnicity, socioeconomic status, and BMI in relation to self-perception of overweight. Obes Res. 2002;10:345–50. 33. Maiman LA, Wang VL, Becker MH, Finlay J, Simonson M. Attitudes toward obesity and the obese among professionals. J Am Diet Assoc. 1979;74:331– 6. 34. Schwartz MB, Chambliss HO, Brownell KD, Blair SN, Billington C. Weight bias among health professionals specializing in obesity. Obes Res. 2003;11:1033–9. 35. Loomis GA, Connolly KP, Clinch CR, Djuric DA. Attitudes and practices of military family physicians regarding obesity. Mil Med. 2001;166:121–5. 36. Price JH, Desmond SM, Krol RA, Snyder FF, O’Connell JK. Family practice physicians’ beliefs, attitudes, and practices regarding obesity. Am J Prev Med. 1987;3:339 – 45. 37. Serdula MK, Mokdad AH, Williamson DF, Galuska DA, Mendlein JM, Heath GW. Prevalence of attempting weight loss and strategies for controlling weight. JAMA. 1999;282: 1353– 8. OBESITY RESEARCH Vol. 13 No. 2 February 2005

331


38. Leonhard ML, Barry NJ. Body image and obesity: effects of gender and weight on perceptual measures of body image. Addict Behav. 1998;23:31– 4. 39. Goodman E, Hinden BR, Khandelwal S. Accuracy of teen and parental reports of obesity and body mass index. Pediatrics. 2000;106:52– 8. 40. Field AE, Camargo CA Jr, Taylor CB, Berkey CS, Roberts SB, Colditz GA. Peer, parent, and media influences on the development of weight concerns and frequent dieting among preadolescent and adolescent girls and boys. Pediatrics. 2001; 107:54 – 60. 41. Flegal KM, Carroll MD, Kuczmarski RJ, Johnson CL. Overweight and obesity in the United States: prevalence and trends, 1960 –1994. Int J Obes Relat Metab Disord. 1998;22: 39 – 47. 42. Schoenborn CA, Adams PF, Barnes PM. Body weight status of adults: United States, 1997–98. Adv Data. 2002;330: 1–15. 43. Roberts RJ. Can self-reported data accurately describe the prevalence of overweight? Public Health. 1995;109:275– 84. 44. Kuskowska-Wolk A, Bergstrom R, Bostrom G. Relationship between questionnaire data and medical records of height, weight and body mass index. Int J Obes Relat Metab Disord. 1992;16:1–9.

332


45. Klag MJ, He J, Mead LA, Ford DE, Pearson TA, Levine DM. Validity of physicians’ self-reports of cardiovascular disease risk factors. Ann Epidemiol. 1993;3:442–7. 46. Engstrom JL, Paterson SA, Doherty A, Trabulsi M, Speer KL. Accuracy of self-reported height and weight in women: an integrative review of the literature. J Midwifery Women’s Health. 2003;48:338 – 45. 47. Strauss RS. Comparison of measured and self-reported weight and height in a cross-sectional sample of young adolescents. Int J Obes Relat Metab Disord. 1999;23:904 – 8. 48. Villanueva EV. The validity of self-reported weight in US adults: a population based cross-sectional study. BMC Public Health. 2001;1:11. 49. Rossouw K, Senekal M, Stander I. The accuracy of selfreported weight by overweight and obese women in an outpatient setting. Public Health Nutr. 2001;4:19 –26. 50. Larson MR. Social desirability and self-reported weight and height. Int J Obes Relat Metab Disord. 2000;24:663–5. 51. Shapiro JR, Anderson DA. The effects of restraint, gender, and body mass index on the accuracy of self-reported weight. Int J Eat Disord. 2003;34:177– 80. 52. Cummings SM, Savitz LA, Konrad TR. Reported response rates to mailed physician questionnaires. Health Serv Res. 2001;35:1347–55.