Correspondence
Body Weight and Mortality among Women
N Engl J Med 1996; 334:732-733March 14, 1996
- Article
To the Editor:
The study of overweight and mortality in women (Sept. 14 issue)1 by Manson and colleagues arrives at some conclusions that may complicate the treatment of women with eating disorders, who are already obsessed with the absolute value of being thin.
Conceptually, there are problems in assuming that weight (like cigarette smoking or exercise) is a controllable variable of habit and lifestyle. Overweight may or may not be primarily a behavioral effect — it may instead embody genetic factors that also predispose people to earlier mortality. Even if the authors are correct in concluding that higher weight is independently associated with increased mortality, this does not necessarily imply that active efforts to reduce weight will alter that risk. Nor can the study address the relative efficacy for an overweight woman of changing her activity level as compared with changing her caloric intake, nor the stressful consequences of prescribed dieting and heightened self-recrimination. The authors are quick to translate their findings regarding weight and mortality into social prescriptions: a revised “optimum” weight chart and a sanction for dieting. We owe it to our patients to be more cautious about recommending painful changes in behavior without demonstrable proof of increased life expectancy.
From the perspective of public policy, this report inadvertently fuels the obsession with thinness and the devotion to altering body shape that beset women in our culture. In their most drastic form, these preoccupations underlie anorexia nervosa and bulimia nervosa, eating disorders that cause extensive morbidity and mortality. Even in their ubiquitous milder form, they foster shame, social prejudice, and self-defeating dieting behavior. Manson and colleagues seem to conclude, like the Duchess of Windsor, that “you can never be too thin. . . .” Our patients with eating disorders remind us daily that this can be a dangerous conviction.
1 ReferencesPaul Hamburg, M.D.
Massachusetts General Hospital, Boston, MA 021141
Manson JE ,Willett WC ,Stampfer MJ , et al. Body weight and mortality among women. N Engl J Med 1995;333:677-685
Full Text | Web of Science | Medline
To the Editor:
The study by Manson et al. has resulted in consternation among many women who have not been blessed (or cursed) with a lean and hungry look. The almost universal interpretation is that “leaner is better.” The true message of the report is that there was no substantial elevation in mortality unless the body-mass index was over 27. Since this was the case, any trend toward increased mortality due to coronary heart disease or cancer in women who were even “mildly overweight” must have been matched by an equal trend toward increased mortality due to unknown causes in the very thin.
Marvin H. Lipkowitz, M.D.
Maimonides Medical Center, Brooklyn, NY 11219To the Editor:
Similar body weights do not always mean similar body compositions. A woman athlete may have the same weight as a sedentary woman of the same height, but the athlete may have 30 to 40 percent less fat.1 Muscles are heavy (80 percent water) as compared with body fat (5 to 10 percent water).2 The body-mass index of an athlete or other well-trained woman could therefore be misleading. A history of physical activity should be included in assessing the body-mass index.
2 ReferencesRose E. Frisch, Ph.D
Harvard School of Public Health, Boston, MA 021151
Frisch RE ,Snow RC ,Johnson LA ,Gerard B ,Barbieri R ,Rosen B . Magnetic resonance imaging of overall and regional body fat, estrogen metabolism, and ovulation of athletes compared to controls. J Clin Endocrinol Metab 1993;77:471-477
CrossRef | Web of Science | Medline2
Moore FD, Olesen KH, McMurrey JD, Parker HV, Ball MR, Boyden CM, eds. The body cell mass and its supporting environment: body composition in health and disease. Philadelphia: W.B. Saunders, 1963.
To the Editor:
Manson et al. clearly demonstrate that in middle-aged white women there is an almost direct relation between body-mass index and overall mortality. When examining the relation between obesity and mortality from coronary heart disease, however, the authors fail to acknowledge the even greater importance of “central obesity,” as measured by the ratio of waist circumference to hip circumference, in predicting death from coronary heart disease.
In the analysis of the association between the waist-to-hip ratio and mortality due to coronary heart disease, the authors found a relative risk of 3.7 and 8.7 in the two highest quintiles of the ratio (P = 0.04 for trend). This relative risk may actually be greater given the follow-up of only six years of those with values for the waist-to-hip-ratio, as well as the small percentage of deaths from cardiovascular causes seen in this middle-aged cohort. Although the data for men1 are clearer in establishing the superiority of the waist-to-hip ratio to the body-mass index in predicting the incidence of and mortality from coronary heart disease, evidence is accruing that the ratio is equally important as a predictor for women.2,3
Thus, although the message is that the increasing prevalence of overweight as measured by weight or the body-mass index may be harmful to women, the most important message is that perhaps not all obesity is the same. The waist-to-hip ratio gives a much more precise indication of the risk of coronary heart disease and thus provides a greater rationale for intervention. The authors need to elaborate further on the role of abdominal or central obesity, since this factor may have great relevance to screening and intervention in both public health and clinical-practice settings.
3 ReferencesTerry A. Jacobson, M.D.
Emory University School of Medicine, Atlanta, GA 30303Shirley Brown, R.D., M.S.
Grady Health System, Atlanta, GA 303351
Larsson B . Regional obesity as a health hazard in men -- prospective studies. Acta Med Scand Suppl 1988;723:45-51
Medline2
Folsom AR ,Kaye SA ,Sellers TA , et al. Body fat distribution and 5-year risk of death in older women. JAMA 1993;269:483-487
CrossRef | Web of Science | Medline3
Freedman DS ,Williamson DF ,Croff JB , et al. Relation of body fat distribution to ischemic heart disease. Am J Epidemiol 1995;142:53-63
Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: In response to Dr. Hamburg, it is important to recognize that the “spin” the media choose in covering a medical report is often quite different from the public health message provided by the investigators. As with any medical story that receives widespread press coverage, the hyperbole and sensationalism of the media should be dissociated from the science presented by the researchers. In the present case, the media often chose to interpret our findings as implying, “You can never be too thin. . . .” These were never the words of the investigators; quite the contrary, we cautioned reporters against that interpretation and emphasized the importance of avoiding unhealthful practices such as smoking or eating improperly to achieve leanness. Nonetheless, the media covered the story as they chose, and that was not under our control. Dr. Hamburg's concern should be tempered, however, by the recognition that the leading eating disorder in the United States is obesity, not anorexia nervosa or bulimia. We have an alarming and escalating epidemic of obesity that currently afflicts at least 58 million Americans.1 The public deserves to have information about the health effects of weight gain and obesity so that it can make informed lifestyle choices. The health benefits of weight loss among the overweight are incontrovertibly clear.2,3
Dr. Lipkowitz is incorrect in stating that the lean women in our study had a counterbalancing increase in mortality “due to unknown causes.” After accounting for biases from cigarette smoking and underlying disease, we found that mortality rates from cardiovascular disease, cancer, and “other causes” were each elevated among the overweight women. The risks of mortality from all causes among the overweight women were diluted, however, not counterbalanced, by causes of death not related to obesity.
Dr. Frisch states correctly that body composition may vary for a given body weight and that physical activity should be included in the assessment. As we noted in our report, control for physical activity in our final multivariate models did not materially alter our results. Also, in our study population of women 30 to 55 years of age on entry into the study, few participants were elite athletes. On the contrary, more than half reported a sedentary lifestyle, which is fairly representative of the U.S. population at large.4
Jacobson and Brown emphasize the importance of “central” obesity and the waist-to-hip ratio in predicting mortality due to coronary heart disease. They are incorrect, however, when they say we failed to acknowledge these associations. We stated clearly in the Results section that “the waist-to-hip ratio was a strong predictor of death due to coronary heart disease in this cohort” and provided the relative risks for each ratio quintile. Although central obesity may be a powerful predictor of mortality due to coronary heart disease, it is important to note that noncentral obesity is not metabolically benign.5 Thus, interventions involving weight reduction should include those with noncentral obesity as well as those with central obesity.
5 ReferencesJoAnn E. Manson, M.D., Dr.P.H.
Walter C. Willett, M.D., Dr.P.H.
Frank E. Speizer, M.D.
Brigham and Women's Hospital, Boston, MA 021151
Kuczmarski RJ ,Flegal KM ,Campbell SM ,Johnson CL . Increasing prevalence of overweight among US adults: the National Health and Nutrition Examination Surveys, 1960 to 1991. JAMA 1994;272:205-211
CrossRef | Web of Science | Medline2
Dattilo AM ,Kris-Etherton PM . Effects of weight reduction on blood lipids and lipoproteins: a meta-analysis. Am J Clin Nutr 1992;56:320-328
Web of Science | Medline3
Goldstein DF . Beneficial health effects of modest weight loss. Int J Obes Relat Metab Disord 1992;16:397-415
Web of Science | Medline4
Caspersen CJ ,Christenson GM ,Pollard RA . Status of the 1990 physical fitness and exercise objectives -- evidence from NHIS 1985. Public Health Rep 1986;101:587-592
Web of Science | Medline5
Young TK ,Gelskey DE . Is noncentral obesity metabolically benign? Implications for prevention from a population survey. JAMA 1995;274:1939-1941
CrossRef | Web of Science | Medline
