Correspondence
Actual versus Self-Reported Intake and Exercise in Obesity
N Engl J Med 1993; 328:1494-1496May 20, 1993
- Article
To the Editor:
Lichtman et al. (Dec. 31 issue)1 investigated why some obese persons have difficulty losing weight, although their diet is reportedly hypocaloric. The authors concluded that “failure to lose weight despite a self-reported low caloric intake can be explained by substantial misreporting of food intake and physical activity.” We do not concur with their inference.
In their analyses, Lichtman et al. studied 10 subjects in the “diet-resistant” group (group 1) and 6 in the control group (group 2). Self-reports of food intake and physical activity were compared with indirect calorimetry and body-composition analysis. The subjects in group 1 underreported their food intake by a mean (±SD) of 47 ±16 percent (95 percent confidence interval, 36 to 58 percent) and overreported their physical activity by 51 ±75 percent (95 percent confidence interval, -3 to +105 percent). (It appears that the authors, who reported a value of P<0.05 for the latter result, used a one-sided test, since the two-sided confidence interval that we calculated indicates a value of P>0.05.) Subjects in group 2 underreported their food intake by 19 ±38 percent (95 percent confidence interval, -21 to +59 percent) and overreported their physical activity by 30 ±43 percent (95 percent confidence interval, -15 to +75 percent). Both confidence intervals for group 2 include upper bounds that are consistent with the findings for group 1.
In their significance tests, the authors did not make the relevant comparisons. Specifically, they did not report whether the discrepancies observed in group 1 are of greater magnitude than those seen in group 2. With a standard two-sample t-test, the differences between self-reported and actual food intake (P>0.10) or physical activity (P>0.50) are not significantly different2 in the two groups.
Thus, the observed differences between group 1 and group 2 could have occurred by chance. These data show apparently striking differences based on very small samples. They cannot provide conclusive evidence that misreporting of food intake and physical activity explains the phenomenon of diet resistance in some obese persons.
2 ReferencesI-Min Lee, M.B., B.S., Sc.D.
Nancy R. Cook, Sc.D.
Charles H. Hennekens, M.D., Dr.P.H.
Harvard University, Boston, MA 021151
Lichtman SW ,Pisarska K ,Berman ER , et al. Discrepancy between self-reported and actual caloric intake and exercise in obese subjects. N Engl J Med 1992;327:1893-1898
Full Text | Web of Science | Medline2
Rosner B. Fundamentals of biostatistics. 3rd ed. Boston: PWS-Kent, 1990.
To the Editor:
Lichtman et al. confirm an impression held by clinicians: patients who say they cannot lose weight despite the fact that they “don't eat anything and exercise every day” are misrepresenting (to themselves, presumably) their caloric intake and energy expenditure. Yet in the accompanying editorial Danforth and Sims appear to use the same denial mechanism as patients who protest, “It's my metabolism, doctor.”1 In the face of demonstrated behavioral abnormalities and a clear absence of metabolic abnormalities, they conclude that “a disorder of metabolism may . . . lead to inappropriate appetite . . . as well [as] to the incorrect estimation of intake and activity described” by Lichtman et al. (i.e., it is the metabolism after all). At best, it is unclear how a metabolic defect might lead patients to misrepresent their food intake and energy expenditure. The authors also propose, “It is possible that the level of physical activity, as well as that of food intake, is genetically determined.”1 This must certainly be a complex gene cluster.
Although it is clear that there are genetic determinants of obesity, it is equally clear that behavior plays some part. Patients cannot change their genes, but they can change their behavior. Change starts with recognizing and taking responsibility for existing behavior. Researchers and clinicians might assist patients in this process, rather than abetting them in their denial.
1 ReferencesAmi Laws, M.D.
Stanford University School of Medicine, Stanford, CA 943051
Danforth E Jr ,Sims EAH . Obesity and efforts to lose weight. N Engl J Med 1992;327:1947-1948
Full Text | Web of Science | Medline
To the Editor:
In connection with the study by Lichtman et al., it is of interest to recall what Samuel Johnson had to say about the cause of obesity as he conversed with his biographer on April 28, 17831:
Talking of a man who was grown very fat, so as to be incommoded with corpulency:Johnson: He eats too much, Sir.Boswell: I don't know, Sir; you will see one man fat, who eats moderately, another lean, who eats a great deal.Johnson: Nay, Sir, whatever may be the quantity that a man eats, it is plain that if he is too fat, he has eaten more than he should have done. One man may have a digestion that consumes food better than common; but it is certain that solidity is encreased by putting something to it.As the saying goes, there seems to be nil novi sub sole.
1 ReferencesGilbert B. Forbes, M.D.
University of Rochester, School of Medicine and Dentistry, Rochester, NY 14642To the Editor:
I am a bit dismayed at some of the editorial opinions expressed by Drs. Danforth and Sims.1 My first concern is with the notions that “there is a strong cultural emphasis on being lean in our society” and that “obesity is not sought after; it occurs spontaneously.” I believe that these statements are true, but only in a very superficial sense. The stronger although perhaps more subliminal emphasis in our society is actually on eating food and drinking beverages that often have poor nutritional quality and contain excess calories. Such behavior is reinforced by various factors, such as advertising, ignorance, and misconceptions about what constitutes a healthful diet. Although I have no personal primary research to back my opinions, I invite Drs. Danforth and Sims to spend a couple of hours at a checkout counter of a busy American supermarket and observe overweight people and what is in their shopping carts.
A number of my friends who do not live in the United States have remarked to me that there seem to be more overweight people in this country than they are used to seeing in their own countries. It would be very helpful to compare statistical data from other industrialized countries over time to assess whether this is actually true. My guess is that such a study would uncover important cultural aspects of diet that might prove useful in further defining the natural history and treatment of obesity.
1 ReferencesDonald E. Casey, Jr., M.D.
1818 Airfield, Kingman, AZ 864011
Danforth E Jr ,Sims EAH . Obesity and efforts to lose weight. N Engl J Med 1992;327:1947-1948
Full Text | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: With respect to the comments of Lee et al., the P values were calculated on the difference in the number of calories between reported and expected energy intake and between reported and expected activity expenditure for each subject in group 1. The actual percentages of overreporting and underreporting were included only for expository purposes. The placement of P values in the text may not make this clear.
We do not agree that the explanation for the failure to lose weight in group 1 requires a comparison with group 2. Rather, the explanation is sought in the discrepancy between reported and expected values in various terms of the energy-balance equation. The doubly-labeled-water technique provided the energy-expenditure side of the equation and, if body composition is assumed to be stable, provided the expected value for the intake. Deviations of reported from expected intake were different from zero (P<0.05) and are the basis for the conclusion that underreporting is part of the explanation for the failure to lose weight. Similar reasoning applies to the levels of energy expended in physical activity. It follows from our conclusion that the percentage of underreporting should be greater in group 1 than in group 2. However, the power of the statistical test is low, and this accounts for the failure to find statistical significance.
Regarding the comments of Dr. Laws, behavior as well as metabolism may have genetic determinants. We currently do not know the degree to which enduring changes can be effected in behavior that contributes to obesity.
The exchange between Boswell and Johnson serves to illustrate the long history of the concept of diet resistance. Von Noorden and others in the late 19th century also observed subjects with “endogenous obesity” whom they described as “resistant to treatment by dieting” and having a “weak metabolism”1. In 1930 Newburgh and Woodwell-Johnston reported the viewpoint of some investigators that there are obese subjects who are the “unfortunate victims of constitutional disease that unrelentingly causes a progressive deposition of adipose tissue, independent of activity or dietary habit”2. The belief that some obese persons have very low energy expenditures and thus low caloric requirements persists today. For example, in 1992 Callaway stated, “the idea that everyone will lose weight on a 1200-calorie diet is silly”3. Some practitioners, not to mention patients, continue to believe that there are euthyroid obese persons with very low energy requirements. We hope this belief can finally be put to rest through the use of convincing modern research techniques, such as doubly labeled water.
3 ReferencesSteven B. Heymsfield, M.D.
Stanley Heshka, Ph.D.
Steven W. Lichtman, Ed.D.
St. Luke's-Roosevelt Hospital Center, New York, NY 100251
Schwartz H. Never satisfied: a cultural history of diets, fantasies, and fat. New York: Anchor Books, 1990:139.
2
Newburgh LH ,Johnston MW . Endogenous obesity -- a misconception. Ann Intern Med 1930;3:815-825
Web of Science3
Bovsun M. The diet dilemma. Medical World News. May 1992:17-25.
Author/Editor Response
In our editorial, we attempted to ward off the conclusion that obesity is only a matter of gluttony and sloth. It is naive to assume that most obese people, perhaps deliberately, simply fool themselves and others about their food intake and physical activity.
Lichtman et al. used the new doubly-labeled-water technique to investigate the possibility that a low “metabolism,” or basal and total energy expenditure, could explain the inability to lose weight. They were not exploring the many metabolic and neurologic mechanisms that contribute to obesity syndromes in animals and humans. There is a strong heritable component to obesity syndromes. This has been unequivocally documented in several studies of twins in which genetic factors were the major determinants of body fatness and could account for 70 percent of its variance1. There are metabolic disorders that lead to inappropriate craving for food, inactivity, and preferential storage of energy as fat2. An inherently low rate of energy expenditure antecedent to the development of obesity has been recognized3. We apparently did not make it clear that, just as many factors contribute to inappropriate eating behavior, some of these same factors, plus perhaps the intense frustration of living with the problem, may contribute to the faulty reporting.
More than 80 percent of obese subjects who lose weight regain it. “Behavioral therapies” have not provided better results. These findings reinforce the view that obesity is an inherent disorder rather than the result of denial or “bad habits.” Blaming the patient for simple misbehavior creates what has been called the “diet relationship trap,”4 in which there is a breakdown in the normal therapeutic rapport between doctor and patient. We prefer not to blame a person for a lack of willpower, when he or she is already overwhelmed by a complex problem and therefore lacks self-esteem.
We agree with Dr. Casey about the importance of social and cultural factors, including the high-fat diet that predisposes people to obesity. We believe that physicians must work with their patients to combat these factors. Because behavioral treatment alone has had limited success, the underlying mechanism of obesity should be further identified and addressed directly.
4 ReferencesElliot Danforth, Jr., M.D.
Ethan A.H. Sims, M.D.
University of Vermont, Burlington, VT 054051
Stunkard AJ ,Harris JR ,Pedersen NL ,McClearn GE . The body-mass index of twins who have been reared apart. N Engl J Med 1990;322:1483-1487
Full Text | Web of Science | Medline2
Bjorntorp P, Brodoff BN, eds. Obesity. Philadelphia: J.B. Lippincott, 1992.
3
Ravussin E ,Lillioja S ,Knowler WC , et al. Reduced rate of energy expenditure as a risk factor for body-weight gain. N Engl J Med 1988;318:467-472
Full Text | Web of Science | Medline4
Garrow JS . Treatment of obesity. Lancet 1992;340:409-413
CrossRef | Web of Science | Medline
- Citing Articles (1)
Citing Articles
1
Aristides M. Machado-Rodrigues, Manuel J. Coelho-E-Silva, Jorge Mota, Edilson Cyrino, Sean P. Cumming, Chris Riddoch, Gaston Beunen, Robert M. Malina. (2011) Agreement in activity energy expenditure assessed by accelerometer and self-report in adolescents: Variation by sex, age, and weight status. Journal of Sports Sciences1-12
CrossRef
