What You Can Change . . . And What You Can't*: The Complete Guide to Successful Self-Improvement (51 page)

21
. This is not an original proposal. It is the interpretation put forward by Polivy and Herman in
Breaking the Diet Habit
, Garner and Wooley in “Confront the Failure of Behavioral and Dietary Treatments,” and Jane Wardle. See particularly J. Polivy and C. P. Herman, “Dieting and Binging: A Causal Analysis,”
American Psychologist
40 (1985): 193–201; and J. Wardle, “Compulsive Eating and Dietary Restraint,”
British Journal of Clinical Psychology
26 (1987): 47–55.

In putting forward a similar theory both of anorexia and bulimia, Peter Slade calls the bulimic a “failed anorexic,” failed in that the anorexic can keep dieting without giving in to binging. See P. Slade, “Towards a Functional Analysis of Anorexia Nervosa and Bulimia Nervosa,”
British Journal of Clinical Psychology
21 (1982): 167–79. See M. Boskind-Lodahl and J. Sirlin, “The Gorging-Purging Syndrome,”
Psychology Today
(March 1977), 50–52, 82–85, for a survey of the rate of dieting just before bulimia starts. For two epidemiological studies linking dieting quite tightly to eating disorders, see G. Patton, E. Johnson-Sabine, K. Wood, et al., “Abnormal Eating Attitudes in London Schoolgirls—A Prospective Epidemiological Study: Outcome at Twelve Month Follow-up,”
Psychological Medicine
20 (1990): 383–94; and K. Kendler, C. MacLean, M. Neale, et al., “The Genetic Epidemiology of Bulimia Nervosa,”
American Journal of Psychiatry
148 (1991): 1627–37. Dieters are at least eight times more likely to get an eating disorder than nondieters.

22
. I take my hat off to S. Dalvit-McPhillips, “A Dietary Approach to Bulimia Treatment,”
Physiology and Behavior
33 (1984): 769–75. It is a shame that this paper has remained obscure.

23
. See Fairburn and Beglin, “Studies of the Epidemiology of Bulimia Nervosa,” for a complete review of the cognitive-behavioral studies, and C. Fairburn, R. Jones, R. Peveler, et al., “Three Psychological Treatments for Bulimia Nervosa,”
Archives of General Psychiatry
48 (1991): 463–69, for an exemplary outcome study. For a fine antidepressant drug study, see U. McCann and W. S. Agras, “Successful Treatment of Nonpurging Bulimia Nervosa with Desimpramine: A Double-Blind, Placebo-Controlled Study,”
American Journal of Psychiatry
147 (1990): 1509–13. The two outcome studies that explicitly compare drugs and cognitive-behavioral therapy are J. Mitchell, R. Pyle, E. Eckert, et al., “A Comparison Study of Antidepressants and Structured Intensive Group Psychotherapy in the Treatment of Bulimia Nervosa,”
Archives of General Psychiatry
47 (1990): 149–57; and W. S. Agras, E. Rossiter, B. Arnow, et al., “Pharmacologic and Cognitive-Behavioral Treatment for Bulimia Nervosa: A Controlled Comparison,”
American Journal of Psychiatry
149 (1992): 82–87.

24
. C. Telch, W. Agras, and E. Rossiter, “Binge Eating Increases with Increasing Adiposity,”
International Journal of Eating Disorders
7 (1988): 115–19. See G. T. Wilson, “Short-term Psychological Benefits and Adverse Effects of Dieting and Weight Loss,” paper delivered at the NIH conference “Methods for Voluntary Weight Loss and Control” (1991), for the argument that dieting may be necessary but not sufficient for bulimia.

25
. J. Brody, “Study Defines ‘Binge Eating Disorder,’”
New York Times
, 27 March 1992, A16.

26
. There is an inconclusive small literature about the effect of artificial sweeteners on weight gain. Some researchers have shown that rats gain weight with artificial sweeteners, but some have shown the reverse. It’s the same with human subjects. (See
Appetite
11 [1988]: supplement 1.) No one has attempted the right study: a long-term follow-up of overweight people who are put on (or taken off) artificial sweeteners. It would cost about $5 million to do this study and find out if the $10-billion-a-year diet-food industry is a scam.

27
. L. Craighead, A. Stunkard, and R. O’Brien, “Behavior Therapy and Pharmacotherapy for Obesity,”
Archives of General Psychiatry
38 (1981): 763–68.

28
. This study may prove to be a watershed, shifting the treatment of obesity from dietary (too ineffective) or surgical (too radical) to pharmacological. It was published in eight articles as “Long-term Weight Control: The National Heart, Lung, and Blood Institute Funded Multimodal Intervention Study,”
Clinical Pharmacology and Therapeutics
(supplement) (May 1992), 581–646.

29
. This footnote could be a chapter in length, but I am going to resist. Rather, I will be selective and highlight just the best studies to guide the student of this topic as to why I come to the conclusions I do:

Enormous Obesity (100 percent above “ideal” weight):
E. Drenick, S. Gurunanjappa, F. Seltzer, and D. Johnson, “Excessive Mortality and Causes of Death in Morbidly Obese Men,”
Journal of the American Medical Association
243 (1980): 443–45, followed two hundred extremely fat young men for ten years. They found that fifty died. Indeed, among the 25-to 34-year-olds, there was twelve times the expected death rate. But their subjects were people who had lost and regained weight from therapeutic fasts. It is just possible that dieting itself, not morbid obesity, contributed to sudden death.

Substantial Obesity (30 to 100 percent above “ideal “weight):
Choosing the low cut point is the most difficult and controversial part of distilling this large literature. Thirty percent overweight looks to me to be approximately the point at which some danger begins. The “Pooling Project” combined data for 12,381 men from eight different American populations who had been followed for five to ten years. Looking at first heart attack, the inflection point seems to be at 30 percent over “ideal” weight. This is a conservative estimate, however, since it can be argued from these data sets that there is no increase at all in coronary risk with weight. See The Pooling Research Project Research Group, “Relationship of Blood Pressure, Serum Cholesterol, Smoking Habit, Relative Weight and ECG Abnormality to Incidences of Major Coronary Events: Final Report of the Pooling Project,”
Journal of Chronic Diseases
31 (1978): 201–306; E. Barrett-Connor, “Obesity, Atherosclerosis, and Coronary Artery Disease,”
Annals of Internal Medicine
103 (1985): 1010–19.

The other essential source is Ancel Keys’s curve fitting of more than a dozen large-scale studies of body mass and coronary heart disease. His curves show two inflection points: one at the 30 percent overweight locus and the other at underweight. Importantly, the direction of the slope with mild overweight (o to 30 percent) is downward, with more weight up to about 30 percent associated with less risk. See A. Keys, “Overweight, Obesity, Coronary Heart Disease.”

See also E. Hammond and L. Garfinkel, “Coronary Heart Disease, Stroke, and Aortic Aneurysm,”
Archives of Environmental Health
19 (1969): 167–82. This is a landmark prospective study of one million people, and it shows increased risk with body weight. But the body weight at which the risk begins is in the upper 20 percent of the population, which seems to correspond to about the 30 percent overweight cut point.

One consoling note to people in this category: This amount of overweight probably does put you at risk in and of itself. Overweight correlates with high blood pressure and serum cholesterol. If your blood pressure is normal, if your cholesterol is normal, if you exercise moderately, and if you don’t smoke, your risk of heart attack is not increased by being overweight. See the classic A. Keys, C. Aravanis, H. Blackburn, et al., “Coronary Heart Disease: Overweight and Obesity as Risk Factors,”
Annals of Internal Medicine 77
(1972): 15–27.

Diabetes mellitus, in contrast to coronary heart disease, presents a clearer picture of risk with moderate overweight. In one prospective study, risk went up 1,000 percent with moderate obesity and 3,000 percent with 45 percent obesity. See K. Westlund and R. Nicholaysen, “Ten-Year Mortality and Morbidity Related to Serum Cholesterol,”
Scandinavian Journal of Clinical and Laboratory Investigations
30 (supplement 127) (1972), 3. See also A. Rimm, L. Werner, B. Van Yserloo, and R. Bernstein, “Relationship of Obesity and Disease in 73,532 Weight-Conscious Women,”
Public Health Reports
90 (1975): 44–51. My conclusion is that if diabetes runs in your family, you should be more attentive to overweight as a risk factor than if heart disease runs in your family.

Mild to Moderate Overweight (10 to 30 percent above “ideal” weight):
I have chosen my language carefully here—“possibly associated with a marginal increase in mortality”—because for every study that shows some health risk in this category, there is at least one that shows no health risk.

The thrust of the Keys and Barrett-Connor reviews is that there is little or no more heart-attack incidence in this range. Indeed, these people may be at lower risk than 0 percent overweight people.

On the other hand, the gray eminence is the Met Life “ideal weight” table, which in spite of its flaws is well entrenched in the public mind. More modern data supporting the view that even a little overweight is a health risk are found in J. Manson, G. Colditz, M. Stampfer, et al., “A Prospective Study of Obesity and Risk of Coronary Heart Disease in Women,”
New England Journal of Medicine
322 (1990): 882–89, which found risk starting at 10 percent overweight among female nurses; and T. Wilcovsky, J. Hyde, J. Anderson, et al., “Obesity and Mortality in the Lipid Research Clinics Program Follow-up Study,”
Journal of Clinical Epidemiology
43 (1990): 743–52, which found risk at low levels of overweight with men but not women. See also the National Institutes of Health Consensus Development and Conference Statement “Health Implications of Obesity,”
Annals of Internal Medicine
103 (1985): 147–51, which echoes the view that even a little overweight is bad. I find this report overly alarmist, and the tone is considerably moderated by the more recent National Research Council, “Obesity and Eating Disorders,” in National Academy of Sciences, National Research Council,
Diet and Health Risk: Implications for Reducing Chronic Disease Risk
(Washington, D.C.: National Academy Press, 1989), 563–92.

Nonetheless, there has never been a study that shows anything but a very small statistical increase of coronary risk in the 10 to 30 percent range of overweight.

Underweight:
See Keys, “Overweight, Obesity, Coronary Heart Disease;” S. Blair, H. Kohl, R. Paffenbarger, et al., “Physical Fitness and All-Cause Mortality: A Prospective Study of Healthy Men and Women,”
Journal of the American Medical Association
262 (1989): 2395–2401; R. Paffenbarger, P. Hyde, A. Wing, and C. C. Hsiesh, “Physical Activity, All-Cause Mortality, and Longevity of College Alumni,”
New England Journal of Medicine
314 (1986): 605–13. The mortality risk of the underweight, unlike that of the mild and moderately overweight, is strong, and it is not attributable to smoking or to subclinical disease. See also P. Sorlie, T. Gordon, and W. Kannel, “Body Build and Mortality: The Framingham Study,”
Journal of the American Medical Association
243 (1980): 1828–31.

One final caveat: No one has ever done the right kind of study to find out if overweight really causes premature death or poor health. The methodologically right study would randomly assign individuals to different weight levels and see who dies when. It seems likely that for ethical reasons this study will never be done in humans. All of the above studies are therefore distant second-best evidence.

30
. Hammond and Garfinkel, “Coronary Heart Disease, Stroke, and Aortic Aneurysm;” P. Hamm, R. Shekelle, and J. Stamler, “Large Fluctuations in Body Weight During Young Adulthood and the Twenty-Five-Year Risk of Coronary Death in Men,”
American Journal of Epidemiology
129 (1989): 312–18; L. Lissner, P. Odell, R. D’Agostino, et al., “Variability of Body Weight and Health Outcomes in the Framingham Population,”
New England Journal of Medicine
324 (1991): 1839–44; L. Lissner, C. Bengtsson, L. Lapidus, et al., “Body Weight Variability and Mortality in the Gothenburg Prospective Studies of Men and Women,” in P. Bjorntorp and S. Rossner, eds.,
Obesity in Europe 88
(London: Libbey, 1989), 51–56.

I use the word
probably
to describe the health risk of dieting. These four studies support it, and their findings do not come out of the blue. A number of smaller-scale studies of humans point to the same health risk for weight cycling, and so does the rat literature. See Garner and Wooley, “Confront the Failure of Behavioral and Dietary Treatments,” for references and a selective review. But I judge the risk “probable,” not “certain,” because two studies fail to find weight variability associated with mortality: L. Lissner, R. Andres, D. Muller, and H. Shimokata, “Body Weight Variability in Men: Metabolic Rate, Health, and Longevity,”
International Journal of Obesity
14 (1990): 373–83; and J. Stevens and L. Lissner, “Body Weight Variability and Mortality in the Charleston Heart Study” (letter to the editor), in
International Journal of Obesity
14 (1990): 385–86.