Howard Steiger Pierre Leichner A. Missagh Ghadirian

Perspectives on the Prevention of Anorexia Nervosa and Bulimia SUMMARY

SOMMAIRE

une Ample evidence suggests a rising incidence of I1 existe suffisamment de preuves suggerant nerveuse de l'anorexie de l'incidence augmentation anorexia nervosa and bulimia over the past et de la boulimie au cours des dernieres decennies. few decades. Correspondingly, medical Parallelement, nous comprenons mieux l'etiologie, la knowledge about the etiology, symptomatologie et le traitement de ces desordres alimentaires. Les medecins de famille, travailleurs de symptomatology and treatment of these eating disorders has increased. Often the front la sante souvent places en premier plan pour affronter ces desordres, sont dans une situation line health-care workers who treat these privilegiee pour les deceler precocement et les disorders, family physicians are in a key prevenir. La comprehension adequate des facteurs position for early detection and prevention of de risque pertinents, des symptomes et des signes these eating disorders. An adequate peut permettre au medecin d'intervenir a temps pour en empecher l'apparition dans certains cas ou, understanding of relevant risk factors, symptoms and signs may allow the physician dans d'autres cas, pour en minimiser la severite. to prevent the onset of an eating disorder in some patients or to minimize the severity of the illness in others. (Can Fam Physician 1987; 33:145-149.) Key words: eating disorders, anorexia nervosa, bulimia

_g_Sg_g___~~~'Eexs Dr. Steiger is a psychologist at the Eating Disorders Unit, Douglas Hospital Centre, Verdun, Que. Dr. Leichner is a psychiatrist at the Eating Disorders Unit, Douglas Hospital Centre. Dr. Ghadirian is a psychiatrist at the Eating Disorders Unit, Douglas Hospital Centre and also at the Allan Memorial Institute, Royal Victoria Hospital, Montreal, Que. Mrs. A, a 35-vear-old housewife, comes to vour office asking for a diet to lose a few pounds. She is 5'4" and weighs 130 lb. She, her husband and two children have recently moved to your town. She is somewhat depressed about the move. A brief historv is unremarkable, except for a family history of alcoholissm on her Jfther's side a(nd obesity oni her mother's. She her.self has obsessivecompulsive personality trcaits. Would you prescribe a diet?

HERE IS LITTLE DOUBT that I the prevalence of anorexia nervosa and bulimia has been increasing CAN. FAM. PHYSICIAN Vol. 33: JANUARY 1987

internationally during the past few decades, in percentages that are not explained solely bv recently increased medical attention to these disorders. " 2 One study done in the United Kingdom suggests that up to 5% of post-pubertal girls exhibit some anorexic symptoms.3 In the United States, the prevalence of bulimia among university-aged women is estimated to range from 4.5%-l13%.' A recent survey of teenaged, Canadian students showed that 22% of females and 6% of males are "at risk" for anorexia nervosa or bulimia in that they endorse beliefs and exhibit behaviours associated with these disorders.6 Alongside these disquieting figures, however, encouraging findings show a good prognosis, especially for anorexics who are managed appropriately; the likelihood of better outcome with early intervention; and a growing understanding of risk factors for these eating disorders." I These findings encourage an interest in early detection and prevention strategies. Family physicians can play a large part in implementing these strategies. Timely intervention might make the

difference between a chronic, deteriorating condition and a brief episode followed by full recovery.

Prevention and Eating Disorders Current thinking on prevention in the mental health sphere has followed Caplan's conceptualization,8 which suggests three levels of prevention: * primary prevention, which consists of early interruption of the chain of factors which causes a disorder, before the onset of symptoms; * secondary prevention, which reduces the prevalence of a disorder through early detection and prompt, effective treatment; * tertiary prevention, which consists of rehabilitation designed to promote a return to the highest-possible level of functioning following onset of the disorder. Because primary prevention requires a thorough knowledge of the cause of a disorder, it remains largely theoretical in its application to eating disorders.9, 10 Nevertheless, family physicians can practise some primary prevention of eating disorders. Al145

ready strides have been made in the secondary prevention of eating disorders. Early diagnosis has been facilitated by enhanced public awareness and increased knowledge of the disorders among health-care professionals. Key among these professionals can be family physicians. (Because tertiary prevention does not usually fall within the scope of the family physician's normal activities, it will not be discussed in this paper.)

Risk Factors for Anorexia Nervosa and Bulimia Eating disorders (EDs) are viewed as the final expression of multiple predisposing, precipitating and perpetuating factors. 1 Prevention-minded physicians niust be aware of multiple psychological, familial, cultural and biological factors that contribute to the development of these disorders. Common symptoms of anorexia nervosa and bulimia are summarized in Tables 1 and 2, respectively. For a discussion of symptomatology, the reader is referred to Leichner.

Psychological risk factors Generally speaking, anorexics who limit their food intake (restrictors) appear to differ psychologically from anorexics and normal-weight individuals who show predominant bulimic symptoms. 12 13 While these differences have been identified retrospectively or cross-sectionally (i.e., in individuals who are already 'ill', it is probable that they reflect significant differences in pre-morbid personality organization. Certain personality features may thus serve as risk factors, and certainly can indicate the possibility of an ED. Restricting anorexics present themselves as obsessive-compulsive. Just

Table 1 Common Symptoms of Anorexia Nervosa a Relentless pursuit of thinness * Refusal to maintain normal weight * Intense fear of becoming obese * Dissatisfaction with body shape * Restriction of food intake * Unusual eating behaviour * Excessive exercising and/or bulimia, vomiting, laxative abuse 146

as they rigidly control their food intake, so they are preoccupied with issues of control and perfection in other aspects of their life. The histories of these patients often reveal a tendency towards overly conscientious behaviour and meticulous attention to work and studies. They tend to sacrifice themselves in their drive to please others. They often experience conflict about sexuality. Parents and teachers view them as well-behaved, achievement-oriented, "model children", but underlying their good manners there is often a terrifying sense of powerlessness to meet high expectations, and fears of losing control and failing dismally. In contrast to anorexics, bulimics, who lose so much control over their diet that they binge and purge, often show a more uncontrolled psychological profile, marked by 'borderline' personality characteristics. Such patients often have chaotic histories characterized by poor social, vocational or academic functioning. Overtly self-destructive behaviour is common: suicide attempts, acts of self-mutilation, sexual promiscuity, substance abuse and illegal acts. Bulimics suffer from unstable moods, and often experience uncontrolled rages or plunge rapidly into despair. While the hypothetical cases described above are over-simplified caricatures, they serve as reminders that it is not simply the over-achieving, 15-year-old, 'straight-A' school girl who is subject to EDS. Physicians encountering either over- or undercontrolled personality types should expand the scope of their usual diagnostic interview, to explore fully the patient's psychosocial system and eating/weight-control patterns. These investigations may reveal a hidden eating disorder or features associated with an incipient ED. In either case, appropriate information about diet, weight management, eating disorders and treatments may "head off" serious problems. In extreme cases of over- or under-control, the individ-

ual's distress level may warrant a referral into psychotherapy, even in the absence of an ED. Although depression has been strongly associated with both anorexia nervosa and bulimia, the specific relationship is not clear.14 EDS frequently occur, however, in the context of a depression. Thus a depressive picture, especially one of chronic dissatisfaction and dysphoria, can itself be considered a psychological risk factor for an eating disorder. On a related note, physicians should remember that weight loss and decreased food intake in a depressed patient may be symptoms of a primary ED, rather than symptoms of the depression itself.

Familial riskfactors Impressionistic evidence suggests that two distinct, pathological, family organizations exist: one predisposes toward anorexia nervosa and the other, toward bulimia. Since these observations have received mixed empirical support,'5' 16 we shall present them only briefly, with the caution that they require further validation. Minuchin et al. 17 describe anorexics' families as superficially 'good' but covertly enmeshed, overprotective, rigid and unable to manage conflict. Similarly, Bruch"8 reports apparently high-functioning (usually middle-class) families, in which parents assume excessive responsibility for the family's welfare and growth. This, she believes, robs the child of a sense of personal worth and effectiveness, lowers self-esteem and forces her into a reluctant state of dependency. Because these families represent many cultural ideals (e.g., the parents are concerned, hard-working and successful), physicians should be careful not to be blinded to internal conflict, of which no family member is able to speak. Families whose members show bulimic traits are thought to be more chaotic and further removed from the " middle-class ideal". They have been found to be disorganized, lacking in cohesion, restrictive of the Table 2 open expression of feelings and inCommon Symptoms of Bulimia clined to de-emphasize intellectual and recreational achievements. 16 Recurrent episodes of binge eating Clinical observation suggests that they Self-induced vomiting, laxative are often families in which emotional, abuse, or dieting after binges physical sometimes, sexual Depressed mood following binges abuse has and, occurred. The important CAN. FAM. PHYSICIAN Vol. 33: JANUARY 1987

point to remember is that eating disorders do occur outside the context of the stereotyped family with high social standing and high achievement. Cultural Risk factors While the average female has increased her body weight over the past 20 years, our female cultural ideal, as represented in pin-ups and fashion magazines, has become slimmer.19 This paradox has set the scene for a huge weight-reduction industry. The glorification of a youthful, often unachievably slim body, virtually 'prescribes' anorexia nervosa to at-risk individuals. Our culture also places new demands on women to abandon traditional roles for career pursuits and to participate in high profile activities. Many women experience conflict, feeling guilt ridden about their more traditional strivings and uncertain about their capacity to succeed in new roles. Rapid changes challenge adaptive mechanisms, so that many young women may begin dieting, as part of the quest to become the "ideal women", and in a misguided attempt to bolster a failing sense of control over their own lives. Evidence shows that eating disorders are spreading into new demographic, socio-economi-c and geographical groups.6 It is no longer possible, therefore, to view middleor upper-class status as a specific risk factor. Similarly, up to 10% of individuals with eating disorders are males,20 and some are middle-aged persons of either sex.

Biological risk factors Major biological risk factors for eating disorders are listed in Table 3. We will not discuss these factors fully here because their specific relationship to eating disorders remains unclear. We do, however, call them to the reader's attention, as they are important components of the complete biopsychosocial evaluation. Precipitating factors The onset of most cases of eating disorder can be linked to a precipitating event. Most often this event is a threat to the patient's self-esteem: a separation from friends or family, entering a new school, changing jobs, or

experiencing

a

debilitating physical

illness. This threat leads the patient to

CAN. FAM. PHYSICIAN Vol. 33: JANUARY 1987

try to regain some control and selfesteem through an improved, slimmer body image: hence the dieting. At first, patients are often unaware of the relationship between the precipitating event and the onset of their illness.

nervosa and bulimia is Insulin-Dependent Diabetes Mellitus (IDDM) .22 It is unclear whether this association arises because of the diet-consciousness required of IDDM patients, or because of the direct effects of IDDM on appetiteregulating functions. Regardless of the reason, counselling patients, who Prevention from the attend diabetes clinics about eating Family Physician's Office disorders and paying special attention Primary Prevention to warning signs seem appropriate Systematic evaluation of the effec- measures. tiveness of primary prevention programs for eating disorders is as yet Secondary Prevention non-existent. We can, however, idenHere, we shall concentrate on the tify several areas in which the family early detection of eating disorders. physician may, by educating the indi- We have already discussed risk facviduals at risk, perform a primary pre- tors to which the physician should be vention function. We encourage attuned. (See Table 3.) Physicians can readers to begin to study the effective- go beyond being alert for individuals in whom various combinations of ness of such efforts. The first, and probably the major, these factors are present. Several easmeans of prevention is related to pre- ily administered questionnaires can be scribing diets. In our clinic we often useful in identifying actual and incipisee anorexics whose disorder began ent cases of eating disorders. Probafollowing a recommendation that they bly the most useful questionnaire for diet. The advisor has sometimes been early detection is the "Eating Disa physician. This finding highlights orders Inventory. "23 Early detection also depenids on the the fact that among young people and other groups at risk, dieting can be a physician's preparedness to consider hazardous treatment. It should be pre- an ED as an explanation of physical scribed conservatively, with careful findings determined from routine exattention to the patient's biopsychoso- amination or investigations into special setting and to the goals and moti- cific complaints. Secretive dieting is vations for weight loss. We recom- often revealed by delayed menarche mend that all dieters should be or secondary amenorrhea. Other informed about eating disorders so gynecological problems, notably unthat they can use their diet like a med- explained infertility or complicated ication: as it was prescribed with full pregnancies, may also occur in women who are tampering with their awareness of the risks. Eating disturbances are unusually food intake. Non-specific gastrointestinal comcommon during pregnancy and after childbirth. Binge eating and other ab- plaints, such as constipation, bloating normal eating patterns, for example, or stomach pain, are often associated have been reported by 20% of women with anorexics' and bulimics' abnorattending a family-planning clinic.21 mal eating patterns. In a related vein, Therefore, pregnancy is another the physician should be wary of reperiod when primary prevention may quests for laxatives and other purgabe possible. Body changes and food tives by patients presenting symptoms cravings during pregnancy, changes that masquerade as obscure gastro-inin the marital relationship, and con- testinal complaints. Ankle edema and cerns about personal identity aroused other edemas may be the result of by the sudden transition to mother- water retention associated with frehood may precipitate an eating dis- quent binging and vomiting. In a superficial oral-dental exam, order in the predisposed woman. Preventive counselling should thus the physician may detect signs that beinclude attention to eating disorders tray hidden episodes of vomiting. Freand proper diet, and should prepare quently, dental sensitivity is associated women at risk to deal with new con- with erosion of dental enamel by stomach acid.24 Sometimes there is enflicts. Another physical condition that ap- largement of the parotid glands. Abpears to be associated with anorexia normal laboratory results may also 147

raise suspicion of an ED.25 Leukopenia, anemia and low fasting-glucose levels may be caused by restricted food intake. Among bulimics, hypokalemia and hypochloremia may be found. In both restricting and binging individuals, relative lymphocytosis and elevated serum alkaline phosphatase and/or amylase may be present. Over 50% of women who request cosmetic breast reductions have been shown to have symptoms of anorexia nervosa and/or bulimia.26 Thus, requests for elective plastic surgery, and the presence of dysmorphophobias of any kind should signal the need for a thorough and frank evaluation of eating pattems and weight concerns. In an interview, one undiagnosed bulimic woman's chief complaint was'"I don't like my droopy eyelids". She had twice sought cosmetic surgery, and had undergone a course of psychotherapy. All the while her bulimia had escaped detection. Individuals in certain professions appear to be particularly predisposed

to eating disorders. Watch out for EDS when they leamed that laxatives proamong women in the performing arts, duce negligible reduction in calorie ab-

fashion or ballet worlds.' Similarly, both male and female athletes may be subject to increased risk of EDS. It is not known whether these professions encourage diet consciousness in their members by placing a premium on attractiveness and conditioning, or whether they attract persons who are predisposed to EDs. In either event, the prevention-oriented physician shoul4 be aware that EDs occur among patients in such professions, much as they are conscious now of the prevalence of heart disease among highlevel executives. Suggestions about a general diagnostic approach to EDs are outlined in Table 4. An elaboration of these points is available elsewhere. 12 Because studies on the effects of psycho-educational approaches to EDs have only recently begun, available data are limited. One recent study, however, made an encouraging finding: many bulimic women in treatment abruptly discontinued laxative abuse

Table 3 Risk Factors Associated With Eating Disorders Restricting (overcontrolled) type Binging (undercontrolled) type Psychological Borderline personality traits Obsessive-compulsive personality traits * Perfectionism * Poor social and vocational * Control issues adjustment * Excessively achievement-oriented * Unstable relationship * Fear of age-appropriate sexuality * Self-destructive ideation * Depressive tendencies and behaviour Unstable mood . Depressive tendencies Fam ilial * Chaotic * Low in cohesion . Limited expression of feelings. * Low emphasis on achievement * Emotional and physical abuse *

Enmeshed Overprotective Rigid * Conflict avoiding * Superficially well organized * Achievement-oriented *

*

Family cconcerns about weight, appearance Cultural * Emphasis on slenderness * Changing role of women . 'Discrimination' against fatness Biological History of weight problems . Being female * Family history of alcoholism, affective disorder, obesity and/or eating disorder *

148

sorption.27 This finding, we believe, a tremendous potential for psycho-education in the treatment and prevention of EDS. A variety of psy-

indicates

cho-educational materials is available from the Anorexia Nervosa and Bulimia Foundation of Canada.28 Is prevention feasible? During our work with individuals suffering from anorexia nervosa and bulimia, we have had many encouraging experiences, not unlike those of the many other health-care profes-

Table 4 Overall Diagnostic Approach* The Setting Should be calm and private. The Intervening Approach Be open, honest, frank and non-judgemental. Clues in the History Look for a precipitating event prior to the onset of the illness. Obtain a detailed account of the patient's current attitude toward food, as well as her eating behaviour. Look for a family history of depression, alcoholism, eating disorders, and an understanding of the family dynamics. The Physical Must be complete and thorough. Look for complications. Tests to Order Do complete blood tests, electrolytes, liver and kidney function studies, as well as an endocrinologic work-up. Electrocardiograms. The Eating Attitude Test or Eating Disorders Inventory. Final Steps Give the patient an open, frank, professional opinion of the situation. Provide educational materials about eating disorders. Outline a treatment plan. Allow the patient to consider his or her options. Provide follow-up for patients who do not agree with the treatment plan. * Reprinted with permission.12 CAN. FAM. PHYSICIAN Vol. 33: JANUARY 1987

sionals who have reported favourable' response to treatment of ED patients., These results, however, do not permit us to ignore many patients, usually those seen when their ED has already been present for many years, who remain refractory to treatment, or who struggle with relapses and milder forms of their disorder even after intensive treatment. This experience highlights the fact that as with many disorders, a strong prognostic indicator for EDs seems to be chronicity. In cases of this sort, physicians have a tremendous role to play. Ordinarily, it is to family physicians that individuals who have developed, or who are at risk for, an ED often first present. Patients may present with medi'cal complications caused by their eating habits, with an associated emotional problem, or because of pressure from concerned family members. Physicians are thus in a position to act before the illness has become chronic and devastating to the physical and psychological health of its victim. We reiterate the point that EDs are multidimensionally determined: psychological and biological factors, familial patterns, social pressures and precipitating life events can finally express themselves as an ED. We firmly believe, therefore, that physicians who are aware of these factors can have substantial influence on the development and course of these disorders. A major therapeutic (and preventive) tool will be the non-judgemental provision of realistic and frank information to individuals at risk, or who already have an ED. Often, we see early improvements in patients' behaviour and mood after we routinely provide them with educational materials during the first encounter. We are convinced that when a developing or existing ED is the problem, a little knowledge is not a dangerous thing. Physicians should certainly be cautious in their approach to patients seeking diets, or otherwise expressing dissatisfaction with their appearance and with themselves. It is probably not overly interpretive to view an eating disorder as, partly, a misguided desire to change unacceptable parts of one's self and one's life. Since ED patients seek to keep control of their lives through such change, it is important to approach them from a non-controlling stance. By offering patients realistic information about eating disorders, it is possible to hand them the means to CAN. FAM. PHYSICIAN Vol. 33: JANUARY 1987

make vital choices for themselves. Hence, they regain the control they wish for. Similarly, it is essential to help patients to identify the areas of their lives which 'beg' for change, be it in intrapsychic, familial or social spheres, and to intervene appropriately before maladaptive solutions such as dieting and binging are adopted. Without intervention, these solutions could become autonomous, entrenched symptoms, distruptive to even the healthy aspects of these patients' lives. At this point, we hope that you have formulated an answer to the question posed at the end of the opening vignette. Would you prescribe a diet? Your answer should be: "Not until I ( have much more information."

References 1. Garfinkel PE, Garner DM. Anorexia nervosa: a multidimensional perspective.

New York: Brunner-Mazel, 1982. 2. Abraham SF, Mira M, Beumont PJV, Sowerbutts TD, Jones DL. Eating behaviours among young women. Med J Australia 1983; Sept:225-8. 3. Button EJ, Whitehouse A. Subclinical anorexia nervosa. Psvchol Med 1981: 2:509-16. 4. Pyle RL, Mitchell JE, Eckert ED, Halvorsan PA, Newman PA, Goff GM. The incidence of bulimia in freshman college students. Int J Eating Disorders 1983; 2(3):75 -85. 5. Halmi KA, Falk JR, Schwartz E. Binge eating and vomiting: a survey of a college population. Psychol Med 1981; 11:697 -706. 6. Leichner P, Arnett J, Parro J. An epidemiology study of maladaptive eating attitudes in a Canadian school age population. Int J Eating Disorders 1986; 5(2):969-82. 7. Schwartz DM, Thompson MG, Johnson CL. Anorexia nervosa and bulimia: the socio-cultural context. Int J Eating Disorders 1982; 1(3):20-36. 8. Caplan G. Principles of preventive psychiatry. New York: Basic Books, 1964. 9. Powers PS, Fernandez RC. Current treatment of anorexia nervosa and bulimia. Karger 1984. 10. Vandereycken W, Meermann R. Anorexia nervosa: a clinician's guide to treatment. Hawthorne, NY: Walter de Gruyter 1984. 11. Leichner P. Detecting anorexia nervosa and bulimia. Diagnosis 1985; 2( 1):3 1-46. 12. Garner D, Garfinkel P, O'Shaughnessy M. Clinical and psychometric comparison between bulimia in anorexia nervosa and bulimia in normal weight

women, in understanding anorexia nervosa and bulimia. Report of the 4th Ross Conference on Medical Research. Ross Laboratories, Columbus, Ohio. 1983: 6-11. 13. Beumont P, George G, Smart D, "Dieters" and "vomiters and purgers" in anorexia nervosa. Psychol Med 1976; 6:617-22. 14. Swift WJ, Andrews D, Barklage NE. The relationship between affective disorder and eating disorders: a review of the literature. Am J Psychiatry 1986; 143(3):290-9. 15. Harding TP, Lachenmeyer JR. Family interaction patterns and locus of control as predictors of the presence and severity of anorexia nervosa. J Clin Psychol. 1986; 42:440-448; 16. Johnson C, Flach A. Family characteristics of 105 patients with bulimia. Am JPsychiat 1985; 142:1321-4. 17. Minuchin S, Rosman BL, Baker L. Psychosomatic families: anorexia nervosa in context. Cambridge, MA: Harvard University Press 1978. 18. Bruch H. The golden cage. The enigma of anorexia nervosa. Cambridge, MA: Harvard University Press, 1978. 19. Gamer DM, Garfinkel PE, Schwartz D, Thompson M. Cultural expectations of thinness in women. Psychol Rep 1980; 47:483-91. 20. Andersen AE, Mickalide AD. Anorexia nervosa in the male: an underdiagnosed disorder. Psychosomatics 1983; 24(12):1066-75. 21. Cooper PJ, Waterman GC, Fairburn CG. Women with eating problems: a community survey. Br J Clin Psychol 1984; 24:45-52. 22. Rodin GM, Johnson LE, Garfinkel PE, Daneman D, Kenshole AB. Eating disorders in female adolescents with insulin dependent diabetes mellitus. Int J Psychiat Med 1986-87; 16(1):49-57. 23. Garner DM, Olmsted MP, Polivy J. Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia. Int J Eating Disorders 1983; 2(2):15-34. 24. Peterson DS, Barkmeier WW. Oral signs of frequent vomiting in anorexia. Am Fam Physician 1983; 27:199-202. 25. Halmi KA, Falk JR, Common physiologic changes in anorexia nervosa. Int J Eating Disorders 1981; 1(1):16-27. 26. Mester H. Women desiring a change in the size of their breasts: a contribution to the issue of dysmorphophobia. Zeitschrift fur Psychosomatische Medizin 1982; 28:69-91. 27. Emmet SW, ed. Theory and treatment of anorexia nervosa and bulimia: biomedical, sociocultural and psychological perspectives. New York: Brunner-Mazel 1985:309. 28. Anorexia Nervosa and Bulimia Foundation of Canada, P.O. Box 3074, Winnipeg, Man. R3C 4E5. 149