Dieting, body weight, body image and self-esteem in young women: doctors' dilemmas

What is normal eating and exercise behaviour in young adolescents and what are normal adolescent thoughts about eating and body shape and weight? Questions like this were raised in the Journal 20 years ago,1 and some remain unanswered. In this article, I examine our current knowledge and discuss some of these unanswered questions.

Adolescent girls often comment that they are "dieting" but what they mean by this differs for premenarchial and postmenarchial girls.2 Before menarche, "dieting" is described by the girls as "normal healthy eating"; only after menarche does "dieting" include the concept of weight loss. Menarche is followed by an increase in body weight and body mass index (BMI). Many young women try to lose weight by restricting their food intake, and exercising. A few experience binge eating — episodes they feel are outside their control and during which they eat large amounts of food — and a few resort to potentially dangerous methods of weight control, such as smoking cigarettes, purging and using social drugs.

Some young Australian women also express beliefs and feelings that are thought to be indicative of the presence of an eating disorder. Standardised measures of eating and exercise behaviour can be provided in both clinical and community settings by the use of a tool known as the Eating and Exercise Examination by Computer (EEE-C)3,4 — a computerised questionnaire which gathers data on thoughts, feelings and behaviour related to eating and exercise. A recent unpublished survey of students from two secondary schools (one working and middle class inner city school and one suburban private girls' school), as well as tertiary students and patients with eating disorders, showed that many young women, both with and without diagnosed eating disorders, feel that body image (weight and shape) and exercise are important for their self-esteem, want to lose weight, are afraid they might gain weight and become fatter, and feel fat (Box 1).

Four large school-based intervention studies have looked at this issue.15-18 One program aimed at improving students' self-esteem, and reducing the focus on weight, food and dieting, appeared to help,18 although such programs would need reinforcement at least every 12 months (randomised, controlled trial). The other studies aimed at providing information and resulted in little or no change in attitudes.19,20 Intervention programs can be harmful, as they can glamorise eating disorders and unintentionally provide negative feedback to students about body image. Even trained teachers may unwittingly be poor role models.21

Improving the self-esteem of school students, encouraging communication, and helping students to be supportive of each other may prevent some young women from developing eating disorders. General practitioners can be helpful in educating school and community groups involved in adolescent care. In clinical practice, GPs need to be aware of the possible risk factors, triggers and perpetuating factors for eating disorders (Box 2).

Does a postpubertal adolescent woman who has been losing weight (in negative energy balance) for 12 months, who has no medical reason for her weight loss and is not particularly interested in her body weight and shape, have an eating disorder? She could be a medium-distance runner who has increased her training to improve her performance (current BMI, 17.5) or she may have changed to "healthy" (low fat) eating and exercising to be fit and healthy for her final-year school exams (current BMI, 16). If we apply the current criteria, set out in the Diagnostic and statistical manual of mental disorders, fourth edition (DSM-IV)22 and shown in Box 3, neither of these women have anorexia nervosa, as their body weight and shape is not important to them.

If these women were unable to achieve a positive energy balance by eating more and exercising less over the next month, even after a consultation with a dietitian experienced in working with eating and exercise problems, an eating disorder should be considered. Many women are not aware they have an eating or exercise disorder until they are challenged and realise they are unable to change their eating and exercise behaviours and thinking. They may find it extremely difficult to trust advice or believe that change is necessary. In these two cases it is unlikely the women would believe anyone who told them that they had an eating disorder, particularly when they are informed that anorexia nervosa is associated with a desire for thinness and an overconcern with body shape and weight. These women are more likely to describe a fear of loss of control over their body, eating or exercise (Box 1).

Soon after beginning to lose weight, these two women may find their physical and academic performance improves. This will reinforce their behaviour and make any discussion about its negative effects appear irrelevant. After losing more weight, finding they are preoccupied with thoughts of food that interfere with reading and studying, and that their behaviour and feelings are affecting their daily activities, relationships and health, an eating disorder is more likely to be diagnosed, and the need for change more readily accepted by the women.

Additional problems with the current diagnostic criteria for the eating disorders relate to amenorrhoea and body weight measures.

One of the first physical signs of losing weight or being in negative energy balance can be secondary amenorrhoea; Patients who are taking oral contraceptive pills and experiencing a withdrawal bleed each month often think they have a "normal" menstrual cycle; and may erroneously report "normal" cycles to health professionals.

Amenorrhoea can occur at a high, normal or low body weight or BMI. It occurs more readily in younger teenage women and over a shorter time if their energy balance is negative.10 It should not be considered an accurate measure of recovery, as age, exercise and time spent in positive energy balance may be interacting factors, particularly in young people. Secondary amenorrhoea can be a useful indicator, but not as a criterion for diagnosis of, or recovery from, an eating disorder.

A diagnosis of anorexia nervosa is usually reserved for people who attain and maintain low body weight, less than BMI 17.5 or 15% less than expected or ideal. BMI is a better measure of body fatness than body weight,23 but there are no "normal" charts for girls at different stages of pubertal development or for women from different cultures. In reality, women who have lost more than 10 kg can show all the features of anorexia nervosa — secondary amenorrhoea, fear of loss of control over their bodies and a preoccupation with weight and eating associated with a fear of returning to obesity. Waiting until a person achieves a certain weight before the need for treatment is recognised is negligent.

It can be seen that certain questions in Box 1 — fearing loss of control over eating, and being preoccupied with weight, shape and eating — discriminate better than other questions between students and patients with eating disorders. If an eating disorder is suspected, it may be useful for physicians to ask young people questions about fear of loss of control over their bodies, eating, weight and shape as well as preoccupation with food, eating, nutrition, body weight and shape, as these may differentiate those at greater risk.

Box 4 lists simple clinical criteria for anorexia nervosa and anorexia-like eating disorders, and bulimia nervosa and bulimia-like eating disorders. I propose these criteria for use by family doctors working with young women, to identify those who may need physical assessment, help with their eating and exercise behaviour, and psychological support. For women identified by these criteria, GPs need to decide on and coordinate appropriate multidisciplinary assessment and treatment; in severe cases, referral to inpatient treatment at a specialised eating disorder unit may be necessary.