Panel Examines Risk for Eating Disorders Throughout the Life Cycle

2004 International Conference on Eating Disorders

During a plenary session at the 2004 International Conference on Eating Disorders, in Orlando, four eating disorders experts explored risk factors for development of eating disorders throughout the lifespan.

Dr. Leann Birch, Distinguished Professor at The Pennsylvania State University, University Park, PA, described individual and familial risk factors identified in an ongoing longitudinal study. The study has followed 197 two-parent families and their 5-year-old daughters; the girls are now 13 and will be followed until they are 15 years old.

The family environment is particularly important for kids, Dr. Birch said, noting that parents often determine control of the eating environment by controlling portion sizes and second helpings. After this, media exposure may lead children to seek out certain foods. Parents’ attempts to restrict certain foods usually backfire.

“Restricting access to snack foods high in fat and sugar leads to enhanced preferences for those foods, increased attention to those foods and, when the foods are present, increased intake of those foods,” said Dr. Birch. Negative self-evaluation about eating those restricted foods has been reported among preschoolers as young as 3 to 5 years of age.

Feeding practices also play a role in the development of unhealthy eating habits—for example, a child’s weight might be interpreted by parents in ways that influence their regular feeding practices, particularly restricting foods, and this could have an impact on a child’s eating pattern. The parents’ own eating patterns and weight play a role both in terms of the environment they are providing for kids and in terms of the genetics they bring to the situation, she added. According to Dr. Birch, development of dieting, overeating, and girls’ self-evaluations are influenced by early weight status because, at least in middle-class America, weight tends to trigger certain kinds of parenting behavior, particularly restricted feeding practices.

Eating in the Absence of Hunger

Dr. Birch told the audience that one behvavioral measure of disordered eating among young girls is eating in the absence of hunger. This is really a response to the presence of palatable food, she said; in addition, it captures some characteristics of binge eating. Some kids consume relatively large amounts of food in a short period, and during debriefing sessions they report feeling that their eating is somewhat out of control, similar to binge eating. This is promoted by restrictions imposed by others, not self-imposed restriction.

Girls who are overweight and have mothers who use a lot of restriction show the greatest increases in eating in the absence of hunger, said Dr. Birch. She added that even at age 5, overweight status increases the risk for maladaptive eating attitudes, including eating in the absence of hunger, which the researchers view as a possible precursor of binge eating.

Maternal Weight

Among normal-weight mothers, maternal restriction is not related to daughters’ eating in the absence of hunger and is not related to change in body mass index (BMI). Dr. Birch noted that about 65% of the mothers in her study are overweight, and the researchers have seen that maternal restriction in this group is quite strongly related to eating in the absence of hunger over time and to daughters’ increases in BMI.

Dr. Birch said that among middle class white girls, early overweight increases risk for at least some sorts of maladaptive eating. In her study, there was greater eating in the absence of hunger, negative self-evaluations of overeating, elevated weight concerns, elevated body dissatisfaction, early dietary restraint, and greater weight gain across middle childhood among those who were overweight at an early age.

One area that needs more study, according to Dr. Birch, is determining which factors are mediating risks conferred by early overweight. Dr. Birch concluded, “Overweight parents create eating and activity environments that are quite different from those of normal-weight parents. If we understood this better we’d have some good ideas about prevention.”

Risk Factors for Eating Disorders

Studying risk factors for eating disorders helps clarify diagnoses and classifications, and also helps direct treatment, according to Ruth Striegel-Moore, PhD, Professor and Chair of Psychology at Wesleyan University, Middletown, CT.

She reported initial findings from the National Growth and Health Study, Wave II, sponsored by the National Institute of Mental Health and the National Institute of Diabetes, Digestive and Kidney Diseases. The goal of the 1-year study is to identify eating disorders early by identifying high-risk populations.

One important question was whether ethnicity was a marker for an eating disorder. In the study, white women were significantly more likely than black women to meet diagnostic criteria for an eating disorder. There was also a significant difference in the number of women who met criteria for bulimia nervosa—4 black women and 23 white women. Less marked ethnic differences were reported for binge eating disorder (BED). However, there were twice as many white women as black women who met diagnostic criteria for BED. Vomiting was much more common among white women, and very uncommon among black women, but no difference was seen in laxative abuse between white and black women. Dr. Striegel-Moore also reported that white women have an earlier onset compared to black women.

Three High-Risk Groups Identified

The researchers identified three high-risk groups: (1) those who had high weight concerns before age 14; (2) those who were less concerned about weight than the first group but who reported high perceived stress; and (3) a subgroup whose weight concerns and perceived stress were lower than those among group 1 or 2, but who reported behavior conduct problems before age 14.

Research on risk factors so far has focused primarily on the question of whether a given variable is a risk factor for an eating disorder. Dr. Striegel-Moore emphasized that her group has attempted to show that for different subgroups different variables or different combinations of variables may increase risk. She also added a word of caution about matching treatment based on a risk profile, pointing out that treatment matching, particularly in substance abuse and alcoholism, has been spectacularly unsuccessful.

Finally, she said, “Our results underscore that we need to be very careful regarding our messages on what factors contribute to risk. When we design prevention programs or public education campaigns, simple messages may make for compelling sound bytes but they don’t necessarily make for compelling truths. Our ultimate goal is to reduce suffering due to an eating disorder. We will succeed if we resist the urge to provide single-factor answers and pursue treatment interventions that keep in mind the complexity of the etiology of eating disorders.”

Psychosocial and Genetic Risk Factors

Andreas Karwautz, MD, Professor at the University Clinic of Neuropsychiatry, Vienna, Austria, described the value of the concept of non-shared environment for development of anorexia nervosa (AN) and the opportunities to use a discordant sister-pair design for clarifying risk for developing anorexia nervosa.

Discordant sister-pairs have several advantages for study over nonrelated case-control studies. These include similar socioeconomic status, living region, family structure, parental socio-academic status, and religious orientation of parents, for example. He added that this structure allows researchers to replicate and investigate found associations in between family studies using a within-family design. Genetic markers can also be included in the same samples and then researchers can focus at the end on gene environment interaction to help clarify the etiology and pathogenesis of these disorders.

Dr. Karwautz shared recent unpublished data on 120 sister-pairs discordant for anorexia nervosa studied in Vienna, London, and Barcelona. Fifty-five of the pairs had restricting type AN and 70 had binge-purge type AN. The two groups were similar in age and the age of onset of AN was around 16 years of age.

Among the 58 sister-pairs with restricting AN, researchers found that personal environment was significant and enhances the risk of developing AN. Some common features in this group included negative self-evaluation, perfectionism, no male friends, parental control, rivalry with the unaffected sister, and a need to compete with the sister’s appearance and shape. Minor but still significant factors included shyness and premorbid anxiety disorders.

Among the 62 sister-pairs with binge-purge type AN, personal environment and dieting contributed to the development of their disorder. Girls with this subtype of AN had a number of the same characteristics as those with the restricting type of AN, but were also distressed by parental arguments, and life events the year before their illness developed. Two main items emerged in the dieting domain: repeated critical comments by family members about weight, shape, and eating, and teasing about shape and weight and appearance,

Vulnerability factors have a significant influence on the development of AN of both subtypes, while dieting vulnerability factors contributed to the development of binge-purge type AN. Data from the multicenter study adds new information about personal vulnerability in the two AN subtypes. Dr. Karwautz reported that personal vulnerability (for example, negative self-evaluation and perfectionism) is highly relevant to restricting AN. Internalizing behavior problems in childhood influence the development of restricting AN, and internalizing and externalizing problems contribute to the development of binge-purge AN.

The next step, said Dr. Karwautz, will be handling candidate gene data together with psychological and psychosocial risk factors in order to develop a gene environment interaction model of development of AN.

Older Adults with Eating Disorders

“We don’t think much about eating disorders in older women, as illustrated by the fact that prevalence rates are often calculated and presented only for young women,” Marika Tiggemann, PHD, Professor of Psychology at Flinders University, Adelaide, Australia, told the plenary session audience. Eating disorders can be a chronic condition throughout life, she said.

Dr. Tiggemann added that body dissatisfaction is one of the few robust and consistent risk factors that have been identified among adult women with eating disorders. She noted that it is reasonable to expect that body image would become more negative as women age—every year moves women further from the thin and youthful ideal

Lifelong Dissatisfaction

Women seem to be more dissatisfied or negative about their bodies at all ages, she said. In fact, she said, body dissatisfaction seems to be remarkably stable across the entire female lifespan. Women aged 30 to 75 years show substantial levels of body dissatisfaction, just like younger women. Between 55% and 95% of women express dissatisfaction with their bodies, and chronic dieting may pose a particular health risk for women as they age.

Risk factors may have differing degrees of influence across the life span. An area that receives too little attention, according to Dr. Tiggemann, includes the biological developmental milestones in a woman’s life. All women have the potential to increase fat deposition through the operation of sex hormones, and this physiological factor moves a women further from the thin and youthful ideal. The effects of menopause on body image haven’t been studied very much, she said. At menopause, weight typically becomes redistributed and women’s shapes change, so they have larger waists, becoming rounder in shape. Pregnancy also brings body dissatisfaction during the postpartum period.

Need for a Better Definition of Body Image

Dr. Tiggemann told the audience that there is a need for a broader definition of body image. When talking about younger women and the predictors of disordered eating, it makes sense to focus on body size and weight; but for older women other aspects might come into play. She mentioned a study in 1996 when older women were asked to select the most attractive physical feature of an older person. Their answer was, “posture.” This demonstrates that there may be many other aspects of body image among older women that are missed because of concentrating on young women and generalizing from what we know about younger women, she added.