Eating Disorders Review
July/August 2010
Volume 21, Issue 4

The Research-Practice Gap:
An Enduring Challenge for the Eating Disorders Field

Dr. Timothy Baker and colleagues caused a furor in the field of clinical psychology with the publication of their monograph on the quality of clinical psychology practice in the U.S. (Psychological Science in the Public Interest, 1990; 9: 67-103).

Baker et al. (2009) state that the practice of clinical psychology is substandard because clinical psychologists demonstrate "insouciance to science" and "disregard of scientific evidence." The authors support this premise by citing surveys revealing that clinical psychologists base their treatment decisions more on "craft" or clinical experience than on scientific research, and point to the low utilization rates of empirically-supported treatments (ESTs) in clinical practice. To remedy this situation, the authors call for a new clinical psychology accreditation system that focuses on an emphasis on training in scientific thought and methods and evidence-based treatments. The article triggered a storm of media attention and professional debate centering on the ethics of practicing clinicians.

Interestingly, the thrust of the Baker et al. (2009) article and the ensuing finger-pointing provide a vivid enactment of the dynamics that fuel the research-practice gap in the field of eating disorders, and in most fields with both applied and basic science components. Baker et al. (2009) are concerned that important research findings are not making their way into clinical practice. We agree. However, we also are concerned that valuable clinical observations and experience are not impacting current research directions. Further, we submit that a top-down approach to the research-practice gap is not a viable solution. Instead, we propose that fostering communication, collaboration and partnership between researchers and practitioners will produce substantive, longstanding improvements in the quality of not only clinical practice but also in the quality of research.

This article briefly reviews evidence regarding the research-practice gap in the field of eating disorders and summarizes possible explanations for the gap from both research and practice perspectives. It also outlines practical strategies clinicians can use to integrate research within their practice, thus enhancing the quality of their clinical work and informing new research directions. (Please note: We are aware that many professionals in our field work in both research and in clinical practice. For the purposes of our discussion, an exaggerated distinction between the two groups is made.)

Evidence for a Research-Practice Divide in the Eating Disorders Field

Our review of the evidence for the research-practice gap is drawn from the work done in this area by the Academy for Eating Disorders (AED), a global multidisciplinary professional association. The AED has conducted informal surveys, focus group discussions, and interviews of its members to gain a clearer picture of the factors perpetuating the research-practice gap. The responses generated by these initiatives suggest that relational, systemic, and attitudinal factors play key roles in generating and perpetuating the research-practice divide.

Relational Factors

A lack of shared respect for the mutual contributions of research and practice to the field, emerged as a significant theme. Responses from the surveys, focus groups, and interviews revealed that clinicians clearly felt "under attack," including references to "a faith-based assault against clinicians" and the view that they are perceived as "unscientific charlatans." This perceived lack of respect may stem from the increased focus on the emerging primacy of empirical research over the last decade, contributing to a perceived power differential between clinicians and researchers and a tendency for researchers to be more commonly at the podium espousing the need for clinicians to change (rather than the reverse). Nonetheless, researchers also reported experiencing a lack of respect, noting that their data fail to be translated into practice and that their research findings are dismissed by clinicians who question their relevance.

Attitudinal and Systemic Factors

Perhaps the most salient attitudinal factor contributing to a divide between researchers and practitioners is the difference in their views about what constitutes valid evidence. Clinicians tend to place greater value on evidence born from clinical observation and experience. Conversely, researchers tend to value the evidence derived from structured research trials. This difference in preference of evidence type reinforces the research-practice gap and the accompanying lack of mutual respect between clinicians and researchers. Systemic factors also contribute to research-practice tension, including a lack of shared language and the lack of institutional and economic support for research-practice collaboration. The quantitative, precise language of research and the expressive, process-oriented language of clinical practice have little in common. The relative institutional and funding factors and simple time constraints affecting researchers and practitioners block opportunities for interface between the two groups.

Strategies for Bridging the Gap

Taking this range of factors into consideration, the AED developed the AED Guidelines, and Action Plan for Research-Practice Integration (www.aedweb.org). The principles and strategies put forth in these documents are aimed at institutions and organizations. Building on the directions outlined in the AED initiative, we suggest practical steps individual treatment professionals and treatment facilities can take to begin bridging the gap in their own practice settings.

Step 1: Join research-practice networks.

Increased interaction and dialogue between researchers and practitioners can help develop a more unified language for describing the complex issues related to eating disorders treatment, research, prevention, and education. Propinquity can also foster mutual respect and understanding and establish partnerships to promote translation of practice into research and research into practice. To this end, practitioners can seek out or develop networking groups within their regions, or join professional associations (e.g., the AED), or even online forums (e.g., AED Research-Practice Listserv).

Step 2: Gain access to treatment research.

The complex challenges of eating disorders treatment require clinicians to expand and improve their treatment tools and to stay abreast of key findings. However, for many clinicians, the time it takes to access and read research articles and the language of research is a challenge. The International Journal of Eating Disorders (IJED)provides jargon-free summaries of each article published along with periodic review articles on current topics in research. Eating Disorders Review provides articles that summarize current findings relevant to clinical professionals. In addition, attending conferences, workshops, or meetings (or purchasing their program CDs) where the latest ED research is reviewed can familiarize practitioners with current findings while accruing necessary continuing education credits.

Step 3: Translate practice to research: The benefits of the case study

Disciplined clinical observation has led to important advancements, not the least of which is the earliest description of eating disorders and subsequent identification of different constellations of eating disorders. The individual case study offers clinicians a format and channel for communicating significant observations gleaned from clinical practice. A case study is the systematic evaluation of an individual treatment case or a small number of treatment cases (case series). The narrative can include clinical observations, interviews, past records, and/or measurement tools such as assessment tools or psychological tests. Hypotheses for single case research can be formulated from the careful observations recorded in case studies. In this way, case studies provide clinicians a channel for directly influencing research directions.

Step 4: Incorporate simple research methods and treatment outcome measures into everyday practice.

The eating disorders field's knowledge base lacks vital information about the quality and effectiveness of ED treatment as it is delivered in community settings. The application of simple treatment assessment instruments in clinical practice provides clinicians with valuable feedback about the effectiveness of their own practices and programs and can also provide data for treatment research. Further, the information provided via these instruments can be used to adapt the therapeutic approach to the needs of individual patients.The AED web site (www.aedweb.org) provides a list of measurement tools that can be easily implemented by busy clinicians and treatment teams. Information is also provided about participating in the AED online Data Network, where unused data can be shared or clinician-researcher collaborations can be forged.

Conclusions

The research-practice gap has presented an ongoing obstacle to the enhancement of the knowledge base in the field of eating disorders. Valuable research findings are not integrated into clinical practice, and valuable practice experience and observations do not impact research directions. A respectful consideration of the value added by the expertise and knowledge of both research and practice is our best hope for a research-practice synthesis that can yield the breakthroughs we need to improve the lives of people with eating disorders and their loved ones.

The AED has produced guidelines and an action plan for research-practice integration within organizations and institutions. We have outlined simple strategies the individual practitioner and/or treatment facility can use to promote research-practice integration within their own practice settings. Through the implementation of these steps, clinicians can help to address the relational, attitudinal and systemic challenges that contribute to the research-practice gap and, at the same time, contribute critical clinical knowledge to advance ED treatment, research, education and prevention.

UPDATE: Anticipated DSM-V Changes Will Enhance Eating Disorders Diagnosis

In the much-anticipated revision of the Diagnostic and Statistical Manual of Mental Disorders (DSM), binge eating disorder (BED) is on track to becoming a formal psychiatric diagnosis, according to Dr. Tim Walsh of Columbia University, emeritus EDR Editorial Board member. Dr. Walsh outlined several anticipated changes in the overall eating disorders category in the DSM-V at the American Psychiatric Association meeting in New Orleans at the end of May. According to Dr. Walsh, currently there does not appear to be enough hard evidence to support creating a psychiatric diagnosis for obesity or overeating. However, he noted that there might be evidence for establishing an overeating syndrome that is distinct from BED.

Dr. Walsh also told the meeting participants that the major eating disorders category, including anorexia nervosa (AN), bulimia nervosa (BN), and eating disorders not otherwise specified (EDNOS), had long cried out for an overhaul. While initially three additional categories/diagnoses were proposed to better differentiate types of patients with EDNOS (which includes some 45% of patients treated for eating disorders), the DSM-V work group decided that such changes would be premature. Instead, the additional categories may be included in an appendix to the DSM-V. Dr. Walsh also noted that some changes are expected in the diagnostic criteria for AN and BN. For AN, the first criterion, which is currently "85% of recommended body weight," now will be reworded as, "restriction of energy intake [leading to] markedly low weight." In addition, the much-debated criterion of amenorrhea has been dropped from the list of symptoms associated with AN.

As for BN, diagnosis will be easier than before because of a change in the required frequency of binge-purge episodes, according to Dr. Walsh. Whereas the current DSM criterion for BN specifies that episodes of binge-eating and purging must occur twice a week for the previous 3 months for a diagnosis of BN to be made, the new criterion allows the diagnosis to be made when the binge/purge frequency is once a week.

Working to Better Define Eating Disorders in Youth

Much remains to be defined in eating disorders among children and adolescents. As the Workgroup for Classification of Eating Disorders in Children and Adolescents recently reported, applying current diagnostic criteria to children and adolescents can be problematic (Eur Eat Disord Rev 2010; 18:79). For example, growth and weight gain are expected in children and adolescents, which is a challenge when using strict weight criteria for a diagnosis of anorexia nervosa (AN). Similarly, children and teens may be premenarcheal, making the amenorrhea criterion irrelevant. And, the cognitive eating disorder criteria, including extreme fear of weight gain, body image disturbances, and overvaluation of weight and shape, are difficult to assess and to apply to young patients, who may not have developed insight into motives for eating behavior. Current thresholds for AN and bulimia nervosa (BN) may be overly narrow, and relaxing the criteria would allow a subset of youth with eating disorders not otherwise specified (EDNOS) to be reclassified within broadened forms of AN or BN.

Study targets eating disorder symptoms

Because the current system may not be adequate for categorizing eating pathology in youth, the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) should consider recognizing children and teens with minimal cognitive eating disorder symptoms, according to Kamryn T. Eddy, PhD and colleagues. The authors recently reported the results of their 8-year study of young patients at the University of Chicago Medical Center (J Am Acad Child & Adoles Psychiatry 2010; 49:277).

Dr. Eddy and colleagues noted that the DSM-IV-TR does not make specific provisions for diagnosing eating disorders in children and adolescents and that to inform the new edition of the DSM-V, research is needed to determine whether the same groups of symptoms and presenting diagnoses seen in adults can be applied to youth. The authors designed a study of 401 youths aged 7 through 19 years who sought treatment for eating disorders at the University of Chicago Medical Center between October 2001 and April 2009. A total of 363 girls (90.5%) and 38 boys (9.5%) participated. Most were Caucasian (73.9%); 12.4% were Hispanic; 8.9% were African American; 2.0% were Asian; and 2.8% were identified as "another race/ethnicity."

Before treatment, all patients were assessed with the Eating Disorder Examination (EDE), version 12.0D/C.2, the version adapted for children. DSM-IV-TR diagnoses of AN and BN were assigned, based on EDE-generated behavioral and cognitive symptoms. Eating disorders presentations that did not meet these criteria were categorized as EDNOS. The Beck Depression Inventory and Rosenberg Self-Esteem Scale were also administered before treatment. Then, latent profile analysis (LPA) was applied. LPA proposes that a heterogeneous group can be broken down to a finite number of homogeneous subgroups by minimizing associations among response across multiple indicators. In this way, LPA identifies the number and composition of unobserved latent groups, which themselves are mutually exclusive.

Three groups were identified

Three latent profile, or LP, groups were identified: LP1, including 144 patients with binge eating and purging (binge/purge); LP2, including 126 patients who were characterized by excessive exercise and extreme eating disorder cognitions (exercise-extreme cognitions); and LP3, 131 patients characterized by LPs that imperfectly resembled DSM-IV-TR eating disorders. Those in the LP1 group had characteristics resembling BN, while those in the LP2 and LP3 groups had characteristics that resembled AN with a relaxed weight criterion.

Dr. Eddy and her co-authors said that the identified LP groups were somewhat similar to those described in adult cases. However, rather than identifying several different bulimic spectrum latent groups, they found that youth with a predominance of these symptoms (such as binge eating and/or purging), clustered in a single latent group. The LP groups were more differentiated than the DSM-IV-TR categories across pretreatment eating disorder and general psychopathology indexes, as well as by weight change at follow-up. However, neither LP nor DSM-IV-TR categories predicted changes in binge/purge behaviors.

One interesting finding was that although there were few boys in this sample of patients, they were relatively more common in the group with the fewest typical eating disorder symptoms. According to the authors, this raises the possibility that boys with eating disorders may be more likely to show atypical symptoms than are girls.

The study results also suggest that latent profile analysis can be used to identify meaningful eating disorder phenotypes in a clinical sample of children and adolescents. Youths who regularly engage in binge eating and/or purging can be differentiated from those who do not. Youth who do not engage in these behaviors are more likely to be of lower weight and are organized into two groups distinguished by the presence of typical eating disorder cognitions and excessive exercise.

The authors suggest that many more ongoing studies, using developmentally sensitive assessment tools and a broad range of clinical indicators, are needed to prepare for the next version of the DSM.

AED Group Clarifies the Role of Family in Eating Disorders

The etiology of anorexia nervosa (AN) and bulimia nervosa (BN) is complex, and treatment poses unique challenges for clinicians. One continuing challenge is an old misconception that will not die, namely that certain types of interactions among family members play a specific role in the etiology or maintenance of an eating disorder.

The Academy for Eating Disorders recently published a position paper specifically refuting the idea that family factors are either the exclusive or even the primary mechanism that underlies risk for a family member developing an eating disorder (Int J Eat Disord 2010; 43:1). A group led by Daniel le Grange, PhD noted that it is disturbing to see recent examples of public misinformation about the role of the family and eating disorders. For example, a high-profile model recently blamed parents and families for the occurrence of eating disorders after media attention given to the deaths of several runway models from complications of AN.

Dr. le Grange and colleagues pointed out that research continues to focus on the role the family may have in causing or contributing to an eating disorder, despite treatment programs like the Maudsley program, which have shown the family as a potential helpful resource in therapy. According to the researchers, the strength of the findings, positive and negative, is called into question by absence of psychiatric controls to determine specificity of the prospective associations identified, lack of power to test for the prediction of full-blown cases of eating disorders, and reliance on risk factor assessments of questionable reliability and validity.

In another area, genetic studies, there is increasing evidence that heritable influences underlie susceptibility to both AN and BN. This type of "family influence," however, is of a much different type, according to the authors, and the nature of the transmissible risk is still unknown.

Evidence that family involvement is helpful

The good news concerning the family and eating disorders is that a small number of studies have shown that family involvement appears to be useful in reducing both psychological and medical morbidity, especially for younger patients with short-term eating disorders. This type of treatment seems to be well accepted by parents and patients alike. Only a few studies have assessed the helpfulness of family therapy for BN. Overall, the results of these studies suggest that family-based treatment method may be promising for some teens with BN, but also that further study is needed before any conclusions can be drawn.

The authors suggest that although no evidence supports the concept of "anorexigenic" parents, or parents who cause AN, in some cases involvement of the family may be contraindicated, for example, when parents have severe psychopathology. Clinicians should carefully assess parents' competencies, motivation, and any history of adverse or traumatizing events. The Study Group concluded that families should be involved routinely in the treatment of most young people with an eating disorder, and that the specific structure and extent of their involvement should be determined on a case-by-case basis.

Relapse in AN: Percent Body Fat is a Risk Factor

As many as half of patients with anorexia nervosa (AN) will relapse at least once, and these patients are at greatest risk for additional relapse during the first year after hospitalization. Identifying factors that place patients at greater-than-normal risk may help lower the recidivism rate.

Lindsay P. Bodell and Laurel E.S. Mayer MD, of New York State Psychiatric Institute, New York City, have targeted what they feel is an important relapse risk factor for AN patients, the percentage of body weight after weight restoration. In an earlier study, the authors had identified the percentage of body fat as a factor that increased the risk of relapse among 26 AN patients (Am J Psychiatry 2007; 164:970). In a newer study reported in the International Journal of Eating Disorders (2010. [Epub ahead of print]), the researchers found that a lower percentage of adipose tissue after short-term weight normalization was associated with a poorer outcome during the first year after inpatient treatment.

In the current study, 22 women between the ages of 18 to 45 years with AN were assessed for changes in body composition and body fat distribution in the year following inpatient treatment on an eating disorders unit. All the women met all the criteria for AN as listed in the Diagnostic and Statistical Manual of Mental Disorders-IV-TR, except for the amenorrhea criterion. Their treatment on the unit consisted of a structured behavioral program aimed at normalizing weight and eating behaviors. If they did not gain 1 lb of weight per week with the prescribed program of three meals and one snack per day, additional calories were added in the form of a liquid nutritional supplement (Ensure™ or Ensure Plus™). One criterion for discharge from the unit was restoration of weight to at least 90% of ideal body weight (IBW) as defined by Metropolitan Life actuarial tables. Body composition was assessed in participants after they maintained 90% of IBW for 2 to 4 weeks.

On the morning of testing, weight was measured to the nearest ¼ lb with a calibrated physician's beam balance scale and total body magnetic resonance imaging was performed. Images were analyzed for body composition, and percent adipose tissue was calculated as total adipose tissue divided by body weight x 100, and this calculation was used as a proxy for percent body fat.

After the inpatient program ended, patients were discharged to treatment in the community. Monthly phone calls by research staff helped follow the patients to obtain information about eating disorder symptoms and weight; every 3 months the participants had in-person evaluations.

More than half had a 'poor' outcome

Follow-up information was available for 21 of the 22 patients (one did not return phone calls and was lost to follow-up). Nineteen of the 21 participants completed enough information to determine Morgan-Russell criteria. Of the 21 participants, the outcome for 10 was categorized as "full," "good," or "fair," but for 11 others, it was "poor." As the authors had expected, there was a significant difference in body mass index, or BMI (mg/k2) between the full, good, or fair results groups and the poor outcome group: 20.8 kg/m2 vs.16.2 kg/m2, respectively.

At the time of initial testing, age, duration of illness, diagnostic subtype and BMI were not significantly different between the women with good, full, or fair outcome and the group with a poor outcome. At the end of one year, however, percent body fat was significantly lower in those with poor outcomes (22% body fat) compared to the three other groups (27% body fat).

According to the authors, the study results suggest that lower percent body fat in recently weight-restored women with AN may be a risk factor for relapse, and normalization of body fat levels may be an important element in long-term recovery from AN. Analyzing body composition may help identify patients at risk for relapse.

Inpatient vs. Outpatient Treatment for Adolescents with AN

Inpatient psychiatric treatment has been identified as the gold standard for treating adolescent patients with anorexia nervosa (AN), but thus far this approach has neither been more successful or cost-effective than outpatient treatment. In fact, most forms of intervention have not been well researched.

Dr. S. G. Gowers and other British researchers recently reported the results of a large population-based randomized controlled trial of the three most common treatments available in the United Kingdom for adolescent patients with AN. The Treatment Outcome for Child and Adolescent Anorexia Nervosa (TOuCAN) trial compared the merits of inpatient psychiatric treatment and two forms of outpatient management, Community Generic Child and Adolescent Mental Health Services (CAMHS) and a specialist multimodal multidisciplinary program developed just for the study (Health Technology Assessment 2010; 14:7).

The aim of this study was to determine if at 1, 2 and 5 years young people treated in specialist inpatient and outpatient services would have any advantages over those who used general outpatient treatment programs. In addition, the study sought to evaluate if inpatient management had any advantages over outpatient treatment and to examine the cost-effectiveness and the satisfaction of patients and physicians with each type of treatment.

The randomized controlled study was conducted among 215 young patients between 12 and 18 years of age presenting with AN at CAMHS. Inclusion criteria included food restriction plus or minus compensatory behaviors, weight below 85% of that expected based on age and height, intense fear of gaining weight or undue influence of weight or shape on self-evaluation, and primary or secondary amenorrhea of at least 3 months in females or menstruation only while on oral contraceptives. Those with severe chronic comorbid physical conditions that affected digestion or metabolism were excluded. Thirty-five centers in northwest England participated.

Study participants were randomized to either treatment as usual within community generic mental health centers; treatment consisting of individual cognitive behavior therapy, dietary advice, parental counseling and feedback on self-report measures; or to inpatient treatment within one of four specialist but not exclusively inpatient units. Outpatient treatment spanned at least 6 months; the length of inpatient treatment was at the service's discretion, with outpatient follow-up for a minimum of 6 months.

What the study results showed

Of the 167 young persons randomized to either inpatient or outpatient treatment, 67% adhered to the allocated treatment; adherence was lower in groups assigned to inpatient management. Each subject was followed up at 1 and 2 years (the main outcome point), and the main outcome measure was completed by 94% at 2 years (only 47% at 5 years). At each time point, there was significant improvement in all groups: 19% achieved a good outcome at 1 year, 33% at 2 years, and 64% at the 5-year point.

Patients who received inpatient treatment had poor results; among these were patients who failed to progress with outpatient treatment and who were transferred for inpatient treatment due to clinical symptoms. Generalist treatment was slightly more expensive over the first 2 years, largely because greater numbers of patients were subsequently admitted to hospital after the treatment phase. The cost-effectiveness analysis showed that specialist outpatient services were incrementally more cost-effective.

And, how satisfied were patients with treatment? Overall, young people were twice as likely to express positive views of outpatient treatment. Parents were also much more satisfied with outpatient treatment, and five times as many expressed positive than negative views of treatment. Parents were consistently more satisfied than were young patients with each type of treatment but both parents and young people were more satisfied with specialist than with general treatments—this was largely due to their confidence in "expertise" and their ability to form a good relationship with an individual therapist, in either an inpatient or an outpatient setting.

Some implications for health care

On the basis of their findings, the authors recommend that for moderately to severely ill adolescents with AN, outpatient services delivered by experienced expert professionals, and supported by medical management of physical complications as required, offer the most cost-effective approach to treatment. In contrast, lengthy psychiatric inpatient treatment does little to add to positive outcomes and is not cost-effective. Treatment by specialists with experience and expertise in managing AN is preferable, owing to its cost-effectiveness and higher levels of satisfaction among both patients and physicians. When young people with AN are managed in community health centers, a consultation and advice link with a specialist service may enable the treatment team to contain the anxiety and reduce the number of unnecessary hospitalizations.

The researchers also feel that further research is needed to help clarify the positive and negative aspects of inpatient care. Physical and psychological risk, parental anxiety and social and educational withdrawal by adolescents often result in inpatient admissions. This is often a logical step because of the availability of intensive psychological therapies, general support, and refeeding and respite from the external world. While satisfaction is generally good, particularly among parents, research outcomes are disappointing. This may suggest that adverse effects are under-recognized, according to the authors.

Overweight and Obesity Delay Puberty in Boys

Unlike overweight girls, who tend to reach puberty earlier than do girls of normal weight, overweight and obese boys in the U.S. may begin puberty later than thinner boys. These findings came from a study by Joyce M. Lee, MD, MPH and colleagues at the University of Michigan (Arch Pediatr Adolesc Med. 2010; 164:139).

At 11.5 years of age, boys with the highest body mass index (mean BMI z score 1.84) were 165% more likely to be prepubertal than were the thinnest boys, according to Dr. Lee and her colleagues. In one of the first longitudinal studies of weight and timing of puberty in males, the study results showed that unlike girls, higher body mass index, or BMI (kg/m2) in earlier childhood may be associated with and may precede later onset of puberty among boys. To further explore the relationship, Lee and colleagues analyzed the records of 401 boys from diverse socioeconomic backgrounds in 10 regions of the U.S. The authors used data from the National Institute of Child Health and Human Development Study of Early Child Care and Youth Development, which included data from full-term single children born in 1991.

The data included height and weight measurements of the children from age 2 years to 12 years, along with a visual assessment of whether the children had begun puberty by the age of 12, using Tanner genitalia staging at 9.5, 10.5, and 11.5 years. Boys were defined as pubertal if they were at Tanner stage 1 at 11.5 years and were otherwise categorized as pubertal.

For children and teens, overweight is defined as BMI at the 85th to 95th percentile and obesity as BMI greater than the 95th percentile, using Centers for Disease Control and Prevention BMI-for -age growth charts. Using these criteria, 14.4% of the teens were found to be overweight and 19.4% were obese by age 11.5. Overall, 49 boys, or 12.2%, were prepubertal at age 11.5, using Tanner staging.

The authors feel that their findings have important implications for understanding gender-based differences in the physiological mechanisms of puberty. Since puberty is regulated by the gonadotropin-releasing hormone axis for both girls and boys, it is not clear why such different associations between body fat and the timing of the onset of puberty would exist between the sexes.

The study did have some limitations, such as an inability to analyze the data according to race; most of the children in the study were Caucasian and the relationship between body fat and BMI is affected by race. In light of the increase in childhood obesity, the authors note that additional studies are needed to further investigate the epidemiologic link between body fat and the start and progression of puberty among boys.

Exploring Quality of Life in Eating Disorders

Although quality of life is often difficult to characterize, the World Health Organization (WHO) defines it as 'an individual's perception of their position in life in the context of the culture and value system in which they live and in relation to their goals, expectations, standards and concerns'(Social Science and Medicine 1995; 10:1403). Increasingly, the quality of life of eating disorders patients has become an important measure of treatment outcome, and it is also helpful for understanding the full impact of the illness.

A recent study evaluated the quality of life of individuals with eating disorders, with an emphasis on the effects of body mass index (BMI, kg/m2 ), and duration and severity of illness (Eur Eat Disorders Rev 2010; 18:147). Drs. Bryony Bamford and Richard Sly, of St. George's, University of London, evaluated 156 adults (148 women, 8 men) who were currently being treated for an eating disorder. Drs. Bamford and Sly evaluated each of the areas of these patients' quality of life according to specific eating disorder diagnoses. The patients ranged in age from 18 to 58 years, and the largest group, 56 patients (36%), had anorexia nervosa (AN), restrictive type. Forty other patients (26%) had bulimia nervosa (BN); 24 (15%) had binge-purge type AN; and 36 (23%) had eating disorder not otherwise specified (EDNOS).

The participants were diagnosed by experienced eating disorder clinicians using a semi-structured clinical interview based on the Eating Disorder Examination. The largest group, 98 patients (63%) came from an outpatient service. The participants also were evaluated with the Eating Disorders Quality of Life (EDQOL) scale, a 25-item self-report measure with four subscales (Psychological, Physical/Cognitive, Work/School and Financial). The EDQOL uses a 5-point scoring scale ranging from "never" to "always" to gauge the degree to which an individual feels his or her eating disorder affects their quality of life. A higher score equates to lower quality of life.

Severity of illness, but not duration, affected quality of life

The authors found that increased severity of illness correlated with lower quality of life. They reported that scores across all subscales of the EDQOL were higher than in the previous eating disorder sample assessed by Engel et al in 2006 (Int J Eat Disord 2006;39:62). This may have been due to the fact that the participants in the current study were all at the beginning of treatment for eating disorders and thus were likely to have more severe illness. Also in this group, BMI emerged as a significant predictor of quality of life, with lower BMIs contributing to a lower self-reported quality of life. Although significant differences were found on psychological and physical/cognitive quality of life subscales between those diagnosed with AN and those diagnosed with EDNOS, this might be in part due to the difference in BMIs between these groups. It suggests that BMI, in addition to other specific behavioral issues related to diagnosis, may have a direct impact on this group's quality of life.

Contrary to expectations, longer duration of illness did not result in poorer quality of life. This finding agrees with results of previous studies that have suggested that the global functioning of eating disorders patients does not decrease even with the continuation of eating symptoms. The authors note that this 'response shift"-- or adaptation to deceased function by changing values and/or expectations and finding ways to carry on with daily function despite the illness--is seen as a desirable outcome of adaptation to physical illness. However, in the case of mental illnesses, such as chronic AN, such a shift is likely to contribute to the low levels of motivation for recovery and possibly to their often strong denial that their illness has had any negative effect upon them.

The authors suggest that those who treat persons with eating disorders spend enough time using motivational approaches to help patients, and take time to explore with their patients ways in which the patients' values or expectations may have changed over time.

Trauma's Effect on Body Image

A person's body image reflects many factors in each individual's unique history of development. It is assumed that development of a coherent body image can be severely impaired by violations of body self-boundaries, leading to symptoms of body dissatisfaction and reduced body vitality. Sexual trauma is particularly serious because it affects extremely personal aspects of life and often occurs within the structure of the family. Results of a recent study showed that victims of sexual trauma have a more complex symptom pattern, including a more negative perception of their body, compared with persons who have undergone non-sexual trauma or no trauma.

Does body image differ by type of abuse?

Martin Sack, MD and co-workers in Munich, Germany, recently reported the results of their study examining the effects of sexual and nonsexual trauma upon a group of outpatients at the Hanover Medical School in Munich. The goal of the study was to see if there were differences in body image disorders depending on the type of abuse. Several questionnaires were mailed to all participants, to be completed prior to the first appointment. Diagnoses were confirmed by clinical interviews according to ICD-10 criteria, and all interviews were performed by board-certified specialists in psychosomatic medicine or by psychiatrists.

Patients completed the Posttraumatic Diagnostic Scale, a self-report of symptoms related to posttraumatic stress disorder. The Body Image Questionnaire was developed for the participants' cognitive and affective self-assessment of their bodies. This questionnaire is commonly used to evaluate body image in German-speaking countries and consists of 20 items, assessed on a four-point Likert scale regarding body perception and one's attitudes toward his or her body. The Somatoform Disorders Screening Instrument-7 Days asks about the presence of all physical complaints on the DSM-IV somatization disorder symptom list, symptoms of the ICD-10 somatization disorder, and the ICD-10 somatoform autonomic dysfunction symptom list during the prior 7 days. Patients rate the severity on a scale of 0 (not present) to 4 (very severe). The Brief Symptom Inventory is a short form of the Symptom checklist SCL-90-R, a self-rating instrument for detection of subjective impairment due to somatic and psychological complaints. The Dissociative Experience Scale used in this study was a German adaptation and extension of the Dissociative Experience Scale, which was developed for screening for dissociative symptoms. This form contains 16 additional questions.

Results: Women had more severe symptoms

Among the 167 females and 73 males who participated in the study, the average age was 37 years, and the males were significantly older, with an average age of 42 years. Approximately 147 (63%) of the subjects were living in a partnership. One hundred (42%) were employed full time, 44 (18%) were employed part time, 42 (17%) were unemployed and 54 (23%) were pensioners or homemakers.

Twenty percent of the patients had PTSD. Further diagnoses included depressive disorders in 41, somatoform disorders in 36 and eating disorders in 33 (14%). Sexual trauma was reported significantly more often by women than by men (63 women vs. 7 men, or 26.3% vs. 9.6%, respectively).Among the female participants, the researchers also found a significant association of trauma history with body image disorders, which was not the case among the men. Sexually traumatized women also showed significantly more complaints in the Brief Symptom index Global Severity Scale, compared with nontraumatized or nonsexually traumatized women, whereas no group differences were found among the men.

The authors concluded that victims of sexual trauma suffer from a more complex pattern of symptoms than do those with nonsexual trauma. Those who have undergone sexual trauma have a more negative perception of their own body. To deal with those who have experienced sexual trauma, the authors suggest that trauma-oriented psychotherapy that aims not only to relieve trauma-related post-traumatic stress disorder (PTSD) symptoms but also to aid psychosomatic reintegration can be helpful. For example, women who are victims of sexual abuse often seek body-oriented forms of psychotherapy in their quest for sexual abuse recovery, to increase their acceptance of their body and to foster the integration of self and bodily experience.

Eating Disorders Disrupt Healthy Sexual Function

Eating disorders can interfere with many aspects of normal life, including sexual function, according to a recent study of 242 women participating in the International Price Foundation Genetic studies (Int J Eat Disord; 2010; 43:123).

When Andrea Poyastro Pinheiro, MD and a group of eating disorders experts assessed physical intimacy, libido, sexual anxiety, partner status, and sexual relationships among the women, they found that nearly half (44.7%) reported either avoidance of or absence of sexual relationships. Women with restricting and binge-purge type anorexia nervosa (AN) had the highest percentage of loss of libido—75% and 74%, respectively, compared to 39.1% for those with bulimia nervosa (BN) and 45.4% for those with eating disorder not otherwise specified (EDNOS).

For comparison purposes only, data were taken from a group of 202 women without eating disorders who were not part of Dr. Pinheiro's study. Compared to the women with no history of eating disorders, more of the women with eating disorders had loss of libido, prevalence of sexual anxiety, tension, frequently changed partners, and reported detached relationships. More women with eating disorders also reported avoiding sexual relationships but fewer reported not having a partner. Nearly two-thirds of women with eating disorders reported loss of libido and elevated levels of sexual anxiety. And, compared to the women without eating disorders, more women with eating disorders reported loss of libido, elevated sexual anxiety, relationships without sex and relationships with tension.

Low BMI and more severe symptoms

The authors reported that one consistently observed finding was the association between low lifetime body mass index (BMI, or kg/m2) and loss of libido and presence of sexual anxiety and problems with sexual relationships. These findings are consistent with the explanation that low body weight impairs the physiological functioning of sexual organs and also echoes evidence from other studies showing that fluctuations in BMI are directly related to changes in sexual interest (Psychol Med 1981; 11:131; Lancet 1979; 1:612). Another explanation might be that, independent of physical changes, individuals with lower BMIs have more severe eating disorders and the increased illness may be associated with more profound body dissatisfaction, body distortion, depression, and discomfort with physical contact, all of which can lead to loss of libido and elevated levels of sexual anxiety.

The authors concluded that sexual intimacy is a fundamental aspect of healthy relationships that can be disrupted by an eating disorder, and thus it should be routinely assessed along with other more commonly evaluated areas of function. According to the researchers, little is now known about how recovery from an eating disorder may affect sexual functioning. However, future studies on the impact of eating disorders upon intimate relationships may help clinicians develop approaches to treatment that will help persons with eating disorders improve intimacy and interpersonal connections.

BOOK REVIEWS: Eating Disorders in Sport

(By Ron A. Thompson and Roberta Trattner Sherman. Routledge: New York, 2010; 284 pages; $49.95)

Eating disorders and sport go hand in hand. The initial presentations of anorexia nervosa and bulimia nervosa and many subclinical cases of eating disorders often entail excessive exercise in a sport context. Time and time again, these issues first present in middle school, high school or college athletics, habitually in association with certain individual-focused sports such as running and gymnastics, more frequently than with team sports such as volleyball or basketball. Athletic coaches and staff members of student health programs are often the first to see the consequences of excessive sport, and they commonly seek guidance from eating disorders specialists as to what is going on, what to expect, and what to do.

In this welcome update and expansion of their 1993 classic, Helping Athletes with Eating Disorders, Ron A. Thompson and Roberta Trattner Sherman, who've now been studying and concerned with the eating disorders-sport connection for several decades, give us a welcome, thorough and authoritative review of the field. These authors have earned their expertise through heavy involvement with national and international sport organizations. They have consulted with and provided educational programs for the International Olympic Committee and the National Collegiate Athletic Association (NCAA), among others. Their discussions offer detailed reviews of the research and clinical literature pertinent to all of the topics considered: an historical and phenomenological overview of eating disorders and disordered eating in sport; the relationships among subclinical and clinical conditions in eating disorder as reflected in athletes; risk factors and ways for identifying problem eating and eating disorders in these populations; how all stakeholders in the athletic environment have a role to play in the management of eating disorders; treatment issues; and, finally, prevention and education. An outstanding chapter on medical considerations is also contributed by Pauline Powers, a member of the EDR Editorial Board.

Addressed to a broad audience of clinicians, educators, athletic personnel and other interested parties, this book covers wide territory that will be accessible and interesting to all who work in these halls. Frequently, practical guidance is offered. In my own practice, the issues raised in this book come up frequently. Patients and parents are exceedingly interested in what is safe, and what are the risks of sport participation. School nurses, athletic faculty, and guidance counselors beg for direction regarding ways to safely run their programs and how to write administrative policies.

Everyone touching on the interface of adolescents and sport– and that includes the large majority of those reading this publication--will find much to learn and to consider in this book.

— JY

Q & A: AN with Borderline Personality Disorder

Q: I'm treating a patient with anorexia nervosa (AN), binge-purge type, who also meets diagnostic criteria for borderline personality disorder. In addition to severe weight restriction and frequent purging she's often suicidal, highly emotional, and habitually engages in cutting behavior. I know that dialectic behavior therapy (DBT) can be helpful for patients with borderline personality disorder, but is there any evidence that DBT might help a patient with both this co-morbid mix of bulimic type of anorexia nervosa plus a severe personality disorder? (G.J., Omaha, NE).

A: The patient you're describing is among the most challenging patients with eating disorders that clinicians are likely to encounter. Not surprisingly, patients with this clinical picture often have the worst prognoses. In efforts to treat such patients, clinical programs often employ amalgams of individual and group psychotherapies, using a variety of individual and group-based cognitive-behavioral, psychodynamic, and family treatments, highly scheduled and structured programming, medications, and, increasingly, DBT. Thus, today's clinicians basically attempt to take advantage of virtually every tool in the contemporary therapeutic bag to treat such patients.

Although randomized controlled trials of DBT have not to my knowledge been conducted with this population, a recent report describing a systematically conducted open trial and 15-month follow-up may be instructive. Overall, the results suggest some improvement, but certainly no miracle cure. In this German study, 24 women with borderline personality disorder (9 with co-morbid AN and 15 with bulimia nervosa [BN]), all of whom had previously failed to respond to eating-disorder-related inpatient treatments, were then treated in a specially adapted inpatient DBT program.

At follow-up, the remission rate was 54% for those with BN, and 33% for those with AN. Of note, 44% of women with AN crossed over to BN, and one woman additionally met criteria for AN. Although the mean weight of women with AN was not significantly increased immediately post-treatment, it had improved at follow-up. For women with BN, the frequency of binge-eating episodes was reduced at post-treatment as well as at follow-up. Self-rated eating-related complaints and general psychopathology, and ratings of global psychosocial functioning were significantly improved at post-treatment and at follow-up (J Behav Ther Exp Psychiatry. 2010 Apr 14. [Epub ahead of print]). On the whole, these findings support the utility of the adapted DBT inpatient program for those who have failed prior treatment attempts. But, we clearly need to do a lot better with our treatment efforts.

— JY