Influence of Overanxious Disorder of Childhood on the Expression of Anorexia Nervosa

REGULAR ARTICLE Influence of Overanxious Disorder of Childhood on the Expression of Anorexia Nervosa T.J. Raney, PhD 1 Laura M. Thornton, PhD 2 Wade Berrettini, MD 3 Harry Brandt, MD 4 Steven Crawford, MD 4 Manfred M. Fichter, MD 5,6 Katherine A. Halmi, MD 7 Craig Johnson, PhD 8 Allan S. Kaplan, MD 9 Maria LaVia, MD 1 James Mitchell, MD 10 Alessandro Rotondo, MD 11 Michael Strober, PhD 12 D. Blake Woodside, MD 8 Walter H. Kaye, MD 2 Cynthia M. Bulik, PhD 1 * ABSTRACT Objective: Childhood anxiety often precedes the onset of anorexia nervosa (AN) and may mark a liability to the emergence of an eating disorder for some women. This study investigates the prevalence of overanxious disorder (OAD) among women with AN and explores how OAD impacts AN symptoms and personality traits. Method: Participants were 637 women with AN who completed an eating disorders history, the Structured Clinical Interview for DSM-IV Axis I Disorders, and assessments for childhood anxiety, eating disorder attitudes, and associated personality traits. Results: Of 249 women (39.1%) reporting a history of OAD, 235 (94.4%) met criteria for OAD before meeting criteria for AN. In comparison to those without OAD, women with AN and OAD self-reported more extreme personality traits and attitudes and they engaged in more compensatory behaviors. Conclusion: Among individuals with AN, those entering AN on a pathway via OAD present with more severe eating disorder pathology. VC 2008 by Wiley Periodicals, Inc. Keywords: anorexia nervosa; overanxious disorder; genetic; childhood anxiety; social phobia; comorbid anxiety disorders; panic disorder; anxiety disorder; twin studies; agoraphobia (Int J Eat Disord 2008; 41:326 332) Introduction Clinical and epidemiological studies indicate that a substantial portion of women with anorexia nervosa Accepted 12 October 2007 Supported by Price Foundation. *Correspondence to: Cynthia M. Bulik, Department of Psychiatry, University of North Carolina at Chapel Hill, 101 Manning Drive, CB #7160, Chapel Hill, NC 27599-7160. E-mail: cbulik@med.unc.edu 1 Department of Psychiatry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 2 Department of Psychiatry, University of Pittsburgh, Pittsburgh, Pennsylvania 3 Center for Neurobiology and Behavior, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 4 Sheppard Pratt Hospital, Baltimore, Maryland 5 Department of Psychiatry, University of Munich (LMU), Munich, Germany 6 Roseneck Hospital for Behavioral Medicine, Prien, Germany 7 New York Presbyterian Hospital, Weill Medical College of Cornell University, White Plains, New York 8 Laureate Psychiatric Clinic and Hospital, Tulsa, Oklahoma 9 Program for Eating Disorders, Toronto General Hospital, Toronto, Toronto, Ontario, Canada 10 Neuropsychiatric Research Institute, Fargo, North Dakota 11 Department of Psychiatry, University of Pisa, Pisa, Italy 12 Semel Institute for Neuroscience and Human Behavior, Department of Psychiatry and Behavioral Science, University of California at Los Angeles, Los Angeles, California Published online 22 January 2008 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/eat.20508 VC 2008 Wiley Periodicals, Inc. (AN) report lifetime comorbid anxiety disorders. 1 3 Across studies, the lifetime prevalence of at least one anxiety disorder among patients with AN ranged between 23 and 54%. 1 Comorbid anxiety disorders occur across AN subtypes with approximately equal frequency. 2 5 Generalized anxiety disorder, obsessive compulsive disorder, and social phobia are the most common comorbid anxiety disorders reported, and they occur significantly more frequently in individuals with AN than in controls. 1,2,5 Anxiety disorders commonly precede the onset of the eating disorder 1,2,6,7 and persist even after long-term weight restoration 8 suggesting a more fundamental underlying anxious trait that is independent of nutritional state. Twin studies have established a substantial contribution of genetic factors to anxiety disorders with heritability estimates for both broadly defined generalized anxiety disorder and panic disorder around 40%. 9 Similarly, twin studies examining the contribution of genetic factors in the development of eating disorders report moderate contributions of additive genetic factors. The heritability of AN has been estimated to be between 33 and 84%, and the heritability of BN between 28 and 83% 10 13 with the remaining variance in liability to both disorders attributable to individual specific environmental factors, and with negligible impact of shared environmental 326 International Journal of Eating Disorders 41:4 326 332 2008

OAD OF CHILDHOOD AND AN factors. In short, there is considerable evidence that risk for eating disorders is influenced by genetic factors. Twin studies also suggest that eating disorders and anxiety disorders may have shared genetic transmission. Keel et al. 14 examined the comorbidity of anxiety and eating disorders in monozygotic twins who were discordant for eating disorder symptoms. They found that twins with eating disorder symptoms were significantly more likely to have an unaffected cotwin with an anxiety disorder, suggesting a shared transmission for anxiety and eating disorders. Silberg and Bulik 15 examined longitudinally reported data in teenage twin girls to identify common genetic liability among anxiety, depression, and eating disorder symptoms in early and late adolescence. Using multivariate genetic modeling, they identified a common genetic factor that influences the liability of childhood anxiety, separation anxiety, depression, and eating problems which remains influential throughout development. They also identified a genetic factor associated with early onset eating problems, but not associated with later eating problems or any mood disturbance. These findings suggest that anxiety and eating disturbances may share a common genetically influenced liability. However, that particular genetic liability may be only one of a variety of pathways to developing an eating disorder. Although shared liability between anxiety disorders and eating disorders has been suggested, and the onset of anxiety disorders often predates the onset of eating disorders, very little attention has been directed to understanding the impact that childhood anxiety disorders have on the subsequent expression and development of AN. Therefore, in a large phenotypically well-characterized sample of individuals with AN we (1) report the prevalence of overanxious disorder (OAD), (2) explore the impact of OAD on the expression and severity of eating disorder symptoms and personality traits, and (3) explore the relation between OAD and other anxiety disorders in a sample of women with AN. Method Participants Participants were recruited for a multisite international Price Foundation Genetic Study of Eating Disorders focusing on individuals with AN and their parents ( AN Trios ). This study was designed to identify susceptibility loci involved in the risk for eating disorders. Informed consent was obtained from all study participants, and all sites received approval from their local institutional review board. Male and female probands affected with AN were recruited from nine sites in North America and Europe including Pittsburgh (W.H.K), New York (K.A.H.), Los Angeles (M.S.), Toronto (A.S.K., D.B.W.), Munich (M.M.F.), Pisa (A.R.), Fargo (J.M.), Baltimore (H.B., S. C.), and Tulsa (C.J.), between 2000 and 2003. Probands were required to meet the following criteria: (a) modified DSM-IV (APA, 1994) lifetime diagnosis of AN, with or without amenorrhea; (b) low weight that is/was less than the fifth percentile of body mass index (BMI) for age and gender on the chart of a National Health and Nutrition Examination Survey epidemiologic sample 16 ; (c) onset before the age of 25 years; (d) weight that is/was controlled through restricting and/or purging, which includes vomiting, use of laxatives, diuretics, enemas, suppositories, or ipecac; (e) age between 13 years and 65 years; (f) Caucasian (one grandparent from another racial or ethnic group was acceptable); and (g) study diagnostic criteria were met at least 3 years before study entry. This last inclusion criterion ensured that AN individuals who were unlikely to develop binge eating were appropriately classified, as research has shown that most binge eating develops within the first 3 years of illness in AN. 17 21 Potential participants were excluded if they reported a maximum BMI since puberty, [27 kg/m 2 for females and [27.8 kg/m 2 for males. This exclusion of individuals who were overweight or obese was designed to increase sample homogeneity. Measures and Interviews Demographic and Clinical Variables. Data relative to current age, age at onset, and current, minimum, and maximum BMI were included in the analyses. Eating Disorder Diagnoses and Associated Features. Along the lines of Lilenfeld et al., 22 Kaye et al., 2 and Godart et al., 4 this study used lifetime histories of eating disorder behaviors in the categorization of eating disorder subtypes to capture subtype crossover for more accurate classification. Lifetime histories of eating disorders and the presence or absence of eating disorder behaviors (e.g., dieting, binge eating, purging) in probands were assessed with the Structured Inventory for Anorexia Nervosa and Bulimic Disorders (SIAB) 23 and with an expanded version of Module H of the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID). 24 The training procedures for the SIAB and SCID have been described in detail elsewhere. 25 Interviewers were mental health professionals with at least a master-level degree (most current interviewers were Ph.D. psychologists) who underwent centralized training and certification on all study interviews. On the basis of detailed clinical interview data, participants were subtyped as follows: AN, restricting subtype International Journal of Eating Disorders 41:4 326 332 2008 327

RANEY ET AL. (RAN, absence of lifetime purging or binging; n 5 228), AN, purging subtype (PAN, purging in the absence of lifetime binging; n 5 172); AN, binging subtype (BAN, binging, with or without purging; n 5 108), or diagnoses of both AN and BN during the course of illness (ANBN, n 5 108). The Eating Disorder Inventory22 (EDI-2) 26 was also used to assess disturbed eating attitudes and related pathology. Anxiety Disorder Diagnoses. The presence of OAD was assessed using the Schedule for Affective Disorders and Schizophrenia - Lifetime version (SADS-L, as modified by Merikangas et al. 27 for DSM-III-R diagnoses) in 90% of the sample and by the K-SADS 28 (modified for retrospective reporting) in 10% of the sample. Diagnostic algorithms were constructed to ensure compatibility across instruments. Other anxiety disorders were assessed with the Structured Clinical Interview for DSM-IV (SCID I). 24 OAD was chosen as our target disorder for two reasons. First, the SCID only assessed GAD in the past 6 months and we were most interested in childhood onset anxiety. Second, the prevalence of SAD was too low to yield meaningful comparisons. Personality and Character Traits. Temperament and character dimensions were measured with the Temperament and Character Inventory (TCI). 29 Perfectionism was assessed with the Frost Multidimensional Perfectionism Scale (MPS), 30 and trait levels of anxiety were assessed with the State-Trait Anxiety Inventory (STAI Form Y-1). 31 Obsessive-compulsive symptoms were assessed with two different measures: the Yale-Brown Obsessive Compulsive Scale (YBOCS) 32 and the Yale-Brown-Cornell Eating Disorder Scale (YBC-EDS). 33 The Y-BOCS is a semistructured interview designed to assess presence and severity of obsessive thoughts and compulsive behaviors typically found among individuals with obsessive compulsive disorder. The YBC-EDS assesses obsessive-compulsive features specific to eating disorders (e.g., those related to food, eating, weight, and exercise). The Revised NEO Personality Inventory (NEO PI-R) 34 is an assessment of personality based on the five-factor model of personality, and it measures the interpersonal, motivational, emotional, and attitudinal styles of adults and adolescents. Statistical Analyses All statistical analyses were conducted using SAS/ STAT 1 9.1 software. 35 To determine the prevalence of OAD in individuals with eating disorders across subtypes, chi-square tests were used to test between-group differences. v 2 tests were also used to determine differences between the OAD and no OAD group in prevalence of anxiety disorders. In order to assess the effects of OAD independent of any other anxiety disorder, the presence or absence of any other anxiety disorder was entered as a covariate in all of the following analyses. Differences between the OAD and no OAD groups in prevalence of the various eating disorder related behaviors was assessed using logistic regressions. Differences between the groups in all continuous measures were determined using analysis of variance with eating disorder subtype entered as an additional covariate. All p-values were corrected for multiple tests using the false discovery rate (FDR). 36 Results A total of 744 participants met criteria for a lifetime diagnosis of AN using the SCID and SIAB. Of those, 13 (1.7%) were males and were excluded from analyses due to small sample size. Seventy participants who were not assessed for OAD or had missing anxiety disorders data were removed from the analysis. An additional 24, who were determined to have a lifetime diagnosis of separation anxiety disorder (SAD), but not OAD were also not included. The resulting sample size was 637. Prevalence of Overanxious Disorder Two hundred forty-nine women (39.1%) met criteria for OAD. Significant differences emerged in the prevalence of OAD across AN subtypes (RAN 5 32.0%, PAN 5 43.8%, BAN 5 40.8%, ANBN 5 49.0%; v 2 5 11.5, df 5 3, p 5.009). Specifically, the PAN and ANBN groups had significantly higher prevalence of OAD than the RAN group. Of the women with OAD, 235 (94.4%) met criteria for OAD before they met criteria for AN. Six (2.4%) had AN at least 1 year prior to the onset of OAD and eight (3.2%) had onset of both AN and OAD in the same year. There was no difference in the age of onset of AN between those with OAD and those without OAD. For those with OAD, there was no difference in the age of onset of eating disorders between those with other anxiety disorders and those without. Eating Disorder Symptoms and Features As shown in Table 1, women with a history of OAD more often reported the use of appetite suppressants (v 2 5 16.60, p \.001), vomiting (v 2 5 20.13, p \.001), enemas (v 2 5 8.55, p 5.008), ipecac (v 2 5 10.45, p 5.004), diuretics (v 2 5 8.30, p 5.009), and laxatives (v 2 5 9.97, p 5.004) to control weight. In addition, women with histories of OAD reported longer total duration of eating disorder symptoms than those without OAD (F 5 27.90, p 5.01). They also reported greater EDI body dissatisfaction (F 5 11.42, p 5.003), drive for thinness (F 5 10.22, p 5.004), bulimia (F 5 11.31, 328 International Journal of Eating Disorders 41:4 326 332 2008

OAD OF CHILDHOOD AND AN TABLE 1. Frequency of various eating disorder behaviors by OAD: % (N) and logistic regression results comparing the two groups, presence or absence of any other anxiety diagnoses was entered as a covariate in all models* No OAD (N 5 388) OAD (N 5 249) v 2 (p-value) OR (95% CI) Binge eating 24.5 (95) 32.1 (80) 4.27 (.06) Fasting 45.4 (176) 59.7 (148) 2.10 (.18) Exercise 46.1 (179) 56.2 (140) 1.08 (.35) Appetite suppressants 22.2 (86) 42.3 (105) 16.60 (\.001) 2.16 (1.49, 3.13) Vomiting 37.2 (144) 56.2 (12) 20.13 (\.001) 2.22 (1.56, 3.15) Enemas 2.3 (9) 8.1 (20) 8.55 (.008) 3.48 (1.43, 8.42) Ipecac 3.6 (14) 9.3 (23) 10.45 (.004) 3.43 (1.58, 7.42) Diuretics 8.3 (32) 19.3 (48) 8.30 (.009) 2.11 (1.26, 3.53) Laxatives 28.4 (110) 44.2 (110) 9.97 (.004) 1.78 (1.24, 2.55) Any purging 48.8 (188) 66.5 (165) 16.35 (\.001) 2.06 (1.45, 2.94) Weight Phobia SIAB 83.5 (324) 92.0 (229) 3.37 (.10) Self Esteem SIAB 90.7 (351) 96.8 (241) 5.21 (.04) 2.48 (1.08, 5.66) Body Image SIAB 92.8 (360) 97.6 (243) 2.12 (.18) *All p-values have been adjusted using FDR. p 5.003), and YBC-EDS food preoccupations (F 5 5.65, p 5.03) (see Table 2). Personality and Temperament In comparison to AN women without OAD, women with OAD reported greater TCI harm avoidance (F 5 17.23, p \.001) and persistence (F 5 9.36, p 5.006), STAI trait anxiety (F 5 28.84, p \.001), and lower TCI self-directedness (F 5 7.81, p 5.01) and novelty seeking, (F 5 5.68, p 5.03) (Table 2). In comparison to AN women without OAD, those with OAD also reported more perceived MPS parental criticism (F 5 36.24, p \.001), higher parental expectations (F 5 11.61, p 5.003), higher personal standards (F 5 11.93, p 5.003), greater concern over mistakes (F 5 32.27, p \.001), and greater doubts about actions (F 5 25.56, p \.001) (Table 2). Also, women with AN and OAD reported lower NEO-PI R extraversion (F 5 5.34, p 5.04) and higher neuroticism (F 5 10.13, p \.001) than those without a history of OAD. OAD and Other Anxiety Disorders Individuals with OAD were more likely than those without OAD to develop additional anxiety disorders (Table 3). More participants with OAD were also diagnosed with generalized anxiety disorder, (v 2 5 42.58, p \.001), panic disorder (v 2 5 9.33, p 5.002), obsessive compulsive disorder (v 2 5 40.74, p \.001), social phobia (v 2 5 37.11, p \.001), or specific phobia (v 2 5 17.84, p \.001). The prevalence of agoraphobia was not related to OAD history; however, agoraphobia was relatively uncommon in this sample. For the participants who developed another anxiety disorder, the presence of OAD was also associated with earlier age of onset for generalized anxiety disorder (t 5 2.79, p 5.01) and obsessive compulsive disorder (t 5 3.78, p \.001), but not for the other anxiety disorders. Conclusion To our knowledge, this is the first study to report estimates of the prevalence of overanxious disorder of childhood among women with AN. Although OAD no longer exists in the DSM-IV, we have benefited from the inclusion of that diagnosis in our assessment in order to explore more carefully patterns of early onset of anxiety disorders and AN. Of 637 women with AN, 39% met criteria for OAD. In this study, the prevalence of OAD varied across the diagnostic subtypes of AN, with OAD more common among PAN and ANBN. This finding is inconsistent with some studies of comorbid anxiety which did not find differences across eating disorder subtypes, 2,3 but consistent with others. 5 Godart et al. 5 found that individuals with RAN and ANBN were more likely than those with BN to have a lifetime diagnosis of OCD, and RAN reported significantly higher prevalences of current OCD and GAD than those with BN. However, in that study, PAN was not included in the comparisons and prevalence of childhood anxiety disorders was not given. Our findings suggest that women with AN who engage in purging behaviors are more likely than those who do not purge to have had OAD in childhood. Strober 37 theorized that women with anxiety disorders and AN may have a similar liability that manifests as a heightened sensitivity to fear conditioning with resistance to extinction. Applying that theory, those women with AN who had preexisting OAD may simply be more vulnerable to this International Journal of Eating Disorders 41:4 326 332 2008 329

RANEY ET AL. TABLE 2. Means and results from analysis of variance assessing differences in personality variables and eating disorder attitudes between those with OAD and those without* Variable No OAD, (N 5 388) Mean (SD) OAD, (N 5 249) Mean (SD) F-Value (p-value) Eating disorder features ED age of onset 16.1 (2.7) 16.1 (3.0) 0.19 (.68) Lowest preferred weight 37.3 (8.1) 36.2 (8.0) 0.48 (.51) Highest BMI 21.3 (2.4) 21.4 (2.4) 0.91 (.39) Lowest BMI 13.8 (1.9) 13.7 (1.9) 0.00 (.95) ED duration 8.8 (7.4) 10.9 (7.6) 7.90 (.01) Lowest preferred daily caloric intake (kcal) 519.0 (328.9) 413.6 (290.0) 4.21 (.06) Temperament and Character Inventory Cooperativeness 34.5 (5.6) 35.1 (6.1) 2.18 (.82) Harm avoidance 20.0 (7.9) 24.2 (6.6) 17.23 (\.001) Novelty seeking 16.4 (7.1) 15.3 (7.0) 5.86 (.03) Reward dependence 16.3 (3.9) 16.9 (4.0) 3.24 (.10) Persistence 6.0 (1.9) 6.5 (1.6) 9.36 (.006) Self-directedness 28.4 (8.9) 24.9 (9.0) 7.81 (.01) Self-transcendence 13.7 (6.8) 14.7 (6.0) 0.60 (.48) Yale-Brown-Cornell Eating Disorder Scale Worst motivation to change 18.6 (5.3) 19.1 (5.5) 0.18 (.68) Worst rituals 12.5 (2.6) 13.6 (2.3) 4.33 (.06) Worst preoccupations 12.8 (2.5) 13.8 (2.2) 5.65 (.03) Yale-Brown Obsessive Compulsive Scale Obsessions 6.6 (6.5) 10.1 (5.8) 2.02 (.19) Compulsions 7.2 (6.7) 10.7 (6.2).056 (.49) Multidimensional Perfectionism Scale Concern over mistakes 32.2 (9.1) 38.1 (6.7) 32.27 (\.001) Doubts about actions 12.5 (3.8) 15.0 (3.4) 25.56 (\.001) Organization 25.1 (5.5) 25.9 (4.9) 2.59 (.15) Parental criticism 9.9 (4.4) 13.0 (4.9) 36.24 (\.001) Parental expectations 13.9 (5.7) 16.4 (6.0) 11.61 (.003) Personal standards 28.1 (5.6) 30.2 (4.4) 11.93 (.003) Eating Disorders Inventory-2 Body dissatisfaction 17.1 (7.0) 20.3 (6.8) 11.42 (.003) Bulimia 2.6 (4.4) 4.3 (5.5) 11.31 (.003) Drive for thinness 14.6 (5.9) 16.9 (5.1) 10.22 (.004) State-Trait Anxiety Inventory Form Y-1 Trait anxiety 50.4 (13.0) 58.3 (12.8) 28.74 (\.001) The Revised NEO Personality Inventory Extraversion 105.7 (22.1) 98.9 (22.7) 5.34 (.04) Neuroticism 105.7 (25.3) 120.8 (21.7) 10.13 (\.001) *The presence or absence of any other anxiety disorder was entered into all models as a covariate. Eating disorder subtype was also entered as a covariate for all models except those predicting eating disorder features. TABLE 3. Frequency of anxiety disorders by presence or absence of a history of OAD: % (N)* Prevalence of Anxiety Disorders Means (std) Age of Onset of Anxiety Disorders No OAD OAD v 2 (p-value) No OAD OAD t-score (p-value) Agoraphobia 2.3 (9) 3.7 (11) 0.94 (.33) 15.8 (8.0) 21.3 (8.3) 21.30 (.25) Generalized anxiety Disorder 7.2 (28) 25.9 (64) 42.58 (\.001) 15.6 (6.7) 11.3 (6.7) 2.79 (.01) Panic disorder 10.1 (39) 18.6 (46) 9.33 (.002) 20.6 (6.8) 19.4 (8.8) 0.70 (.49) Obsessive compulsive Disorder 42.9 (165) 68.8 (170) 40.74 (\.001) 13.6 (5.6) 11.1 (5.9) 3.78 (\.001) Social phobia 13.5 (52) 33.9 (83) 37.11 (\.001) 12.5 (4.8) 10.8 (4.8) 1.95 (.07) Specific phobia 5.4 (21) 15.5 (38) 17.84 (\.001) 8.7 (6.9) 9.3 (7.2) 20.28 (.80) *Comparison made by chi-square. Also, means (std) of age of onset of anxiety disorders by presence or absence of a history of OAD. excessive fear conditioning, have more intense anxiety responses, and turn to purging as an immediate intervention to reduce the extreme emotional distress. An alternative explanation is that those women with OAD may have a generalized heightened sensitivity to fear conditioning that, with the emergence of AN, is complicated by the addition of a weight based specific fear. The additive impact of the general and specific foci for the extreme fear conditioning may result in a more intense perceived need to reduce anxiety through purging. Another possible explanation is that anxious symptoms may be related to altered striatal dopamine function in AN. 38 Sustained activation of dopamine neurons is related to uncertainty. 39 Perhaps a contributing component of anxiety in AN individuals is a height- 330 International Journal of Eating Disorders 41:4 326 332 2008

OAD OF CHILDHOOD AND AN ened sensitivity to uncertainty, resulting in extreme distress reduction strategies such as purging. The presence of OAD in this sample was associated with a more severe and complex clinical presentation. Previous research has not consistently demonstrated a relationship between anxiety disorders and increased severity of eating disorder symptoms. 40 In contrast, our explicit focus on OAD with onset by definition in childhood identified a consistent and robust association between anxiety and severity of eating disorder symptoms and personality traits. Although all women with AN displayed the characteristic personality profiles expected in this population (e.g., harm avoidance, persistence, perfectionism, and lower self-directedness), those with a history of OAD displayed a profile that was magnified. The same pattern held for eating disorder behaviors, with those with a history of OAD engaging in more problematic weight control practices and displaying greater body dissatisfaction, higher drive for thinness, more eating preoccupation, and longer duration of illness. OAD was also associated with the presence of additional anxiety disorders. As may be expected, those individuals with a history of OAD were significantly more likely to have GAD, OCD, specific phobia, social phobia, and panic disorder. Agoraphobia was infrequently endorsed by participants and did not differ significantly between women with or without OAD. The presence of OAD was also associated with younger ages of onset for GAD and OCD than those without OAD who developed these conditions. Collectively, these observations suggest that childhood OAD is a pernicious disorder that is associated with greater severity and longer duration of eating disorder. The results of this investigation should be viewed within the context of several limitations. The OAD diagnoses and age at onset estimates were made retrospectively and subject to recall biases inherent in that approach. Individuals with anxiety disorders persisting into adulthood or who developed additional anxiety disorders later in life may have been more sensitized to recognize anxiety symptoms in childhood which could inflate the association between childhood onset and other anxiety disorders. Another limitation is that the SCID assesses only current GAD, so many cases of remitted GAD are missed. Therefore, many participants may, at an earlier time, have met criteria for GAD and would have gone undetected if the GAD had remitted at the time of assessment. Furthermore, the SCID screen for GAD is limited in that people who are always anxious may not report having been more anxious in the past 6 months. One of the strengths of studying OAD is that it captures those individuals with an early pervasive anxiety, even if the symptoms remitted at the time of assessment. In addition, the distinction of OAD as a diagnosis of childhood is somewhat artificial as many of the patients reported the onset of other anxiety disorders close in time to their childhood anxiety disorder. In the DSM-IV 41 the diagnosis of OAD was subsumed by the diagnosis of GAD, with the only remaining distinction being that children can present with fewer symptoms to meet diagnostic criteria. Also noteworthy, this study is part of multicenter genetic study, so the participants were recruited from populations of primarily European descent to ensure sample homogeneity that is critical in genetic studies. This homogeneity limits the generalizability of these findings to non-european groups. An OAD pathway may represent one entrée into AN, which portends persistent and pervasive anxious symptoms, obsessive and perfectionistic personality styles, and later anxiety disorders. Although these observations are based on retrospective recall, they highlight the importance of early detection of OAD as a potential means of averting the emergence of both later anxiety and eating disorders symptoms. Support for the clinical collection of participants and data analysis was provided by the Price Foundation. The authors thank the staff of the Price Foundation Collaborative Group for their efforts in participant screening and clinical assessments and the participating families for their contribution of time and effort in support of this study. This study was also supported by grant MH-66117 from the National Institutes of Health, Bethesda, MD. References 1. Godart NT, Flament MF, Perdereau F, Jeammet P. Comorbidity between eating disorders and anxiety disorders: A review. International J Eat Disord 2002; 32:253 270. 2. Kaye WH, Bulik CM, Thornton L, Barbarich N, Masters K, the Price Foundation Collaborative Group. Comorbidity of anxiety disorders with anorexia and bulimia nervosa. Am J Psychiatry 2004;161:2215 2221. 3. Blinder B, Cumella E, Sanathara V. Psychiatric comorbidities of female inpatients with eating disorders. Psychosom Med 2006;68:454 462. 4. Godart N, Berthoz S, Perdereau F, Jeammet P. Comorbidity of anxiety with eating disorders and OCD. Am J Psychiatry 2006;163:326. 5. Godart N, Berthoz S, Rein Z, Perdereau F, Lang F, Venisse J-L, et al. Does the frequency of anxiety and depressive disorders differ between diagnostic subtypes of anorexia nervosa and bulimia? Int J Eat Disord 2006;39:772 778. 6. Bulik C, Sullivan P, Fear J, Joyce P. Eating disorders and antecedent anxiety disorders: a controlled study. Acta Psychiatr Scand 1997;96:101 107. International Journal of Eating Disorders 41:4 326 332 2008 331

RANEY ET AL. 7. Deep AL, Nagy LM, Weltzin TE, Rao R, Kaye WH. Premorbid onset of psychopathology in long-term recovered anorexia nervosa. Int J Eat Disord 1995;17:291 297. 8. Pollice C, Kaye WH, Greeno CG, Weltzin TE. Relationship of depression, anxiety, and obsessionality to state of illness in anorexia nervosa. Int J Eat Disord 1997;21:367 376. 9. Kendler KS. Twin Studies of psychiatric illness: An update. Arch Gen Psychiatry 2001;58:1005 1014. 10. Bulik CM, Sullivan PF, Wade TD, Kendler KS. Twin studies of eating disorders: A review. Int J Eat Disord 2000;27:1 20. 11. Bulik CM, Sullivan PF, Tozzi F, Furberg H, Lichtenstein P, Pedersen NL. Prevalence, heritability, and prospective risk factors for anorexia nervosa. Arch Gen Psychiatry 2006;63:305 312. 12. Klump K, Miller K, Keel P, McGue M, Iacono W. Genetic and environmental influences on anorexia nervosa syndromes in a population-based twin sample. Psychol Med 2001;31:737 740. 13. Wade TD, Bulik CM, Neale M, Kendler KS. Anorexia nervosa and major depression: Shared genetic and environmental risk factors. Am J Psychiatry 2000;157:469 471. 14. Keel PK, Klump KL, Miller KB, McGue M, Iacono WG. Shared transmission of eating disorders and anxiety disorders. Int J Eat Disord 2005;38:99 105. 15. Silberg JL, Bulik CM. The developmental association between eating disorders symptoms and symptoms of depression and anxiety in juvenile twin girls. J Child Psychol Psychiatry 2005;46: 1317 1326. 16. Hebebrand J, Himmelmann GW, Heseker H, Schäfer H, Remschmidt H. Use of percentiles for the body mass index in anorexia nervosa: Diagnostic, epidemiological, and therapeutic considerations. Int J Eat Disord 1996;19:359 369. 17. Bulik C, Sullivan P, Fear J, Pickering A. Predictors of the development of bulimia nervosa in women with anorexia nervosa. J Nerv Ment Dis 1997;185:704 707. 18. Eckert E, Halmi K, Marchi P, Grove W, Crosby R. Ten-year follow-up of anorexia nervosa: Clinical course and outcome. Psychol Med 1995;25:143 156. 19. Strober M, Freeman R, Morrell W. The long-term course of severe anorexia nervosa in adolescents: Survival analysis of recovery, relapse, and outcome predictors over 10 15 years in a prospective study. Int J Eat Disord 1997;22:339 360. 20. Eddy KT, Keel PK, Dorer DJ, Delinsky SS, Franko DL, Herzog DB. Longitudinal comparison of anorexia nervosa subtypes. Int J Eat Disord 2002;31:191 201. 21. Tozzi F, Thornton LM, Klump KL, Fichter MM, Halmi KA, Kaplan AS, et al. Symptom fluctuation in eating disorders: Correlates of diagnostic crossover. Am J Psychiatry 2005;162:732 740. 22. Lilenfeld LR, Kaye WH, Greeno CG, Merikangas KR, Plotnicov K, Pollice C, et al. A controlled family study of anorexia nervosa and bulimia nervosa: Psychiatric disorders in first-degree relatives and effects of proband comorbidity. Arch Gen Psychiatry 1998;55:603 610. 23. Fichter MM, Herpertz S, Quadflieg N, Herpertz-Dahlmann B. Structured interview for anorexic and bulimic disorders for DSM-IV and ICD-10: Updated (third) revision. Int J Eat Disord 1998;24:227 249. 24. First M, Spitzer R, Gibbon M, Williams J. Structured Clinical Interview for DSM-IV Axis I Disorders, Research Version, Patient Edition. New York: New York State Psychiatric Institute, 1997. 25. Kaye WH, Devlin B, Barbarich N, Bulik CM, Thornton L, Bacanu S-A, et al. Genetic analysis of bulimia nervosa: Methods and sample description. Int J Eat Disord 2004;35:556 570. 26. Garner DM. Eating Disorder Inventory-2 Professional Manual. Odessa, FL: Psychological Assessment Resources, 1991. 27. Merikangas KR, Stevens D, Fenton B, Stolar M, O Malley S, Woods S, et al. Co-morbidity and familial aggregation of alcoholism and anxiety disorders. Psychol Med 1998;28:773 788. 28. Puig-Antich J, Orvaschel H, Tabrizi M, Chambers W. The Schedule for Affective Disorders and Schizophrenia for School-Age Children Epidemiologic Version (Kiddie-SADS-E), 3rd ed. New York: New York State Psychiatric Institute and Yale University School of Medicine, 1980. 29. Cloninger CR, Przybeck TR, Svrakic DM, Wetzel RD. The Temperament and Character Inventory (TCI): A guide to its Development and Use. St. Louis, MO: Washington University Press, 1994. 30. Frost RO, Marten P, Lahart C, Rosenblate R. The dimensions of perfectionism. Cogn Ther Res 1990;14:449 468. 31. Spielberger C, Gorsuch R, Luchene R. STAI manual for the State- Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press, 1970. 32. Goodman WK, Price LH, Rasmussen SA, Mazure C, Fleischmann RL, Hill CL, et al. The Yale-Brown obsessive compulsive scale. I. Development, use, and reliability. Arch Gen Psychiatry 1989;46: 1006 1011. 33. Sunday SR, Halmi KA. Comparison of the Yale-Brown-Cornell eating disorders scale in recovered eating disorder patients, restrained dieters, and nondieting controls. Int J Eat Disord 2000;28:455 459. 34. Costa PTJ, McCrae RR. NEO PI-R: Professional manual. Odessa, FL: Psychological Assessment Resources, 1992. 35. SAS Institute. SAS/STAT 1 Software: Version 9. Cary, NC: SAS Institute, 2004. 36. Benjamini Y, Hochberg Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J R Stat Soc Ser B 1995;57:289 300. 37. Strober M. Clinical and research forum pathologic fear conditioning and anorexia nervosa: On the search for novel paradigms. Int J Eat Disord 2004;35:504 508. 38. Frank GK, Bailer UF, Henry SE, Drevets W, Meltzer CC, Price JC, et al. Increased Dopamine D2/D3 receptor binding after recovery from anorexia nervosa measured by positron emission tomography and [11C]Raclopride. Biol Psychiatry 2005; 58:908 912. 39. Fiorillo CD, Tobler PN, Schultz W. Discrete coding of reward probability and uncertainty by dopamine neurons. Science 2003; 299:1898 1902. 40. Saccomani L, Savoini M, Cirrincione M, Vercellino F, Ravera G. Long-term outcome of children and adolescents with anorexia nervosa: Study of comorbidity. J Psychosom Res 1998;44:565 571. 41. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric Association Press, 1994. 332 International Journal of Eating Disorders 41:4 326 332 2008