Prevalence and Correlates of Eating Disorder Symptoms in Adolescents with Bipolar I Disorder

Abstract

Objective:

To investigate the prevalence and correlates of eating disorder symptoms in adolescents with bipolar I disorder (BP I).

Methods:

We retrospectively collected a DSM-IV-TR-based diagnostic assessment of 179 adolescents with BP I and evaluated clinical variables in those with and without eating disorder symptoms. For comparison, we retrospectively evaluated eating disorder symptoms in adolescents with generalized anxiety disorder (GAD).

Results:

Thirty-six percent of adolescents with BP I experienced lifetime eating disorder symptoms; among comorbid adolescents, 74% reported eating disorder cognitions and 40% reported symptoms related to bingeing, 25% purging, and 17% restricting. BP I adolescents with (vs. without) eating disorder symptoms had higher Children's Depression Rating Scale-Revised scores (40.5 vs. 34.5; p < 0.001; effect size = 0.59) and were more likely to be female (75% vs. 45%; p < 0.001; odds ratio = 3.8). There were no differences in Young Mania Rating Scale scores (p = 0.70); lifetime presence of attention-deficit/hyperactivity disorder (p = 0.86) and alcohol (p = 0.59) or substance (p = 0.89) abuse/dependence symptoms; age of BP I onset (p = 0.14); inpatient hospitalization status at baseline (p = 0.53); presence of lifetime inpatient hospitalization (p = 0.64) or suicide attempt (p = 0.35); seriousness of suicidality (p = 0.86); body mass index (p = 0.48); and second-generation antipsychotic (SGA; p = 0.32) or non-SGA mood stabilizer (p = 0.09) use. Eating disorder cognitions (rather than behaviors) were higher in the GAD group (58%) compared with the BP I group (27%; p = 0.004).

Limitations:

A retrospective study is subject to recall bias and limits our understanding of the temporal relationship between eating disorder and mood symptoms.

Conclusions:

Eating disorder symptoms are frequently comorbid in adolescents with BP I. The comorbidity is associated with more severe depression but does not confer a more severe illness course.

Introduction

Bipolar disorders (BP) are chronic disorders that account for significant economic, social, and health burden worldwide. Compared with the general population, eating disorders occur more frequently in adults with BP and the comorbidity is associated with greater physical and psychiatric burden (Álvarez Ruiz and Gutiê´rrez-Rojas, 2015; Balzafiore et al., 2017; Brietzke et al., 2011; Fagiolini et al., 2002; McElroy et al., 2016a; McElroy et al., 2013; McElroy et al., 2011; Seixas et al., 2012). Conversely, ∼8% of patients with eating disorders have comorbid BP, conferring greater eating disorder severity (Thiebaut et al., 2019b). Bipolar and eating disorders share phenomenologic similarities of behavioral activation, impulsivity, emotion dysregulation, and disturbances in eating behavior and weight regulation (McDonald et al., 2019). They also share familial and neurobiologic underpinnings; compared with controls, there are higher rates of BP in relatives of probands with anorexia nervosa and bulimia nervosa (McElroy et al., 2005) and several common neurotransmitter and neurotrophin systems contribute to the pathophysiology of both disorders (McElroy et al., 2011).

Correspondingly, eating disorders are more prevalent in BP than in the general population (i.e., a prevalence of 0.9%, 1%–2.3%, and 1.9%–3.5% for anorexia nervosa, bulimia nervosa, and binge-eating disorder, respectively, in the general population vs. up to 8.6% for anorexia nervosa, 24.6% for bulimia nervosa, and 28.8% for binge-eating disorder in the BP population) (Thiebaut et al., 2019a). The comorbidity is more frequent for eating disorders involving bingeing or purging, likely related to shared features of impulsivity and emotion dysregulation that etiologically separate them from restricting pathology (McElroy et al., 2016a; McElroy et al., 2011; Olatunji et al., 2012; Thiebaut et al., 2019a; Wildes et al., 2008).

In adults with BP (i.e., bipolar I disorder [BP I]; bipolar II disorder [BP II]; schizoaffective disorder, bipolar type; cyclothymic disorder), a comorbid eating disorder correlates with more severe mood and cognitive disturbances (McElroy et al., 2011), lower functioning, more depressive episodes, higher anxiety disorder comorbidity, earlier age of BP onset (Jen et al., 2013; McElroy et al., 2016a; McElroy et al., 2011), more frequent suicide attempts (McElroy et al. 2016a; McElroy et al. 2013; McElroy et al. 2011), heightened alcohol and substance abuse/dependency (Fornaro et al., 2010; Jen et al., 2013; McElroy et al., 2013), and delayed recovery from depression (Balzafiore et al., 2017) and, in women with only BP I, with earlier onset of BP I, more depressive episodes, more suicide attempts, and higher rates of alcohol and drug abuse or dependence (Brietzke et al., 2011).

Although adolescence is a critical time for the development of both bipolar and eating disorders, their relationship in adolescents is understudied and poorly characterized (Ward et al., 2019). Among adolescents with BP I, BP II, or bipolar disorder not otherwise specified (BP NOS), 28.4% report a lifetime eating disorder (3.6% anorexia nervosa; 8.1% bulimia nervosa; 16.8% eating disorder not otherwise specified [ED NOS]), whereas 11.2% report a lifetime history of binge eating unassociated with an eating disorder diagnosis (Khoubaeva et al., 2022). In addition, 12% of adolescents or young adults with BP demonstrate binge-eating behavior (Martin et al., 2016); adolescents with BP and overweight/obesity are more likely to have a lifetime history of binge-eating behavior or bulimia nervosa compared with their normal-weight counterparts (Goldstein et al., 2016); adolescents in the community with hypomania (but not mania) are more likely to have eating disorders; and full and subthreshold eating disorders significantly co-occur in high school students with subthreshold BP (McElroy et al., 2005).

Given its significant impact on disease course and severity, it is important to evaluate the effect of comorbid eating disorder symptoms in BP, particularly among adolescents who may present a critical developmental window for mitigation of disease severity and progression. To our knowledge, the present study is the first to evaluate the prevalence and impact of a full range of eating disorder symptoms—bingeing, purging, and restricting symptoms—in adolescents with BP I.

Anxiety is instrumental in the development and maintenance of eating disorders (Bulik, 1995) and, correspondingly, anxiety disorders often co-occur in individuals with eating disorders (Swinbourne and Touyz, 2007; Watson et al., 2014). In comorbid adolescents, generalized anxiety disorder (GAD) is one of the most common anxiety disorders and is particularly associated with subthreshold anorexia nervosa (Hughes et al., 2013). Conversely, in adults with anxiety disorders, 13.5% met criteria for an eating disorder (ED NOS in all cases) with subthreshold symptoms noted across eating disorder subtypes (Swinbourne et al., 2012). Insomuch that the relationship between anxiety and eating disorders is relatively better characterized, we investigated the prevalence of eating disorder symptoms in adolescents with GAD to compare comorbidity rates and the distribution of symptoms among eating disorder subtype.

We hypothesized that eating disorder symptoms are highly prevalent in adolescents with BP I and that binge/purge symptoms are more prevalent than restricting symptoms. In addition, we hypothesized that eating disorder symptoms comorbid with BP I confers a more severe illness course and is associated with second-generation antipsychotic (SGA) use and a higher body mass index (BMI). Finally, we hypothesized that eating disorder symptoms, particularly related to bingeing and purging, are more prevalent in adolescents with BP I than those with GAD.

Methods

Participants

Participants included 179 adolescents 12–21 years of age with BP I, recruited from outpatient community clinics and inpatient psychiatry units at Cincinnati Children's Hospital Medical Center. The comparison group included 24 participants 12–17 years of age with GAD, recruited from outpatient community clinics and the University of Cincinnati Pediatric Anxiety Disorders outpatient clinic.

Assessments

After participants signed informed consent for study participation, we collected a DSM-IV-TR-based diagnostic assessment (i.e., the Washington University in St. Louis Kiddie Schedule for Affective Disorders and Schizophrenia [WASH-U-KSADS]) (Geller et al., 2001) from two Institutional Review Board (IRB)-approved longitudinal prospective studies of adolescents with BP I. In addition, we collected Anxiety Disorders Interview Schedule (ADIS) for DSM-IV: Child Version diagnostic assessments (Silverman and Albano, 1996) from an IRB-approved longitudinal prospective study of adolescents with GAD.

Demographic and eating disorder symptom measures

In the BP I and GAD cohorts, we evaluated age, sex, and race, BMI, and the presence of any lifetime eating disorder symptom. Given that subthreshold eating disorder symptoms closely resemble and are as impairing as threshold disorders, it is clinically meaningful to evaluate individual eating disorder symptoms (Martin et al., 2000).

In the BP I cohort, assignment of a subthreshold or threshold score on a 0–3 scale (i.e., 0—no information; 1—not present; 2—subthreshold; 3—threshold) in the following areas was counted toward the analyses: (1) eating disorder cognitions (e.g., body image disturbance; subjectively feeling “fat” and often being bothered by these thoughts merited a subthreshold score while self-perceptions about weight unaltered by contrary objective evidence merited a threshold score); (2) restricting symptomatology (e.g., emaciation; a weight below 90% of ideal merited a subthreshold score while a weight below 85% of ideal merited a threshold score); (3) binge symptomatology (e.g., engaging in “eating attacks” or binges; binges occurring less than once a week merited a subthreshold score while binges occurring once per week or more merited a threshold score); (4) purge symptomatology (e.g., use of weight loss methods such as diet pills, and/or self-induced vomiting, and/or laxatives, and/or excessive exercise, and/or food intake restriction; use of a weight loss method less than one time per week merited a subthreshold score while use of a method one or more times per week merited a threshold score).

In the GAD cohort, we evaluated the corresponding ADIS questions: eating disorder cognitions (e.g., “Do you wish that you could be thinner than you are now?”); restricting symptomatology (e.g., “Do you have times that you go without eating {fasting} so that you can control our weight?”); binge symptomatology (e.g., “Do you ever find yourself eating a lot of food at one time?”) and purge symptomatology (e.g., “People try all sorts of things to control or lose weight. They might go on diets, use medications, exercise for hours, or even try to vomit so they don't lose any weight. Do you do anything like that to control your weight?”).

Measures of comorbidity and clinical severity

In the BP I cohort, we also evaluated the lifetime presence of attention-deficit/hyperactivity disorder (ADHD) and substance abuse/dependence symptoms and for SGA and non-SGA mood stabilizer use at screening. To assess BP I clinical course and severity, we evaluated the age of BP I onset, inpatient hospitalization status at baseline, and the presence of lifetime inpatient hospitalization and/or suicide attempt. If suicidal behavior was present, we assessed the seriousness of the suicidality based on the level of intent and potential for lethality; a score of ≥3 on either measure was counted as “serious.” A score of 3–5 indicated serious intent [i.e., 3; definite but ambivalent or 4; serious or 5; very serious] and a score of to 3–6 indicated potential for greater lethality (i.e., 3; mild [e.g., took 10 aspirin] or 4; moderate [e.g., took 10 Seconals; brief unconsciousness] or 5; severe [e.g., cut his/her throat] or 6; extreme [e.g., experienced coma as a result of a suicide attempt]).

We reviewed baseline Children's Depression Rating Scale-Revised (CDRS-R) scores (a clinician-administered semistructured 17-item interview with scores ≥40 indicating at least moderate depressive symptomatology) (Poznanski et al., 1984) and Young Mania Rating Scale (YMRS) scores (an 11-item clinician-administered interview with a score of ≥25 suggesting severe manic symptoms) (Young et al., 1978) to evaluate affective symptoms.

Statistical analyses

General linear and logistic regression models were used to evaluate clinical variables between BP I adolescents with and without eating disorder symptoms. Given the clinically relevant associations among race, sex, and eating disorder symptoms in the BP I group, these demographic variables were included as covariates in the analyses of clinical variables. In addition, the types of eating disorder symptoms in BP I adolescents were examined by sex while controlling for race. Fisher's exact tests were used for some BP I versus GAD comparisons due to the small GAD sample size. Although the BP I and GAD groups differed significantly by sex, this variable was not included as a covariate in BMI and eating disorder symptom comparisons due to the small sample size of the GAD group, the confounding of sex in the GAD group (i.e., 21 of 24 adolescents with GAD were female), and the unbalanced distribution of eating disorder symptom type. All hypothesis tests were two-sided with a significance level of 0.050 and conducted using SAS software, version 9.4 (Cary, NC).

Results

As shown in Table 1, among the total sample adolescents with BP I, 36% experienced lifetime eating disorder symptoms. Among comorbid adolescents (n = 65), 74% experienced eating disorder cognitions, 40% experienced symptoms related to bingeing, 25% purging, and 17% restricting (Table 2). As in Table 2, adolescents with comorbid BP I and eating disorder symptoms were similar in age and race and were more likely to be female (p < 0.001) compared with BP I adolescents without eating disorder symptoms. In addition, females reported significantly more purging behavior compared with males (Table 3; p = 0.05).

Table 1.

Comparision of the Bipolar I Disorder and Generalized Anxiety Disorder Groups

	BP I (n = 179)	GAD (n = 24)	p
Age, years	15.7 (2.5)	15.2 (1.7)	0.199
Sex, female	100 (56%)	21 (87%)	0.003
Race, Caucasian	137 (77%)	22 (92%)	0.091
BMI, kg/m²	24.1 (6.3)	22.9 (8.3)	0.426
Any lifetime ED subthreshold or threshold symptom	65 (36%)	14 (58%)	0.038
Cognitions	48 (27%)	14 (58%)	0.004
Binge symptoms	26 (15%)	0 (0%)	0.049
Purge symptoms	16 (9%)	0 (0%)	0.226
Restricting symptoms	11 (6%)	0 (0%)	0.368

WASH-U-KSADS responses in the BP I group and ADIS responses in the GAD group. Mean (standard deviation) or number (percentage) shown. Of the 14 GAD subjects with ED symptoms, 13 were female.

ADIS, Anxiety Disorders Interview Schedule for DSM-IV: Child Version; BMI, body mass index; BP I, bipolar I disorder; ED, eating disorder; GAD, generalized anxiety disorder; WASH-U-KSADS, Washington University in St. Louis Kiddie Schedule for Affective Disorders and Schizophrenia.

Table 2.

Comparison of the Bipolar I Disorder Group With and Without Eating Disorder Symptoms

	Without ED symptoms (n = 114)	With ED symptoms (n = 65)	p
Age, years	15.6 (2.7)	16.0 (2.2)	0.298
Race, Caucasian	82 (72%)	55 (85%)	0.054
Sex, female	51 (45%)	49 (75%)	<0.001
BMI, kg/m²	23.7 (6.4)	24.7 (6.1)	0.479
Type of ED symptom
ED cognitions		48 (74%)
Binge symptoms		26 (40%)
Purge symptoms		16 (25%)
Restricting symptoms		11 (17%)
BP I diagnosis––manic (vs. mixed)	42 (37%)	15 (23%)	0.094
SGA use at screening	25 (22%)	19 (29%)	0.317
Non-SGA mood stabilizer use at screening	6 (5%)	0 (0%)	0.088
Lifetime attention-deficit/hyperactivity disorder	60 (54%)	27 (42%)	0.859
YMRS score at baseline	25.9 (5.8)	25.6 (5.0)	0.704
YMRS score change	−15.1 (8.6)	−15.5 (8.0)	0.865
CDRS-R score at baseline	34.5 (8.4)	40.5 (8.3)	<0.001
CDRS-R score change	−7.9 (10.4)	−11.8 (11.6)	0.081
Age of BP I onset	13.4 (2.5)	13.9 (1.9)	0.135
Lifetime alcohol abuse/dependence symptoms	11 (10%)	10 (17%)	0.591
Lifetime substance abuse/dependence symptoms	56 (52%)	30 (50%)	0.887
Inpatient hospitalization at baseline	63 (55%)	39 (60%)	0.534
Lifetime inpatient hospitalization	76 (71%)	43 (68%)	0.635
Lifetime suicide attempt	31 (28%)	20 (32%)	0.347
Seriousness of suicidality	26 (23%)	17 (26%)	0.854
Seriousness of suicidal behavior or suicide attempt	2.4 (1.5)	2.5 (1.1)	0.759
Medical lethality of suicidal behavior or suicide attempt	2.2 (1.3)	2.5 (1.0)	0.722

Mean (standard deviation) or number (percentage) shown.

BMI, body mass index; BP I, bipolar I disorder; CDRS-R, Children's Depression Rating Scale-Revised; ED, eating disorder; SGA, second-generation antipsychotic; YMRS, Young Mania Rating Scale.

Table 3.

Bipolar I Disorder Group: Type of Eating Disorder Symptom by Sex (Controlling for Race)

	Female (n = 49)	Male (n = 16)	p
ED cognitions	37 (76%)	11 (69%)	0.190
Binge symptoms	18 (37%)	8 (50%)	0.335
Purge symptoms	15 (31%)	1 (6%)	0.050
Restrict symptoms	9 (18%)	2 (12%)	0.622

ED, eating disorder.

While controlling for these demographic differences, comorbid adolescents had significantly higher baseline CDRS-R scores (mean score = 40.5; standard deviation = 8.3) than adolescents without eating disorder symptoms (mean score = 34.5; standard deviation = 8.4; p < 0.001). Between BP I adolescents with and without lifetime eating disorder symptoms, there were no differences in age of BP I onset (p = 0.14), baseline YMRS scores (p = 0.70), lifetime presence of ADHD (p = 0.86), presence of alcohol (p = 0.59) or substance (p = 0.89) abuse/dependence symptoms, seriousness of suicidality (p = 0.85), presence of lifetime inpatient hospitalization (p = 0.64), inpatient hospitalization status at study baseline (p = 0.53), presence of lifetime suicide attempt (p = 0.35), BMI (p = 0.48), and SGA (p = 0.32) and non-SGA (p = 0.09) mood stabilizer use (Table 2).

Threshold lifetime eating disorder diagnoses were less common: 2.8% met criteria for a current disorder, ED NOS in all cases, and 3.9% met criteria for a past eating disorder: bulimia nervosa (n = 3), anorexia nervosa (n = 1), and ED NOS (n = 3). All adolescents meeting criteria for current ED NOS reported bingeing and/or purging symptoms and 60% reported concomitant symptoms associated with restricting pathology. Among those with past ED NOS, all reported bingeing and/or purging and the majority reported concomitant restricting symptoms.

Table 1 shows the comparisons between BP I and GAD adolescents. The GAD group had a significantly higher proportion of females (p = 0.003) and adolescents with any eating disorder symptom (p = 0.04). Specifically, rates of eating disorder cognitions were higher in the GAD group (58%) compared with the BP I group (27%; p = 0.004) while binge symptoms were nonexistent in the GAD group compared with the BP I group (15%; p = 0.049). No threshold eating disorders were reported in the GAD group.

Discussion

To our knowledge, we are the first to characterize the relationship between BP I and bingeing, purging, and restricting symptoms in adolescents. We found that eating disorder symptoms occur frequently in adolescents with BP I with 36% experiencing any lifetime eating disorder symptom. Among these, binge (40%) and purge (25%) symptoms occurred most commonly. Consistent with extant bipolar adolescent studies, threshold lifetime eating disorder diagnoses were less common (Khoubaeva et al., 2022; Martin et al., 2016; McElroy et al., 2005). In addition, we found that BP I comorbid with eating disorder symptoms confers greater depressive symptomology but not a more severe BP illness course.

Varied sample characteristics, threshold criteria, and assessment tools account for substantial heterogeneity in reported prevalence rates of eating disorders in BP; with this in mind, some studies demonstrate higher comorbidity rates for a threshold eating disorder than the present study. We evaluated adolescents with BP I, who experience manic, not hypomanic, episodes. While the mostly adult evidence base suggests that eating disorder comorbidity rates are similar across BP diagnoses (McDonald et al., 2019), our findings support contrasting evidence that adolescents who experience hypomania, but not mania, are more likely to have a comorbid eating disorder (McElroy et al., 2005). Indeed, in a sample of youth with BP, participants with BP II and BP NOS were significantly more likely than those with BP I to report a lifetime or current threshold eating disorder (Khoubaeva et al., 2022).

Higher prevalence of bingeing and purging symptoms relative to restricting symptoms in the present study aligns with adult evidence of higher prevalence rates of bulimia nervosa and binge-eating disorder (McElroy et al., 2016a; McElroy et al., 2011, Thiebaut et al., 2019a; Wildes et al., 2008) and with findings in comorbid adolescents indicating that anorexia nervosa is the least prevalent among the eating disorders (Khoubaeva et al., 2022).

Our preponderance of binge-eating symptoms (15% of the total sample) supports adult findings that binge eating, compared with anorexia nervosa, bulimia nervosa, and binge-eating disorder, carries the highest prevalence across BP studies (McDonald et al., 2019) and with findings in comorbid adolescents indicating that binge eating is more prevalent than anorexia nervosa and bulimia nervosa (Khoubaeva et al., 2022). Notably, binge-eating disorder, new in DSM-5, could not be evaluated in our DSM-IV-TR-based study; moreover, binge-eating disorder could be underrepresented in an adolescent population given its later onset (mean age 25.4 years) relative to the other eating disorders (Hudson et al., 2007).

There is a higher prevalence of anorexia nervosa in BP than in the general population, particularly for the binge-purge subtype (McDonald et al., 2019; Wildes et al., 2008). In the present study, 6% of adolescents with BP I reported symptoms associated with restricting pathology (and one reported past threshold anorexia nervosa). We evaluated restricting symptoms without regard for their association with a categorical eating disorder and therefore could not characterize bingeing/purging related to subthreshold anorexia nervosa; however, our findings suggest that restricting symptoms unassociated with threshold anorexia nervosa are not uncommon in adolescents with BP I.

Our findings provide further evidence that females are significantly more likely than males to report eating disorders comorbid with BP (McElroy et al., 2016b, McElroy et al., 2011; Seixas et al., 2012) and they extend the adolescent literature (Khoubaeva et al., 2022) by corroborating that a female preponderance remains true for adolescents with BP I and when subthreshold eating disorder symptoms are considered. Consistent with adult BP findings in which 25.9% of males reported a lifetime history of clinically significant eating disturbance (Wildes et al., 2007), males represented 25% of our comorbid sample, highlighting the need to screen for eating disorder symptomatology in males and females. In addition, as in comorbid adults (Ramacciotti et al., 2005), in comorbid adolescents, females reported significantly more purging behavior compared with males.

The present findings do not support our hypothesis that comorbid adolescents experience a more severe illness course. We did not find group differences in inpatient hospitalization status at baseline or in the presence of lifetime inpatient hospitalization, and, in contrast to adult findings, there was no difference in age of BP I onset (Brietzke et al., 2011; Jen et al., 2013; McElroy et al., 2016a; McElroy et al., 2011). In contrast to adult (Brietzke et al., 2011; McElroy et al., 2016a; McElroy et al., 2016b; McElroy et al., 2013; McElroy et al., 2011) and adolescent (Khoubaeva et al., 2022) findings, there were no differences in lifetime suicide attempt or seriousness of suicidality. It is possible that suicidality is more likely to be associated with eating disorder symptoms comorbid with BP II or BP NOS or with a comorbid threshold eating disorder. In some cases, it is also possible that the relationship between the comorbidity and clinical severity strengthens with time, rendering it more likely to be observed in adulthood.

Greater depressive (but not manic) symptom severity in comorbid adolescents in the present study corroborates evidence of more symptomatic depressive mood episodes (Álvarez Ruiz and Gutiê´rrez-Rojas, 2015; Brietzke et al., 2011; Fagiolini et al., 2002; Seixas et al., 2012) and delayed recovery from depression (Balzafiore et al., 2017) in comorbid adults and higher lifetime most severe depression in adolescents (Khoubaeva et al., 2022). Indeed, eating pathology (i.e., an eating disorder or eating disorder symptoms) and depression (i.e., a depressive disorder diagnosis or depressive symptoms) are commonly comorbid and share a bidirectional relationship; eating pathology predicts depression and vice versa. Accordingly, it is possible that depressed individuals develop eating pathology to regulate negative mood, while factors such as calorie deprivation or persistent striving to achieve a physical ideal effect depression (Puccio et al., 2016).

Moreover, the co-occurrence of eating pathology and depression in adolescents may be explained by shared characteristics of low self-esteem, low self-worth (Sahlan et al., 2021), and fear of negative social evaluation (Kenny et al., 2021). Our finding of more severe depression in comorbid adolescents supports the association between depression and eating pathology and extends the literature by suggesting that eating disorder symptoms significantly impact mood in adolescents with BP I.

In contrast to previous adult (Brietzke et al., 2011; Fornaro et al., 2010; Jen et al., 2013; McElroy et al., 2013) and adolescent (Khoubaeva et al., 2022) studies, we did not find a higher rate of substance abuse/dependence symptoms in comorbid adolescents. In addition, although ADHD and eating disorders share neurocognitive deficits and, in the case of bulimia nervosa and binge-eating disorder, impulsivity as a core characteristic (Reinblatt, 2015), we did not find higher rates of ADHD in comorbid adolescents, corroborating extant adolescent findings (Khoubaeva et al., 2022) and suggesting that ADHD does not increase the risk of developing eating disorder symptoms in adolescents with BP I.

Aligned with extant adolescent BP findings (Khoubaeva et al., 2022), neither SGA or non-SGA mood stabilizer use was associated with the presence of eating disorder symptoms in adolescents with BP I. This is noteworthy, especially since SGAs may disrupt serotonin neurotransmission and hormones (e.g., leptin and ghrelin) involved in the regulation of satiety mechanisms, heightening vulnerability to developing bingeing and purging (McDonald et al., 2019). However, we did not consider medication adherence, dosage, duration of use, and past medication trials, important factors warranting future investigation in understanding the relationship SGA/mood stabilizer use in comorbid adolescents.

Correspondingly, we found no difference in BMI in BP I adolescents with and without comorbid eating disorder symptoms. This corroborates BP adolescent findings (Khoubaeva et al., 2022) and contrasts with opposing evidence that lifetime binge eating is associated with overweight/obesity in adolescents with BP (Goldstein et al., 2016) and increased BMI (McElroy et al., 2016b; Wildes et al., 2008) and weight gain and obesity in adults with BP (Yaramala et al., 2020). Insomuch that weight gain is associated with depression severity in adults (Fagiolini et al., 2002) and the relationship between BP and weight gain may strengthen with time as a result of cumulative mood episodes (Goldstein et al., 2016), our finding of more severe depression in comorbid adolescents highlights the need to monitor weight in this population, who may be particularly vulnerable to obesity/overweight in adulthood.

The lifetime prevalence of at least one anxiety disorder is as high as 75% among individuals with eating disorders (Black Becker et al., 2004). Our evaluation of adolescents with GAD suggests that the converse is not true and aligns with extant findings that eating disorders, particularly bulimia nervosa, are the least comorbid of the psychiatric disorders in children and adolescents with GAD (Mohammadi et al., 2020). Compared with adolescents with BP I, those with GAD reported significantly more eating disorder symptoms; however, all reported eating disorder cognitions rather than behaviors. This suggests that eating-disordered thinking rather than behavior is more prevalent in adolescents with GAD compared with adolescents with BP I, possibly related to traits of impulsivity and emotion dysregulation shared between bingeing/purging behavior and BP I (and not GAD).

Limitations

Our findings should be interpreted within the context of several limitations. First, we did not adjust for multiple statistical comparisons to maximize our statistical power. In addition, our evaluation of a DSM-IV-TR-based diagnostic assessment limited our ability to evaluate for binge-eating disorder. Also, the association between specific mood states and eating disorder symptoms is clinically important but could not be evaluated in this cross-sectional study. Finally, the retrospective nature of our study is subject to recall bias and limits our understanding of the temporal relationship between eating disorder and mood symptoms.

Our comparison with adolescents with GAD should be cautiously interpreted within the context of several limitations: (1) we used different diagnostic assessments for the BP I and GAD groups; the ADIS eating disorders questions are less specific and not well-matched with the WASH-U-KSADS; (2) based on our available data, we evaluated a large sample of adolescents with BP I relative to GAD, potentially challenging the validity of our group comparisons; (3) the GAD group was mostly female and the sample size was small, which limited our ability to determine if sex also contributed to our results; (4) investigating GAD alone does not consider the co-occurrence of eating disorder symptoms with other anxiety disorders, limiting the scope of our comparison.

Conclusions

To our knowledge, this is the first study to investigate a full range of eating disorder symptoms in adolescents with BP I. Our results add to the literature by documenting the clinical significance of eating disorder symptoms in BP I and extend adult and adolescent findings by characterizing the prevalence and correlates of eating disorder symptoms unique to adolescents with BP I. We found that eating disorder symptoms are highly comorbid in adolescents with BP I and are associated with more severe depression. Our findings underscore the importance of screening for eating disorder symptoms in adolescents with BP and formulating a comprehensive treatment approach that does not exacerbate the co-occurring condition. Prospective studies exploring eating disorders in adolescents with BP are warranted, as this population presents an important developmental window for improving outcomes for individuals with BP.

Clinical Significance

In adolescents with BP I, we found that comorbid eating disorder symptom are common and associated with more severe depression. Our findings underscore the clinical relevance of the presence of eating disorder symptom adolescent with BP I and inform the clinical approach to the assessment and treatment of this population.

Footnotes

Disclosures

Dr. Farrow's spouse is a consultant for Avid Radiopharmaceuticals. Dr. Strawn has received research support from Patient-Centered Outcomes Research Institute (PCORI), the National Institutes of Health, and the Yung Family Foundation. He has provided consultation to Cerevel, Intracellular Therapeutics, Otsuka, and Roche. He receives royalties from Springer Publishing, Cambridge University Press, and UpToDate and received material support from Myriad. He has also received honoraria from Medscape Live and the Neuroscience Education Institute. Dr. DelBello has received research support from the National Institute of Health, the PCORI, AbbVie, Alkermes, Janssen, Johnson and Johnson, Lundbeck, Myriad, Otsuka, Sage, Shire, Sunovion, Supernus, and Vanda and has provided consultation or advisory board services for Alkermes, Johnson and Johnson, and Sage. All other authors have no financial disclosures or conflicts of interest.

Acknowledgments

The authors thank the National Institute on Drug Abuse (Quetiapine Plus Topiramate or Placebo for Bipolar Mania and Cannabis Use in Adolescents; R01DA022221); National Institute of Health (Neurofunctional Predictors of Escitalopram Treatment Response in Adolescents with Anxiety; K23MH106037); National Institute of Mental Health (Multimodal Neuroimaging of Treatment Effects in Adolescent Mania; ROIMH083924); American Academy of Child and Adolescent Psychiatry 68th Annual Meeting (Poster Presentation); Prevalence and Correlates of Eating Disorder Symptoms in Adolescents with Bipolar Disorder; and the Warren Liang Medical Student Scholars Program Fund (W.Y.K.).

Authors' Contributions

J.E.F.: conceptualization, investigation, writing—original draft, writing—review and editing, visualization, and project administration; T.J.B.: methodology, formal analysis, data curation, writing—original draft; and writing—review and editing; W.Y.K.: investigation, data curation, writing—original draft, and writing—review and editing; K.E.H.: investigation, and data curation; J.R.S.: conceptualization, and writing—review and editing; M.P.D.: conceptualization; methodology; writing—review and editing; and supervision.

References

Álvarez Ruiz

, Gutiérrez-Rojas

. Comorbidity of bipolar disorder and eating disorders. Rev Psiquiatr Salud Ment, 2015; 8(4):232–241; doi: 10.1016/j.rpsm.2014.12.001

Balzafiore

, Rasgon

, Yuen

, et al. Lifetime eating disorder comorbidity associated with delayed depressive recovery in bipolar disorder. Int J Bipolar Disord, 2017; 5(25):1–9; doi: 10.1186/s40345-017-0094-4

Black Becker

, DeViva

, Zayfert

. Eating disorder symptoms among female anxiety disorder patients in clinical practice: The importance of anxiety comorbidity assessment. J Anxiety Disord, 2004; 18(3):255–274; doi: 10.1016/S0887-6185(03)00002-1

Brietzke

, Moreira

, Toniolo

, et al. Clinical correlates of eating disorder comorbidity in women with bipolar disorder type I. J Affect Disord, 2011; 130(1–2):162–165; doi: 10.1016/j.jad.2010.10.020

Bulik

. Anxiety disorders and eating disorders: A review of their relationship. N Z J Psychol, 1995; 24(2):51–62.

Fagiolini

, Frank

, Houck

, et al. Prevalence of obesity and weight change during treatment in patients with bipolar I disorder. J Clin Psychiatry, 2002; 63(6):528–533; doi: 10.4088/jcp.v63n0611

Fornaro

, Perugi

, Gabrielli

, et al. Lifetime co-morbidity with different subtypes of eating disorders in 148 females with bipolar disorders. J Affect Disord, 2010; 121(1–2):147–151; doi: 10.1016/j.jad.2009.06.007

Geller

, Zimerman

, Williams

, et al. Reliability of the Washington University in St. Louis Kiddie Schedule for Affective Disorders and Schizophrenia (WASH-U-KSADS) mania and rapid cycling sections. J Am Acad Child Adolesc Psychiatry, 2001; 40(4):450–455; doi: 10.1097/00004583-200104000-00014

Goldstein

, Blanco

, He

, et al. Correlates of overweight and obesity among adolescents with bipolar disorder in the National Comorbidity Survey-Adolescent Supplement (NCS-A). J Am Acad Child Adolesc Psychiatry, 2016; 55(12):1020–1026; doi: 10.1016/j.jaac.2016.08.010

10.

Hudson

, Hiripi

, Pope

Jr , et al. The prevalence and correlates of eating disorders in the National Comorbidity Survey Replication. Biol Psychiatry, 2007; 61(3):348–358; doi: 10.1016/j.biopsych.2006.03.040 Erratum in: Biol Psychiatry 2012; 72(2):164.

11.

Hughes

, Goldschmidt

, Labuschagne

, et al. Eating disorders with and without comorbid depression and anxiety: Similarities and differences in a clinical sample of children and adolescents. Eur Eat Disord Rev, 2013; 21(5):386–394; doi: 10.1002/erv.2234

12.

Jen

, Saunders

, Ornstein

, et al. Impulsivity, anxiety, and alcohol misuse in bipolar disorder comorbid with eating disorders. Int J Bipolar Disord, 2013; 1(13):1–9; doi: 10.1186/2194-7511-1-13

13.

Kenny

, Orellana

, Fuller-Tyszkiewicz

, et al. Depression and eating disorders in early adolescence: A network analysis approach. Int J Eat Disord, 2021; 54(12):2143–2154; doi: 10.1002/eat.23627

14.

Khoubaeva

, Dimick

, Roane

, et al. Comorbid eating disorders in a sample of youth with bipolar disorder: Elevated burden of dimensional and categorical psychopathology. J Clin Psychiatry, 2022; 83(4):21r14201; doi: 10.4088/JCP.21r14201

15.

Martin

, Williamson

, Thaw

. Criterion validity of the multiaxial assessment of eating disorders symptoms. Int J Eat Disord, 2000; 28(3):303–310; doi: 10.1002/1098-108x(200011)28:3<303:aid-eat7>3.0.co;2-i

16.

Martin

, Woo

, Timmins

, et al. Binge eating and emotional eating behaviors among adolescents and young adults with bipolar disorder. J Affect Disord, 2016; 88–95; doi: 10.1016/j.jad.2016.02.030

17.

McDonald

, Rossell

, Phillipou

. The comorbidity of eating disorders in bipolar disorder and associated clinical correlates characterised by emotion dysregulation and impulsivity: A systematic review. J Affect Disord, 2019; 228–243; doi: 10.1016/j.jad.2019.08.070

18.

McElroy

, Crow

, Biernacka

, et al. Clinical phenotype of bipolar disorder with comorbid binge eating disorder. J Affect Disord, 2013; 150(3):981–986; doi: 10.1016/j.jad.2013.05.024

19.

McElroy

, Crow

, Blom

, et al. Prevalence and correlates of DSM-5 eating disorders in patients with bipolar disorder. J Affect Disord, 2016a;191:216–221; doi: 10.1016/j.jad.2015.11.010

20.

McElroy

, Crow

, Blom

, et al. Clinical features of bipolar spectrum with binge eating behaviour. J Affect Disord, 2016b;201:95–98; doi: 10.1016/j.jad.2016.05.003

21.

McElroy

, Frye

, Hellemann

, et al. Prevalence and correlates of eating disorders in 875 patients with bipolar disorder. J Affect Disord, 2011; 128(3):191–198; doi: 10.1016/j.jad.2010.06.037

22.

McElroy

, Kotwal

, Keck

Jr , et al. Comorbidity of bipolar and eating disorders: Distinct or related disorders with shared dysregulations?. J Affect Disord, 2005; 86(2–3):107–127; doi: 10.1016/j.jad.2004.11.008

23.

Mohammadi

, Pourdehghan

, Mostafavi

, et al. Generalized anxiety disorder: Prevalence, predictors, and comorbidity in children and adolescents. J Anxiety Disord, 2020; 73:102234; doi: 10.1016/j.janxdis.2020.102234

24.

Olatunji

, Broman-Fulks

, Ciesielski

, et al. A taxometric investigation of the latent structure of eating disorders. Psychiatry Res, 2012; 197(1–2):97–102; doi: 10.1016/j.psychres.2011.12.016

25.

Poznanski

, Grossman

, Buchsbaum

, et al. Preliminary studies of the reliability and validity of the children's depression rating scale. J Am Acad Child Psychiatry, 1984; 23(2):191–197; doi: 10.1097/00004583-198403000-00011

26.

Puccio

, Fuller-Tyszkiewicz

, Ong

, et al. A systematic review and meta-analysis on the longitudinal relationship between eating pathology and depression. Int J Eat Disord, 2016; 49(5):439–454; doi: 10.1002/eat.22506

27.

Ramacciotti

, Paoli

, Marcacci

, et al. Relationship between bipolar illness and binge-eating disorders. Psychiatry Res, 2005; 135(2):165–170; doi: 10.1016/j.psychres.2004.04.014

28.

Reinblatt

. Are eating disorders related to attention deficit/hyperactivity disorder?. Curr Treat Options Psychiatry, 2015; 2(4):402–412; doi: 10.1007/s40501-015-0060-7

29.

Sahlan

, Williams

, Forrest

, et al. Disordered eating, self-esteem, and depression symptoms in Iranian adolescents and young adults: A network analysis. Int J Eat Disord, 2021; 54(2):132–147; doi: 10.1002/eat.23365

30.

Seixas

, Miranda-Scippa

, Nery-Fernandes

, et al. Prevalence and clinical impact of eating disorders in bipolar patients. Braz J Psychiatry, 2012; 34(1):66–70.

31.

Silverman

, Albano

. Anxiety Disorders Interview schedule for DSM-IV: Child version. Graywind Publication: New York, NY; Oxford University Press: New York; 1996.

32.

Swinbourne

, Hunt

, Abbott

, et al. The comorbidity between eating disorders and anxiety disorders: Prevalence in an eating disorder sample and anxiety disorder sample. Aust N Z J Psychiatry, 2012; 46(2):118–131; doi: 10.1177/0004867411432071

33.

Swinbourne

, Touyz

. The co-morbidity of eating disorders and anxiety disorders: A review. Eur Eat Disord Rev, 2007; 15(4):253–274; doi: 10.1002/erv.784

34.

Thiebaut

, Godart

, Radon

, et al. Crossed prevalence results between subtypes of eating disorder and bipolar disorder: A systematic review of the literature. Encephale, 2019a;45(1):60–73; doi: 10.1016/j.encep.2018.06.001

35.

Thiebaut

, Jaussent

, Maimoun

, et al. Impact of bipolar disorder on eating disorders severity in real-life settings. J Affect Disord, 2019b;246:867–872; doi: 10.1016/j.jad.2018.12.128

36.

Ward

, Rodriguez

, Wright

, et al. Estimation of eating disorders prevalence by age and associations with mortality in a simulated nationally representative US cohort. JAMA Netw Open, 2019; 2(10):e1912925; doi: 10.1001/jamanetworkopen.2019.12925

37.

Watson

, Hoiles

, Egan

, et al. Normative data for female adolescents with eating disorders on the Multidimensional Anxiety Scale for Children. Int J Eat Disord, 2014; 47(5):471–474; doi: 10.1002/eat.22252

38.

Wildes

, Marcus

, Fagiolini

. Eating disorders and illness burden in patients with bipolar spectrum disorders. Compr Psychiatry, 2007; 48(6):516–521; doi: 10.1016/j.comppsych.2007.05.012

39.

Wildes

, Marcus

, Fagiolini

. Prevalence and correlates of eating disorder co-morbidity in patients with bipolar disorder. Psychiatry Res, 2008; 161(1):51–58; doi: 10.1016/j.psychres.2007.09.003

40.

Yaramala

, McElroy

, Geske

, et al. The impact of binge eating behavior on lithium- and quetiapine-associated changes in body weight, body mass index, and waist circumference during 6 months of treatment: Findings from the bipolar CHOICE study. J Affect Disord, 2020; 266:772–781; doi: 10.1016/j.jad.2018.09.025

41.

Young

, Biggs

, Ziegler

, et al. A rating scale for mania: Reliability, validity and sensitivity. Br J Psychiatry, 1978; 133:429–435; doi: 10.1192/bjp.133.5.429