Open Collections

UBC Faculty Research and Publications

Eating disorders in biological males: clinical presentation and consideration of sex differences in a… Coelho, Jennifer S; Lee, Tiffany; Karnabi, Priscilla; Burns, Alex; Marshall, Sheila; Geller, Josie; Lam, Pei-Yoong Nov 26, 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-40337_2018_Article_226.pdf [ 433.57kB ]
JSON: 52383-1.0374302.json
JSON-LD: 52383-1.0374302-ld.json
RDF/XML (Pretty): 52383-1.0374302-rdf.xml
RDF/JSON: 52383-1.0374302-rdf.json
Turtle: 52383-1.0374302-turtle.txt
N-Triples: 52383-1.0374302-rdf-ntriples.txt
Original Record: 52383-1.0374302-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessEating disorders in biological males: clinicalpresentation and consideration of sexdifferences in a pediatric sampleJennifer S. Coelho1,2* , Tiffany Lee1, Priscilla Karnabi1, Alex Burns1, Sheila Marshall3,4, Josie Geller2,5and Pei-Yoong Lam1,4AbstractBackground: The growing body of research on eating disorders among male adolescents reveals some sexdifferences in clinical presentation. The current study set out to replicate and extend recent research on the clinicaland medical characteristics of male youth with eating disorders, and examine sex differences between biologicalmales and females in a tertiary pediatric eating disorder treatment setting.Methods: A retrospective chart review was conducted with all biological males who were admitted to the EatingDisorders Programs at British Columbia Children’s Hospital (2003–2015) or the Looking Glass Residence (2011–2015).Clinical data, including demographics, percentage of median body mass index (% mBMI), and psychiatric diagnoses,were recorded along with medical data (i.e., vital signs, basic biochemistry investigations, and bone mineraldensity). A comparison group of females with eating disorders who received treatment at British ColumbiaChildren’s Hospital in the inpatient or outpatient streams (2010–2015) were included, to examine sex differenceswith males who were admitted during the same period.Results: A total of 71 male youth were included in the chart review. Males had significant medical complications,with 26.5% of the sample presenting with a heart rate of less than 50 beats per minute and 31.4% presenting witha bone mineral density z-score for the lumbar spine ≤ − 1. Sex differences between the subset of males who weretreated between 2010 and 2015 (n = 41) and the females (n = 251) were examined. Females were more likely thanwere males to have a diagnosis of anorexia nervosa or bulimia nervosa, and to be underweight (< 95% mBMI) atadmission. Males were younger than females, but no differences emerged in the duration of the eating disordersymptoms. No sex differences emerged relating to medical instability (e.g., bradycardia).Conclusions: A large proportion of male children and youth with eating disorders are medically compromised atadmission. Males were younger than females, and were less likely than females to have a diagnosis of anorexianervosa or bulimia nervosa. Males who were underweight at admission had also lost a lower percentage of bodyweight in comparison to females. The current study replicates previous sex differences reported in pediatricsamples.Keywords: Males, Pediatric, Eating disorders, Anorexia nervosa, Sex differences* Correspondence: jennifer.coelho@cw.bc.ca1Provincial Specialized Eating Disorders Program for Children & Adolescents,British Columbia Children’s Hospital, Box 178, 4500 Oak St., Vancouver, BCV6H 3N1, Canada2Department of Psychiatry, University of British Columbia, Vancouver, BC,CanadaFull list of author information is available at the end of the article© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (, which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver( applies to the data made available in this article, unless otherwise stated.Coelho et al. Journal of Eating Disorders            (2018) 6:40 English SummaryThere is limited research on the presentation of maleswith eating disorders. The overwhelming majority of in-dividuals presenting to specialized eating disorder treat-ment centers are female; however, there are concernsthat eating disorders in males are underdiagnosed andmisunderstood. The current study set out to examinethe clinical characteristics of a large sample of maleswho presented for treatment between 2003-2015 at aspecialized treatment program for eating disorders. Wecompared a group of females who had received treat-ment at this program between 2010-2105 with maleswho were admitted during the same period. The maleswho were treated during this period had significant med-ical complications, with approximately one quarter ofthe sample presenting with very low heart rate and onethird presenting with a low bone mineral density. Sexdifferences among adolescents with eating disorders alsoemerged: males were younger than females, and had ayounger age of onset than did females. Males were alsoless likely than females to have a diagnosis of anorexianervosa or bulimia nervosa. The current diagnostic clas-sification system for eating disorders may not capturesome of the eating- and weight-related concerns experi-enced by males, given the larger proportion of atypicaland other eating disorder diagnoses in males.BackgroundThe published literature on eating disorder presentationin adolescent males has been growing over the past sev-eral years. In 2012, a special issue on males and eatingdisorders was published in Eating Disorders: The Journalof Treatment and Prevention. Male eating disorders werehighlighted to be “under-diagnosed, undertreated, andmisunderstood” by many clinicians (p. 346), [1]. A smallbody of literature on the clinical characteristics of maleswith eating disorders had been established over the pastseveral decades, prior to the publication of this specialissue. In one of the largest samples of males with eatingdisorders (135 adult males), bulimia nervosa was themost common diagnosis, with reports of long delays inseeking treatment [2]. Delays in referrals of males witheating disorder symptoms to specialist services were alsohighlighted in a sample of adolescents [3].In contrast to the prevalence of bulimia nervosa in thelarge adult male sample reported by Carlat and col-leagues [2], anorexia nervosa (AN) was the most com-mon diagnosis in a sample of adolescent males, followedby atypical eating disorder diagnoses [4]. In line with thispresentation, a sample of 10 males with AN between theages of 9–22 who were treated in a tertiary hospital set-ting all presented with very low weight (less than 80% ofsuggested body weight) [5]. In contrast, in a small sam-ple of males with an early onset eating disorder (i.e.,onset at or before age 13), the majority of patients didnot meet criteria for AN despite significant weight lossand medical instability [6]. A recent review suggests thatthe current diagnostic classification scheme for fullthreshold eating disorders, including AN and bulimianervosa (BN), may have less applicability for males,given the inclusion of the pursuit of thinness as a prom-inent feature of the diagnostic criteria [7].Muscularity-related concerns are one of the purportedfeatures of eating disorder concerns in males, which arenot currently accounted for in the diagnostic criteria foreating disorders [7]. Muscularity concerns emerged asone of four symptom classes of eating disorder symptompatterns (along with binge-eating/purging, body image,and a mostly asymptomatic group with some muscular-ity concerns) in a large sample of male youth [8]. Thehigh prevalence of excessive exercise that has been re-ported in males with eating disorders (e.g., [9]) may re-flect the high level of muscularity concerns in somemales.Younger, pre-adolescent patients who present to eatingdisorder treatment services are more likely to be malethan are older adolescents [10, 11]. One of the chal-lenges of interpreting these results are that males appearto be more reluctant to seek treatment than females,due to perceived stigma about seeking help [12]. Thehigher rates of younger males in treatment that havebeen reported in the literature may reflect a decrease inhelp-seeking behaviour for older males with eating disor-ders (rather than a difference in prevalence of eating dis-orders across children and adolescents). The higher rateof younger male patients, however, suggests a particularrelevance of studying male eating disorders in pediatricsettings. One of the largest pediatric samples of malesstudied to date examined the medical and psychologicalcharacteristics of male adolescents with eating disorders[13]. Norris and colleagues [13] described a total of 52male adolescents with eating disorders between the agesof 10 to 17, who had received treatment at a tertiary eat-ing disorders program in Canada. The most commondiagnosis in this group was Eating Disorder Not Other-wise Specified (EDNOS; 52% of the sample), followed byAN and BN. Food restriction was reported by the over-whelming majority of the sample, with over-exercise,binge episodes, vomiting and laxative use reported by asubset of the males. A notable proportion of males forwhom bone mineral density z-scores were available werefound to have osteopenia (31%) or osteoporosis (8%).Several male-specific studies have recently emerged,which further examined the medical and clinical charac-teristics of males with eating disorders. Vo and col-leagues [14] reported on a sample of 33 male outpatients(ages 11–25), with a focus on medical characteristics. Ofnote, a large proportion (51.5%) of the sample metCoelho et al. Journal of Eating Disorders            (2018) 6:40 Page 2 of 12criteria for admission for medical stabilization due tolow heart rate and large orthostatic shifts in heart rate(more than 20 beats per minute) at presentation. Al-though males presented with a mean of 88% medianbody mass index (mBMI), there was an average weightloss of 20% of initial body weight. Male adolescents witheating disorders present at a higher percentage of theirsuggested body weight than females [15]. Similarly, adultmales with AN have a higher body mass index (BMI) atadmission than do females [16]. There have been criti-cisms of the use of BMI in examining sex differences inanthropometric measures, due to sex- and age-relateddifferences in body composition (i.e., body fat versusmuscle mass) [17]. Nagata and colleagues have employeddual-energy x-ray absorptiometry (DXA) scans to exam-ine sex differences in body composition, and report thatthere are deficits in fat mass and lean body mass in bothadolescent males and females [18]. However, there arenot currently guidelines recommending the use of DXAscans to evaluate body composition in males [18].Woodside and Kaplan [19] report that the clinicalpresentation of males and females who were admitted toa day treatment program was very similar, and thatmales could be effectively treated in a group settingcomprised predominantly of females. However, somemixed findings are emerging with respect to sex differ-ences in the clinical presentation of males and females.Several studies suggest no sex differences in admissionage (e.g., [15, 20–25]), while one study with an adoles-cent sample reported females were older than males[26]. Adult males have been reported to have a later ageof onset [16, 27], whereas studies with adolescent sam-ples report a younger age of onset in males [26, 28].Across studies, some conflicting findings between thepediatric and adult literature have therefore emergedwith respect to age at admission and age of onset. Withthe exception of the two studies on sex differences in ad-olescents [26, 28] all studies were conducted prior to thepublication of the 5th edition of the Diagnostic and Stat-istical Manual (DSM-5) [29] and therefore do not con-sider sex or age-related differences in the prevalence ofavoidant/restrictive food intake disorder (ARFID). Someof the inconsistencies across the literature can poten-tially be attributed to low power due to small samplesizes; however, there are a growing number of largestudies (sample sizes greater than 50 males; (e.g., [20–23, 26, 28]).Further discrepancies in the literature arise when con-sidering eating disorder diagnostic presentation. Someresearchers have reported no sex differences in eatingdisorder diagnostic categories [22, 24], while others havereported that males are less likely than are females topresent with a diagnosis of AN or BN or more likely topresent with an atypical or other eating disorder (e.g.,[26, 28]). Children and adolescents who were diagnosedwith ARFID were also more likely to be male in com-parison to those diagnosed with AN or BN [30]. Incon-sistencies also arise in examination of sex differences insymptom presentation. While some studies report nodifferences in eating disorder symptom presentation [26,27], others report that males have lower levels of laxativeabuse [20, 31] and vomiting [28] than do females witheating disorders. Males have been reported to be morelikely to have over-exercise as an eating disorder symp-tom than are females (e.g., [3]), while a more recentstudy with a larger sample size reported no sex differ-ences in the prevalence of excessive exercise [28]. Re-searchers have reported the absence of sex differences inmedical instability (e.g., bradycardia and hypotension)(e.g., [28]).In addition to developmental considerations of pa-tient samples, another possible factor that could ac-count for differences in clinical presentation acrossstudies is the intensity of the treatment setting.Whereas some studies investigating presentation ofmales have focused on individuals presenting to anoutpatient clinic (e.g., [14, 26]), other studies havebeen conducted at centers that offer a continuum ofcare (including outpatient, day treatment, and in-patient services) (e.g., [28]). A report from our grouprevealed that among a sample of male inpatients, thepredominant diagnosis was AN (restrictive orbinge-eating/purging subtype), with 82.6% of thesample receiving this diagnosis [32]. The high preva-lence of AN reported by Coelho and colleagues [32]exceeds that reported by Norris and colleagues [13],in which 42% of males were diagnosed with AN.Similarly, Shu and colleagues [28] reported lowerlevels of AN diagnoses in their sample, at 34% ofmales (and 38.1% of females). The fact that Coelhoand colleagues [32] focused exclusively on an in-patient sample may account for some of the differ-ences in the prevalence of AN across studies. Overthe past decade there has been an increasing focusof outpatient services as the first-line interventionfor eating disorders, including family-based treatment[33]. Furthermore, some of the youth who have beenreferred to treatment in a pediatric eating disorderssetting may have lost weight and be experiencing sig-nificant medical complications associated with eatingdisorders, but may not yet meet criteria for AN.Atypical eating disorders may be more common inthose who are admitted to lower intensity treatmentfor eating disorders (i.e., outpatient services), as theseindividuals may be more medically stable or have lostless weight than those admitted to a higher level ofcare. Individuals with other eating disorders may alsobe more likely to be admitted to outpatient services.Coelho et al. Journal of Eating Disorders            (2018) 6:40 Page 3 of 12For example, those with a diagnosis of ARFID areless likely than those with AN to present with lowheart rate and hypotension [34]. The current studytherefore set out to examine diagnostic differencesacross males who were treated in outpatient versusintensive (day treatment/inpatient) services.Aims and objectivesThe goal of the current research is to replicate and ex-tend research on the clinical features of male youth witheating disorders, and examine sex differences inpediatric eating disorders. A retrospective chart reviewwas conducted to achieve this goal. We endeavored toextend the report by Vo and colleagues [14] on clinicaland medical characteristics of male youth. We hypothe-sized that males admitted to outpatient would be morelikely than those admitted to a more intensive treatment(i.e., day treatment and/or inpatient) to have an eatingdisorder diagnosis other than AN or BN. We further setout to replicate previous reports of sex differences inmale and female adolescents [26, 28]. We therefore ex-pected a higher prevalence of diagnoses other than ANor BN in males than in females, and a younger age ofonset than females. Given the diversity of youth present-ing to our service, we also set out to examine whethermales would be more likely to come from an ethnic mi-nority group than would females, as reported by Kinaszand colleagues [26]. We also conducted exploratory ana-lyses to assess sex differences on key medical and clinicalcharacteristics, including presence of eating disordersymptoms, medical stability (bradycardia, orthostaticshift, and bone mineral density z-score), andweight-related presentation at admission (i.e., % mBMIat admission, and percentage of weight loss at presenta-tion). Bradycardia and orthostatic shift were classifiedbased on admission criteria published by the Society forAdolescent Health and Medicine [35].MethodsParticipantsInclusion criteria for the study were any biologicalmales (based on assigned sex at birth) who wereassessed and admitted to treatment at either BritishColumbia (BC) Children’s Hospital (Provincial Special-ized Eating Disorders Program for Children and Ado-lescents) between January 2003 and July 20151 or theLooking Glass Residence (formerly Woodstone Resi-dence) between April 2011 (when the residenceopened) and July 2015. Data from the first admissionduring the study period were recorded. Admission tothe program was defined as attending at least one med-ical or therapy appointment (for outpatients), or admis-sion to one of the more intensive treatment programs(i.e., having at least 1 day of admission to day program,inpatient, or residential treatment).BC Children’s Hospital offers outpatient treatment, aday treatment program, and inpatient treatment. Youth(and their families) attend an initial assessment in theprogram, and the level of care (outpatient, day treatmentor inpatient) is recommended by the multi-disciplinaryassessment team, in collaboration with youth and theirfamily, after evaluation of medical stability, symptom se-verity, and other characteristics of the patient and family(including past treatment history and availability of par-ents/caregivers to provide meal support). The assess-ment team includes a pediatrician, nurse, and either apsychiatrist or psychologist. BC Children’s Hospital isthe only tertiary hospital-based program in the province.Patient and family readiness for treatment and recoveryare considered when evaluating an appropriate level ofcare at assessment; however, some youth may be certi-fied for involuntary treatment given the potential forpermanent medical consequences of eating disorders.Admission criteria for outpatient treatment (e.g.,family-based treatment) include medical and psychiatricstability, willingness to engage in treatment, and avail-ability of parent(s) or caregiver(s) to provide meal sup-port. Admission criteria for day treatment includesmedical stability (i.e., not meeting hospital admission cri-teria [35]), no acute safety-related concerns, ability andwillingness of youth and family to engage in treatment,significant life interference, and need for tertiary levelservices. Inpatient treatment would be recommended forindividuals who have failed to make sufficient progressin a lower level of care, and/or whose symptoms requireintensive treatment. Youth who are not medically orpsychiatrically stable may be admitted to a medical orpsychiatric bed for stabilization prior to transfer to theinpatient Eating Disorders Program. Given the wide geo-graphical area served by the hospital setting, inpatienttreatment may be recommended for children and ado-lescents who do not have eating disorders services intheir local community, and whose family are not able tosupport their child to attend a lower level of care.The Looking Glass Residence offers treatment for indi-viduals with eating disorders between ages 16–24 in aresidential setting, with 24-h support provided by amulti-disciplinary team of health professionals (includingpsychiatrists, nursing, allied health professionals, and anurse practitioner). Admission criteria include medicaland psychiatric stability. Looking Glass Residence is avoluntary recovery-focused treatment program, whichoffers an intensive 12-week program of structured ther-apy in a home-like setting. Details of this program areavailable in from the authors, and are detailed in: Wil-liams KD, O’Reilly C, Coelho JS: Residential treatmentfor eating disorders in a Canadian treatment Centre:Coelho et al. Journal of Eating Disorders            (2018) 6:40 Page 4 of 12clinical characteristics and treatment experiences of resi-dents, submitted.Information regarding participants’ eating disorderand other psychiatric diagnoses was recorded fromdocumentation by staff psychiatrists or psychologists, inaccordance with the 4th edition of the Diagnostic andStatistical Manual of Mental Disorders (DSM-IV-TR)[36] criteria, or DSM-5 [29] criteria for those admittedafter May 2013. A retrospective review of the charts ofbiological males who met inclusion criteria was under-taken. A standardized data extraction manual was usedto ensure consistency among members of the researchteam in data entry and checking. Study data were col-lected and managed using REDCap (Research ElectronicData Capture) tools hosted at BC Children’s HospitalResearch Institute. REDCap is a secure, web-based appli-cation designed to support data capture for researchstudies [37].To facilitate an analysis of sex differences, a compari-son group of biological females (based on assigned sexat birth) who were admitted to treatment was analyzed.Given the small number of males admitted to the Look-ing Glass Residence during the study period, and the dif-ferences in admission criteria between hospital-basedand residential treatment, the comparison group waslimited to those who were admitted to the Eating Disor-ders Program at BC Children’s Hospital. Data from thefirst admission of biological females with eating disor-ders who were admitted to the inpatient unit (March2010–June 20152) or outpatient family-based therapy(January 2010 – July 20153) was available as part ofother on-going research in the Eating Disorders Pro-gram, and was included as a convenience sample for thefemale comparison group.Procedures in this study were reviewed and approvedby the Research Ethics Board at the Children’s andWomen’s Health Center of British Columbia (see Decla-rations section for details).ResultsClinical and demographic characteristics of male youthCharts were available for 71 male youth, 68 of whomhad their first admission at BC Children’s Hospital dur-ing this period, and 3 who were admitted to the LookingGlass Residence. One male who was first admitted to BCChildren’s Hospital had a subsequent admission to theLooking Glass Residence during the study period; how-ever, only data from the first admission were analyzed.Pairwise deletion of data was performed, in which casesthat did not have data for a particular variable were ex-cluded from that analysis. The sample size for each ana-lysis is indicated for variables for which there wasmissing data.Male youth had an average age of 14.8 years (SD = 2.8,range = 9–24 years). Due to a non-normal distributionfor the duration of eating disorder symptoms, medianand interquartile range (IQR) were evaluated. Durationof eating disorder symptoms was calculated based onthe number of months that had elapsed between the firstreported onset of eating disorder symptoms and the dateof assessment. For cases in which a range in the timeframe was given for the onset of symptoms (e.g., a sea-son, or school grade), the midpoint of the time periodwas recorded. The median duration of symptoms priorto admission was 11 months (IQR = 18.0, range = 2–148;N = 69). Participants were from a diverse ethnic back-ground, including: Caucasian (n = 27, 38.0%), Asian (in-cluding South Asian, West Asian, and Chinesebackgrounds; n = 15, 21.1%), mixed background (n = 5,7.0%), Latin American (n = 2, 2.9%), Aboriginal (n = 1,1.4%), and Pacific Islander (n = 1, 1.4%). Information onethnic background was not available for 20 participants(28.2%). One of the males reported questioning his gen-der identity at assessment, but did not have a formaldiagnosis of gender dysphoria at admission. Demo-graphic information from the sample is presented inTable 1.For the youth who received treatment at BC Children’sHospital, the majority were first admitted to outpatienttreatment (n = 35, 51.5%), 28 (41.2%) were admitted toinpatient treatment, and 5 (7.4%) were admitted to daytreatment. Several youth who were admitted to the out-patient stream had shared care with other treatmentprograms at BC Children’s Hospital (i.e., an inpatientmental health admission for child or adolescent psych-iatry, n = 5, 7.3%), or with other community-based men-tal health teams (n = 2, 2.9%) for support withco-occurring mental health concerns. A majority of themales (n = 44, 62.0%) had received previous treatmentfor an eating disorder prior to admission to the EatingDisorders Program (see Table 1 for details). Previoustreatment was reported at assessment based on parentor youth report and/or information included in the re-ferral, and was defined as admission for treatment of aneating disorder to: a community hospital, specialized eat-ing disorder inpatient or day treatment program, or out-patient services for an eating disorder (e.g., secondaryservice programs or private therapist providing eatingdisorder treatment). A subgroup of the 68 youth whowere admitted to BC Children’s Hospital had beentreated in another department immediately prior to theirtransfer to the eating disorders program, including amedical stabilization admission (n = 12, 17.6%), admis-sion to inpatient psychiatry or psychiatric emergency(e.g., for concerns about psychiatric safety or suicidalityn = 5, 7.3%), or a visit to emergency without subsequentadmission (n = 1, 1.1%).Coelho et al. Journal of Eating Disorders            (2018) 6:40 Page 5 of 12The most common eating disorder diagnosis was AN(restrictive subtype, n = 27, 38%; binge-eating/purgingsubtype, n = 9, 12.7%), followed by EDNOS/other speci-fied feeding or eating disorder (OSFED) with primarysymptoms of restriction (n = 20, 28.2%), EDNOS/OSFEDwith symptoms of binge-eating and/or purging (n = 3,4.2%), and BN (n = 2, 2.8%). Three males (4.2%) met thecriteria for ARFID. Additional diagnoses included failureto thrive (n = 1, 1.4%) and food avoidance emotional dis-order (n = 3, 4.2%) under DSM-IV-TR [36], with symp-toms that would likely meet DSM-5 [29] criteria forARFID. Finally, three males (4.2%) reported symptomsof disordered eating that did not meet full criteria for aneating disorder diagnosis (or were on a differential diag-nosis with EDNOS/OSFED) but were causing significantimpairment in well-being or functioning, and thereforerequired specialized treatment.To examine the hypothesis that males who receivedoutpatient treatment would be more likely than those inan intensive treatment to present with a diagnosis otherthan AN or BN, eating disorder diagnoses were catego-rized into one of three groups: (1) diagnosis of AN orBN; (2) atypical AN or BN (including EDNOS with re-strictive or binge-eating/purging symptoms); and (3)other eating disorder diagnoses (e.g., unspecified eatingdisorder, purging disorder, ARFID). We selected onlythose who were admitted to BC Children’s Hospital(given the limited sample size of those who receivedresidential treatment, and the differences with residentialtreatment admission criteria, participants from theLooking Glass Residence were excluded). There was asignificant difference in diagnostic classification acrosstreatment intensity (outpatient versus day treatment/in-patient), χ2 (2, N = 68) = 17.05, p ≤ .001. Z-tests (withBonferroni correction applied) to assess differences incolumn proportion suggested that there were fewermales in outpatient with a diagnosis of AN or BN (n =10, 28.6%) than in the inpatient or day treatment pro-grams (n = 25, 75.8%). There were more individuals withan “other” eating disorder diagnosis in outpatient (n =11, 31.4%) than in inpatient/day treatment (n = 1, 3.0%).No differences in treatment intensity emerged for thosewith atypical eating disorders who were treated in out-patient (n = 14, 40.0%) versus inpatient/day treatment (n= 7, 21.2%).Eating disorder symptoms reported at assessment (byeither the youth or the parent) were recorded. Eatingdisorder symptoms were marked as present if there wasa lifetime history of the symptom. Restriction was themost prevalent eating disorder symptom reported at as-sessment, with nearly all of the males (n = 69, 97.2%)restricting their intake. Excessive exercise was also acommon symptom, with 70.4% (n = 50) of the youth pre-senting with symptoms of excessive exercise. ExcessiveTable 1 Demographic and presenting characteristics of male youthLiving Arrangements (n = 71)Living with both parents (biological or adoptive/fosterparents)45(63.4%)Living with mother (biological or adoptive/foster parent) 16(22.5%)Living equally in two homes (parents or other caregivers) 5 (7.0%)Living with mother and a step-parent 3 (4.2%)Living with other caregivers (e.g., grandparents) 2 (2.8%)Prior Eating Disorder Treatment (n = 71)Community hospital admission (medical stabilization) 27(38.0%)Outpatient eating disorder treatment 29(40.8%)Inpatient admission to a specialized eating disordersprogram1 (1.4%)Day treatment/residential treatment admission 0 (0%)Certification for Involuntary Treatment (n = 66)Voluntary admission 54(81.8%)Certification for part or all of admission 12(18.2%)Vegetarianism (n = 66)Partial Vegetarian (most meals are vegetarian) 3 (4.5%)Full vegetarian 5 (7.6%)Vegan 1 (1.5%)Not Vegetarian 57(86.4%)Self-harm and Suicidality (n = 71)Current self-harm behaviours 5 (7.0%)Past history of self-harm behaviours 5 (7.0%)Previous suicide attempt 5 (7.0%)History of abusePhysical abuse reported at assessment (n = 69) 4 (5.8%)Sexual abuse reported at assessment (n = 68) 1 (1.5%)Co-occurring psychiatric diagnoses at assessment (n = 71)Major depressive disorder 10(14.1%)Other mood disorder (including not otherwise specified/unspecified, and persistent depressive disorder/dysthymia)2 (2.8%)Generalized anxiety disorder 3 (4.2%)Social anxiety disorder 3 (4.2%)Panic disorder 2 (2.8%)Other anxiety disorder (including not otherwise specified/unspecified anxiety disorder, selective mutism, andspecific phobia)11(15.5%)Obsessive-compulsive disorder 3 (4.2%)Attention-deficit/hyperactivity disorder 3 (4.2%)Autism spectrum disorder 2 (2.8%)Substance-related disorder 1 (1.4%)Coelho et al. Journal of Eating Disorders            (2018) 6:40 Page 6 of 12exercise was defined as activity that was aimed at con-trolling weight or shape, and that was significant in itsduration, frequency and/or intensity (including exercis-ing despite injury), and/or interferes with activities, as inthe DSM-5 ([29], p., 346). A subgroup of males reportedregular episodes of binge-eating (n = 20, 28.2%) and pur-ging through vomiting (n = 14, 19.7%) and/or laxativeuse (n = 2, 2.8%).The percentage of median BMI (% mBMI) was calcu-lated for youth ages 19 and under (n = 68), according toreference values from the World Health Organization[38]. At admission, the mean percentage mBMI of youthwas 87.5% (SD = 12.2). The majority of males wereunderweight (< 95% mBMI) at admission (n = 51, 75% ofsample). There was a significant increase in weight overthe course of admission, t(45) = 9.6, p ≤ .001, d = 1.3;Madmission = 82.2, SD = 7.4; Mdischarge = 94.3, SD = 10.3).Length of treatment was also investigated. Due to anon-normal distribution of length of treatment, medianand IQR were investigated. The median length of treat-ment for the 61 males who completed treatment duringthe study period and for whom discharge data was avail-able, was 90 days (IQR = 124, range = 1–661 days).A large proportion of the sample (n = 47, 66.2%) hadsymptoms of a co-occurring psychiatric disorder at as-sessment, and received either a full diagnosis of aco-occurring psychiatric disorder, or symptoms of aco-occurring psychiatric disorder were present but wereon a differential diagnosis and therefore were notassigned (for example, due to malnutrition at the time ofassessment). For details of the diagnoses for which par-ticipants met full criteria, see Table 1.Biochemistry results of the male sample are presentedin Table 2. Approximately one third of the sample forwhom transaminases were available (alanine aminotrans-ferase; n = 46) presented with levels that were out ofrange. A portion of the sample (13.3%) for whom testos-terone levels were available (n = 30) had levels that werebelow the expected range. Medical characteristics of themale youth are presented in Table 3. Indices of medicalcompromise included heart rate, with 26.5% of the sam-ple (n = 68) presenting with bradycardia (i.e., a heart rateof below 50 beats per minute), and 33.3% (20 of samplesize of 60) presenting with significant orthostatic shift inheart rate (an increase of more than 20 beats per mi-nute). A portion of the sample (n = 35) had bone mineraldensity results available, with 31.4% of the sample pre-senting with z scores equal to or less than − 1 for thelumbar spine and 34.3% for hip z-scores.Sex differences: Youth treated at BC Children’s HospitalData from a subset of males who entered treatment afterJanuary 2010 were compared with data from biologicalfemales who were admitted to the inpatient oroutpatient treatment streams at BC Children’s Hospitalduring the study period (2010–2015). A total of 41 malesand 251 females were included in the dataset. In the fe-male comparison group, 58 were treated in an outpatientsetting (23.1% of female sample), and 193 were treatedin the inpatient program (76.9% of female sample). Inthe male sample, 48.8% (n = 20) were initially treated inan outpatient setting, 12.2% (n = 5) in day treatment,and 39.0% (n = 16) in the inpatient program.Due to unequal sample sizes and a violation of nor-mality and/or homogeneity of variance assumptions,non-parametric tests were employed to assess sex differ-ences in age at admission, duration of eating disordersymptoms, and age of onset. Male youth were signifi-cantly younger at admission than females (Mann-Whit-ney U = 3989.0, z = − 2.31, p = .021, r = − 0.14). Themedian age of males at admission was 14.9 years (IQR =3.7, range = 9–17), and that of females was 15.5 years(IQR = 2.3, range = 9–18). No significant difference be-tween males and females emerged for the duration ofeating disorder symptoms at the time of admission(Mann-Whitney U = 4704.0, z = − 0.24, p = .81, r = −0.01). Males reported a median duration of eating dis-order symptoms of 11.0 months (IQR = 8.5, range = 2–96 months; n = 40) prior to admission, while females re-ported a median duration of 10.0 months (IQR = 12.0,range = 0–84 months; n = 241). The age of onset of eat-ing disorder symptoms was calculated by subtracting theage at assessment by the duration of eating disordersymptoms. Data for the age of onset variable were nor-mally distributed; however, there was a violation ofhomogeneity of variance. Given the large discrepancy insample size between the two groups, non-parametrictests were performed, which demonstrated a significantsex difference, Mann-Whitney U = 2927.5, z = − 1.98, p= .047, r = − 0.13. Males had a younger age of onset (M= 13.1 years, SD = 2.7; n = 40) than did females (M =14.2 years, SD = 1.6; n = 183). Females were more likelythan males to come from a Caucasian ethnic back-ground, χ2 (1, N = 243) = 11.36, p ≤ .001, odds ratio = 3.6(CI = 1.7, 8.0). See Table 4 for details of sex differences.No significant differences in lifetime history ofbinge-eating (p = .20), vomiting (p = .56), laxative use (p= .27), or excessive exercise emerged (p = .55). Examin-ation of sex differences in diagnostic presentation wereperformed, with youth classified into one of 3 groups:(1) diagnosis of AN or BN; (2) atypical AN or BN (in-cluding EDNOS with restrictive or binge-eating/purgingsymptoms); and (3) other eating disorder diagnoses (e.g.,unspecified eating disorder, purging disorder, ARFID).Significant sex differences in diagnostic presentationemerged, χ2 (2, N = 292) = 18.91, p ≤ .001. Z-tests wereused to compare column proportions, with a Bonferronicorrection applied. Females were more likely than malesCoelho et al. Journal of Eating Disorders            (2018) 6:40 Page 7 of 12to be diagnosed with AN or BN, whereas males weremore likely to be diagnosed with an atypical eating dis-order, or another eating disorder diagnosis. See Table 4for details.A subset of medical parameters was selected for as-sessment of sex differences, including heart rate andbone mineral density (see Table 4). The medical stabilityof males and females was similar, with no significant dif-ferences between sexes on the proportion of youth whomet criteria for bradycardia (less than 50 beats per mi-nute; χ2 (1, N = 225) = 0.07, p = .79), large orthostaticshift (greater than 20 beats per minute; χ2 (1, N = 210) =0.11, p = .74), nor were there differences in the propor-tion of youth who had a bone mineral density z-scorefor the lumbar spine that was equal to or less than − 1,χ2 (1, N = 224) = 1.79, p = .18.Females were more likely than were males to beunderweight [less than 95% of mBMI; χ2 (1, N = 286) =Table 2 Laboratory findings for male youthMeasure of Central Tendencya,Sample sizeRange % Out of Range(reference range)Metabolic panelSodium (mmol/L) 141 (IQR = 3), 59 135–146 1.7% (135–145)Potassium (mmol/L) 4.3 (SD = 0.42), 59 3.4–5.2 5.1% (3.5–5.0)Chloride (mmol/L) 102 (IQR = 3), 50 95–111 4.0% (95–107)Magnesium (mmol/L) 0.86 (SD = 0.08), 55 0.65–1.05 9.1% (0.74–0.99)Phosphate (mmol/L) 1.37 (SD = 0.26), 56 0.22–1.95 23.2% (0.87–1.52)TransaminasesAspartate aminotransferase (AST; U/L) 31 (IQR = 14), 44 15–112 25.0% (10–40)Alanine aminotransferase (ALT; U/L) 27.5 (IQR = 23.75), 46 0–241 32.6% (10–45)Cholesterol PanelCholesterol (mmol/L) 3.65 (SD = 0.89), 24 1.3–5.2 16.7% (2.6–5.2)Triglycerides (mmol/L) 0.70 (IQR = 0.72), 23 0.24–2.12 26.1% (0.4–1.5)HormonesTestosterone (nmol/L) 6.1 (IQR = 14)b, 30 0.4–30.1b 13.3% (< 0.6)aMean (Standard Deviation) is reported for variables that were normally distributed. Median (Interquartile Range) is reported for variables with anon-normal distributionb3 males had undetectable testosterone levels. The values presented are for those for whom hormone levels could be recorded by the laboratory (n = 27);however, the column for the percentage of participants for whom the values were out of range included those who had undetectable testosterone levelsTable 3 Medical characteristics of male youthMeasure of Central Tendencya,Sample sizeRange % Out of Range(n; Reference Range)Heart RateBeats per minute (bpm, supine) 61.6 (SD = 18.5), 68 28 – 108 26.5% (18; < 50 bpm)Orthostatic Shift (standing - lying) 16.5 (IQR = 15), 60 -6 – 87 33.3% (20; > 20 bpm)Blood PressureSystolic blood pressure (supine) 105.4 (SD = 12.8), 68 79 – 140 8.8% (6; < 90 mmHg)Orthostatic shift in systolic pressureb -1.54 (SD = 7.5), 63 -18 – 17 0% (0; > 20 mmHg drop)Diastolic blood pressure (supine) 60.43 (SD = 8.6), 68 43 – 82 2.9% (2; < 45 mmHg)Orthostatic shift in diastolic pressureb -4.89 (SD = 7.4), 63 -24 – 11 22.2% (14; > 10 mmHg drop)Bone Mineral Density (BMD) Reference Range:BMD z-score≤ -1Spine z-score -0.45 (SD = 1.0), 35 -2.4 – 1.9 31.4% (n = 11)Hip z-score -0.6 (SD = 1.2), 35 -3.0 – 2.1 34.3% (n = 12)Total body z-score -0.29 (SD = 1.2), 35 -3.2 – 2.1 28.6% (n = 10)aMean (Standard Deviation) is reported for variables that were normally distributed. Median (Interquartile Range) is reported for variables with anon-normal distributionbValues for orthostatic shift in blood pressure are presented as lying-standing to examine postural drop. Reference range for heart rate and orthostatic shifts inpulse and blood pressure are based on admission criteria recommended by the Society for Adolescent Medicine [35]Coelho et al. Journal of Eating Disorders            (2018) 6:40 Page 8 of 127.63, p = .006, odds ratio = 2.8 (CI = 1.3, 5.9)], with 85.7%of females (n = 210) and 68.3% of males (n = 28) classi-fied as underweight at admission. Given a non-normaldistribution of data for this variable, non-parametrictests were performed. Females who were underweighthad a lower %mBMI at admission (Median = 78.0, IQR =12.1) than did males (Median = 83.9, IQR = 11.0),Mann-Whitney U = 2020.0, z = − 2.69, p = .007, r = −0.17). Females also had lost a larger percentage of weight[(maximum weight-admission weight)/maximum weight*100] than males, F(1,252) = 4.67, p = .032, partial η2= .018. Females lost an average of 20.3% of their max-imum body weight (SD = 10.1, CI = 18.9, 21.7; n = 217),whereas males lost an average of 16.3% (SD = 11.9, CI =12.9, 19.7; n = 37).DiscussionThe males who presented to our service demonstratedsignificant medical compromise, despite being only mod-erately underweight (with an average of 87.5% mBMI).Vo and colleagues [14] reported a similar clinical presen-tation of males, with a mean of 88% mBMI in their sam-ple. As with Vo and colleagues, a portion of the samplepresented with indices of medical instability, includingvery low heart rate (26.5%) and a large orthostatic shiftin heart rate (33.3%). Bone mineral density results wereavailable for a portion of the males included in the chartreview. Approximately one-third of males presented withbone mineral density z-scores equal to or less than − 1(with 31.4% presenting with low z-scores for the lumbarspine and 34.3% with low z-scores for the hip), which issimilar to the report by Norris and colleagues [13] that39% of males had osteopenia or osteoporosis based onz-scores for the lumbar spine. Similarly, previous reportsindicated that 35% of male adolescents with AN pre-sented with results consistent with a diagnosis of osteo-penia [39]. Testosterone levels were low in 13% of thesample, including 3 males for whom testosterone wasundetectable. Previous reports have highlighted diversityin endocrinopathies in a case series of adult males withAN [40]. Testosterone levels in males with AN havebeen demonstrated to rise with weight gain, though therate of increase in testosterone levels varied across indi-viduals [41]. Given that testosterone values were avail-able for only 42% of the total sample (n = 30) in thecurrent study, it was not possible to evaluate the rela-tionship between testosterone levels and % mBMI. Oneof the factors driving the lack of availability of hormonelevels (and other biochemistry investigations) for a por-tion of the sample was that a standardized process forbiochemistry investigations was not in place in the Eat-ing Disorders Program until 2011, and was also not yetin place at Woodstone (now known as Looking GlassResidence) at the time of assessment for the maleresidents.The current results extended previous findings re-ported by Kinasz and colleagues [26] and Shu and col-leagues [28], in demonstrating a higher prevalence ofatypical and other eating disorder diagnoses in malesthan in females. Males in the current study were alsoTable 4 Sex differences in male and female youth who were treated at BC Children’s Hospital between 2010 and 2015Males, Sample Size Females, Sample SizeDemographics n = 30 n = 213Ethnicity: Caucasian* 12 (40.0%) 151 (70.9%)Ethnic Minority Group (includes Asian, East Indian, Aboriginal, mixed)* 18 (60.0%) 62 (29.1%)Eating Disorder Diagnosis n = 41 n = 251Anorexia Nervosa or Bulimia Nervosa* 23 (56.1%) 208 (82.9%)Atypical Anorexia Nervosa or Bulimia Nervosa* 10 (24.4%) 32 (12.7%)Other Eating Disorder Diagnosis* 8 (19.5%) 11 (4.4%)Eating Disorder Symptoms (Lifetime History)Binge-eating 13 (31.7%), 41 55 (22.4%), 245Vomiting 9 (22.0%), 41 64 (26.2%), 244Laxative Use 2 (4.9%), 41 25 (10.3%), 243Excessive Exercise 28 (68.3%), 41 177 (72.8%), 243Medical CharacteristicsBradycardia (heart rate < 50 bpm) 9 (22.0%), 41 44 (23.9%), 184Orthostatic shift in heart rate (> 20 bpm) 12 (32.4%), 37 61 (35.3%), 173Bone Mineral Density – Spine z-score (≤ −1) 5 (22.7%), 22 75 (37.1%), 202Note: This table does not capture all variables used for sex comparisons. Additional comparisons of variables not included in this table (i.e., treatment settings, ageat admission, age at eating disorder onset, duration of eating disorder symptoms, and weight) are detailed in the results section*denotes significant sex difference, p < .05Coelho et al. Journal of Eating Disorders            (2018) 6:40 Page 9 of 12less likely to be diagnosed with AN or BN. Kinasz andcolleagues [26] have suggested that the diagnosis of another specified eating disorder may represent a “catchall” category for males (p. 415). The higher prevalence ofatypical eating disorder diagnoses may be accounted forin part by the lower percentage of weight loss in malesobserved in the current study. Males were also morelikely to come from an ethnic minority group than werefemales, further substantiating the report by Kinasz andcolleagues [26]. The diversity in ethnic background inour sample of males is reflective of the population in thearea served by the hospital, with 51.8% of the populationin Vancouver belonging to a visible minority group (and27.3% of individuals in the province of British Columbiabelonging to a visible minority) [42]. Also supporting thehypotheses was the younger age of onset in the malesample; however, no significant differences emerged inthe duration of eating disorder symptoms.The medical stability of male and female youth wassimilar, which mirrors previous findings [28]. There wereno sex differences in the proportion of males and fe-males who had an abnormal z-score for the lumbarspine from the bone mineral density assessment. Past re-search has suggested that males with AN have more pro-nounced bone loss than do females [22]; however, thisresearch included adult males in the sample. We also ex-amined sex differences in body weight at presentation.Given the higher proportion of females who were under-weight at admission, we limited this analysis to those whowere underweight. As in previous research (e.g., [15]),males presented at a higher % mBMI than did females.The diagnostic distribution of males included a por-tion of individuals who did not meet full criteria for aneating disorder, as well as a significant minority of thesample who met criteria for EDNOS/OSFED. Maleswere less likely than were females to be diagnosed withAN or BN. Previous studies with male adolescents havereported mixed results. Some studies report that atypicaldiagnoses are the most common diagnoses of males pre-senting for eating disorder treatment (e.g., [6, 13]), whileothers report that anorexia nervosa is the most commondiagnosis (e.g., [4]). A previous report on a subsample ofmales treated in our inpatient eating disorders programalso found AN to be the most common diagnosis [32].We hypothesized that the presentation of males with aneating disorder diagnosis other than AN or BN may bemore common in outpatient programs, and the resultsof this study partially support this prediction. Males witha diagnosis of AN or BN were more likely to be treatedin intensive (inpatient/day treatment) services, whilethose with an other eating disorder diagnosis were morelikely to receive treatment in outpatient services. Therewere no differences in the proportion of males with adiagnosis of atypical AN or BN diagnosis across theintensive and outpatient services. Given the differencesin diagnostic presentation across services, it is importantto consider the intensity of treatment when comparingthe clinical characteristics across samples of males witheating disorders. This also suggests that the results re-garding the more common prevalence of atypical andother eating disorder diagnoses in males need to beinterpreted with caution, given that the majority (76.9%)of the female sample was from an inpatient treatmentsetting, whereas only 39% of the 41 males who were in-cluded in the analyses on sex differences were initiallytreated in the inpatient setting.One of the limitations of this study is that the datasetof females does not represent all the individuals whowere treated in the program over the time period, butrather a convenience sample of a portion of females(treated in outpatient family-based therapy or inpatient)for whom data were available. Further prospective re-search investigating sex and gender differences acrosstreatment intensity is warranted to clarify whether theincreased prevalence of atypical eating disorder diagno-ses in males, as reported in Shu et al. [28], are foundacross the continuum of care.There were several variables for which there werenon-normal distributions of the data, both in clinical pres-entation (e.g., duration of eating disorder symptoms) aswell as medical characteristics (e.g., testosterone levels,and measures within the metabolic panel). Therefore,there appears to be a heterogeneity in males who presentto a tertiary pediatric eating disorder service. Thenon-normal distributions led to use of non-parametrictests to detect sex differences, which have a lower relativepower than would a parametric test (assuming the under-lying assumptions of the test can be met) [43]. Further-more, due to relatively small sample sizes across thediagnoses, we were not able to examine differences acrosseach of the eating disorder subtypes. Furthermore, giventhe long study period, there were changes in eating dis-order diagnostic criteria from DSM-IV-TR [36] to DSM-5[29]. We grouped together diagnostic presentations (e.g.,considering whether those with an EDNOS diagnosis pre-sented with symptoms of restriction or binge-eating/pur-ging), to facilitate comparison of diagnoses across thestudy period. However, due to a lack of detail in the fre-quency and duration of symptoms noted in the older as-sessment notes that were examined as part of thisretrospective chart review, it was not possible to reclassifyDSM-IV-TR diagnoses into DSM-5 diagnoses. It is there-fore not possible to confirm whether some of the otherdiagnoses reported in the sample (e.g., food avoidanceemotional disorder) would now meet criteria for ARFID.Given the report by Vo and colleagues [14] that there wasa decrease in the prevalence of EDNOS/OSFED diagnosiswhen applying DSM-5 criteria, it will be important toCoelho et al. Journal of Eating Disorders            (2018) 6:40 Page 10 of 12continue investigating the pattern of diagnostic presenta-tion in adolescent males.Strengths of the current study include the large samplesize of 71 male children and youth for the descriptiveanalyses, the wide range of physical assessments, and thedetailed psychiatric assessments that were obtained aspart of clinical practice. Self-report tools to assess eatingdisorder and related symptoms were not included in thecurrent study. Males have reported less severe eatingpathology than do females (e.g., [26, 28]), as measuredby the Eating Disorder Examination [44]. However, themost common measures in the field of eating disordershave been developed to assess female concerns, and rec-ommendations have been put forth by Darcy and Lin[45] regarding tools that include male-specific concerns,including muscularity. A recent review by Murray andcolleagues [7] highlighted the lack of focus on muscular-ity concerns in current diagnostic criteria for eating dis-orders, and suggests that changes may be necessary tothe conceptualization of eating disorder pathology to bemore inclusive across the gender spectrum. Our grouphas undertaken a prospective study of all biologicalmales who are admitted to our service (along with agroup of matched females), to follow-up on some of theobserved sex differences in pediatric eating disorderpresentation. We have included validated measures de-signed to assess concerns that may be more relevant formales, such as muscularity, in accordance with the rec-ommendations of Darcy and Lin [45].ConclusionSome consistent sex differences in pediatric eating disor-ders appear to be emerging, including males having ayounger age of presentation, younger age of onset, ahigher body weight at admission (as measured by%mBMI), and a lower prevalence of AN or BN eatingdisorder diagnoses. There appear to be significant med-ical sequelae in male children and youth with eating dis-orders, despite the smaller percentage of weight lossrelative to females for those who were underweight atadmission.Endnotes1This group of participants represents an expansion ofa previous sample of male youth (n = 23, who were ad-mitted to inpatient treatment) reported in Coelho et al.,2015 [32]2A portion of this sample (females with AN who weretreated between 2010 and 2014) were included in a re-port by Janzen and colleagues (Janzen M, Cheung C,Steinberg C, Lam PY, Krahn A: Changes on the electro-cardiogram in anorexia nervosa: a case control study,submitted). The current sample was expanded to includefemales who were admitted to the inpatient unit (March2010 – June 2015) across all eating disorder diagnoses3A portion of this sample was included in a reportby Coelho and colleagues (Coelho JS, Beach B, O’BrienK, Marshall S, Lam PY: Effectiveness of family-basedtreatment for pediatric eating disorders in a tertiary caresetting, submitted). Only biological females who wereadmitted during the study period (2010–2015) were in-cluded in the analyses for the current studyAbbreviations% mBMI: Percentage median body mass index; AN: Anorexia Nervosa;ARFID: Avoidant/Restrictive Food Intake Disorder; BC: British Columbia;BMI: Body Mass Index; BN: Bulimia Nervosa; CI: Confidence Interval;DSM: Diagnostic and Statistical Manual of Mental Disorders; EDNOS: EatingDisorder Not Otherwise Specified; IQR: Interquartile Range; OSFED: OtherSpecified Feeding or Eating Disorder; REDCap: Research Electronic DataCapture; SD: Standard DeviationAcknowledgementsWe would like to thank Katie Coopersmith, Mona Maleki and MacKenzieRobertson for their assistance with data entry for this project. Portions of thisdata have been presented at the Eating Disorders Association of Canada bi-annual meeting (Winnipeg, 2016), and the Eating Disorders Research Societyannual meeting (Leipzig, 2017).FundingThis research was supported by a grant from the Swiss Anorexia NervosaFoundation (Project # 40–14), awarded to the first author. The funding bodydid not have a role in the final design of the study, the analysis, nor theinterpretation of data or the writing of the manuscript.Availability of data and materialsThe datasets analyzed during the current study are available from thecorresponding author on reasonable request.Authors’ contributionsJSC designed the study, oversaw data collection, analyzed and interpretedthe results, and prepared the manuscript. TL, PK, and AB contributed to thedesign of the study database, collected and entered data, and providedassistance with manuscript preparation. SM, JG, and PYL were involved instudy design, interpretation of the results, and manuscript preparation. Allauthors read and approved the final manuscript.Ethics approval and consent to participateEthics approval was obtained from the Research Ethics Board at theChildren’s and Women’s Health Center of British Columbia (study H14–00136). Given the retrospective nature of the study, a waiver of participantconsent was granted.Consent for publicationNot applicable.Competing interestsJennifer S. Coelho and Josie Geller are members of the editorial board of theJournal of Eating Disorders. The other authors declare that they have nocompeting interests.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1Provincial Specialized Eating Disorders Program for Children & Adolescents,British Columbia Children’s Hospital, Box 178, 4500 Oak St., Vancouver, BCV6H 3N1, Canada. 2Department of Psychiatry, University of British Columbia,Vancouver, BC, Canada. 3School of Social Work, University of BritishColumbia, Vancouver, BC, Canada. 4Division of Adolescent Health & Medicine,Coelho et al. Journal of Eating Disorders            (2018) 6:40 Page 11 of 12Department of Pediatrics, University of British Columbia, Vancouver, BC,Canada. 5Eating Disorders Program, St. Paul’s Hospital, Vancouver, BC,Canada.Received: 16 August 2018 Accepted: 24 October 2018References1. Strother E, Lemberg R, Stanford SC, Turberville D. Eating disorders in men:underdiagnosed, undertreated, and misunderstood. Eat Disord. 2012;20(5):346–55.2. Carlat DJ, Camargo CA Jr, Herzog DB. Eating disorders in males: a report on135 patients. Am J Psychiatry. 1997;154(8):1127.3. Margo JL. Anorexia nervosa in males: a comparison with female patients. BrJ Psychiatry. 1987;151(1):80–3.4. Eliot AO, Baker CW. Eating disordered adolescent males. Adolescence. 2001;36(143):535.5. Siegel JH, Hardoff D, Golden NH, Shenker IR. Medical complications in maleadolescents with anorexia nervosa. J Adolesc Health. 1995;16(6):448–53.6. Bayes A, Madden S. Early onset eating disorders in male adolescents: aseries of 10 inpatients. Australasian Psychiatry. 2011;19(6):526–30.7. Murray SB, Nagata JM, Griffiths S, Calzo JP, Brown TA, Mitchison D, BlashillAJ, Mond JM. The enigma of male eating disorders: a critical review andsynthesis. Clin Psychol Rev. 2017;57:1–1.8. Calzo JP, Horton NJ, Sonneville KR, Swanson SA, Crosby RD, Micali N, EddyKT, Field AE. Male eating disorder symptom patterns and health correlatesfrom 13 to 26 years of age. J Am Acad Child Adolesc Psychiatry. 2016;55(8):693–700.9. Touyz SW, Kopec-Schrader EM, Beumont PJ. Anorexia nervosa in males: areport of 12 cases. Aust N Z J Psychiatry. 1993;27(3):512–7.10. Peebles R, Wilson JL, Lock JD. How do children with eating disorders differfrom adolescents with eating disorders at initial evaluation? J AdolescHealth. 2006;39(6):800–5.11. Walker T, Watson HJ, Leach DJ, McCormack J, Tobias K, Hamilton MJ, ForbesDA. Comparative study of children and adolescents referred for eatingdisorder treatment at a specialist tertiary setting. Int J Eat Disord. 2014;47(1):47–53.12. Griffiths S, Mond JM, Li Z, Gunatilake S, Murray SB, Sheffield J, Touyz S. Self-stigma of seeking treatment and being male predict an increased likelihoodof having an undiagnosed eating disorder. Int J Eat Disord. 2015;48(6):775–8.13. Norris ML, Apsimon M, Harrison M, Obeid N, Buchholz A, Henderson KA,Spettigue W. An examination of medical and psychological morbidity inadolescent males with eating disorders. Eat Disord. 2012;20(5):405–15.14. Vo M, Lau J, Rubinstein M. Eating disorders in adolescent and young adultmales: presenting characteristics. J Adolesc Health. 2016;59(4):397–400.15. Geist R, Heinmaa M, Katzman D, Stephens D. A comparison of male andfemale adolescents referred to an eating disorder program. Can J Psychiatry.1999;44(4):374–8.16. Gueguen J, Godart N, Chambry J, Brun-Eberentz A, Foulon C, Snezana M,Guelfi JD, Rouillon F, Falissard B, Huas C. Severe anorexia nervosa in men:comparison with severe AN in women and analysis of mortality. Int J EatDisord. 2012;45(4):537–45.17. Gallagher D, Visser M, Sepulveda D, Pierson RN, Harris T, Heymsfield SB.How useful is body mass index for comparison of body fatness across age,sex, and ethnic groups? Am J Epidemiol. 1996;143(3):228–39.18. Nagata JM, Golden NH, Peebles R, Long J, Murray SB, Leonard MB, CarlsonJL. Assessment of sex differences in body composition among adolescentswith anorexia nervosa. J Adolesc Health. 2017;60(4):455–9.19. Woodside DB, Kaplan AS. Day hospital treatment in males with eatingdisorders—response and comparison to females. J Psychosom Res. 1994;38(5):471–5.20. Button E, Aldridge S, Palmer R. Males assessed by a specialized adult eatingdisorders service: patterns over time and comparisons with females. Int JEat Disord. 2008;41(8):758–61.21. Crisp A. Collaborators. 1.5. Anorexia nervosa in males: similarities and differencesto anorexia nervosa in females. Eur Eat Disord Rev. 2006;14(3):163–7.22. Mehler PS, Sabel AL, Watson T, Andersen AE. High risk of osteoporosis inmale patients with eating disorders. Int J Eat Disord. 2008;41(7):666–72.23. Welch E, Ghaderi A, Swenne I. A comparison of clinical characteristicsbetween adolescent males and females with eating disorders. BMCPsychiatry. 2015;15(1):45.24. Weltzin T, Bean P, Klosterman E, Lee HJ, Welk-Richards R. Sex differences inthe effects of residential treatment on the quality of life of eating disorderpatients. Eating Weight Disorders. 2015;20:301–10.25. Striegel-Moore RH, Leslie D, Petrill SA, Garvin V, Rosenheck RA. One-year useand cost of inpatient and outpatient services among female and malepatients with an eating disorder: evidence from a national database ofhealth insurance claims. Int J Eat Disord. 2000;27(4):381–9.26. Kinasz K, Accurso EC, Kass AE, Le Grange D. Does sex matter in the clinicalpresentation of eating disorders in youth? J Adolesc Health. 2016;58(4):410–6.27. Bramon-Bosch E, Troop NA, Treasure JL. Eating disorders in males: acomparison with female patients. Eur Eat Disord Rev. 2000;8(4):321–8.28. Shu CY, Limburg K, Harris C, McCormack J, Hoiles KJ, Hamilton MJ, WatsonHJ. Clinical presentation of eating disorders in young males at a tertiarysetting. J Eat Disord. 2015;3(1):39.29. American Psychiatric Association. Diagnostic and statistical manual ofmental disorders (DSM-5). Amer Psychiatr Pub. 2013.30. Fisher MM, Rosen DS, Ornstein RM, Mammel KA, Katzman DK, Rome ES,Callahan ST, Malizio J, Kearney S, Walsh BT. Characteristics of avoidant/restrictive food intake disorder in children and adolescents: a “new disorder”in DSM-5. J Adolesc Health. 2014;55(1):49–52.31. Núñez-Navarro A, Agüera Z, Krug I, Jiménez-Murcia S, Sánchez I, Araguz N,Gorwood P, Granero R, Penelo E, Karwautz A, Moragas L. Do men witheating disorders differ from women in clinics, psychopathology andpersonality? Eur Eat Disord Rev. 2012;20(1):23–31.32. Coelho JS, Kumar A, Kilvert M, Kunkel L, Lam PY. Male youth with eatingdisorders: clinical and medical characteristics of a sample of inpatients. EatDisord. 2015;23(5):455–61.33. Lock J, Le Grange D. Treatment manual for anorexia nervosa: a family-basedapproach: Guilford Publications; 2015.34. Brigham KS, Manzo LD, Eddy KT, Thomas JJ. Evaluation and treatment ofavoidant/restrictive food intake disorder (ARFID) in adolescents. Curr PediatrRep. 2018;6(2):107–13.35. The Society for Adolescents Health and Medicine. Position paper of thesociety for adolescent health and medicine: medical management ofrestrictive eating disorders in adolescents and young adults references. JAdolesc Health. 2015;56(1):121–5.36. American Psychiatric Association, American Psychiatric Association. DSM-IV-TR: Diagnostic and statistical manual of mental disorders, text revision, vol.75. Washington, DC: American Psychiatric Association; 2000. p. 78–85.37. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Researchelectronic data capture (REDCap)—a metadata-driven methodology andworkflow process for providing translational research informatics support. JBiomed Inform. 2009;42(2):377–81.38. Onis MD, Onyango AW, Borghi E, Siyam A, Nishida C, Siekmann J.Development of WHO growth reference for school-aged children andadolescents. Bull World Health Organ. 2007;85:660–7.39. Castro J, Toro J, Lazaro L, Pons F, Halperin I. Bone mineral density in maleadolescents with anorexia nervosa. J Am Acad Child Adolesc Psychiatry.2002;41(5):613–8.40. Skolnick A, Schulman RC, Galindo RJ, Mechanick JI. The endocrinopathies ofmale anorexia nervosa: case series. AACE Clin Case Reports. 2016;2(4):e351–7.41. Wheeler MJ, Crisp AH, Hsu LK, Chen CN. Reproductive hormone changesduring weight gain in male anorectics. Clin Endocrinol. 1983;18(4):423–9.42. Statistics Canada (2013). National Household Survey Focus on geographyseries, National Household Survey year 2011. Website Accessed 26 July2018: Accessed 31 Oct 2018.43. Field A. Discovering statistics using SPSS. London: Sage publications Ltd; 2009.44. Fairburn CG, Cooper Z, O'Connor M. Eating disorder examination (16.0D). In:Fairburn CG, editor. Cognitive Behavior Therapy and Eating Disorders.45. Darcy AM, Lin IH. Are we asking the right questions? A review ofassessment of males with eating disorders. Eat Disord. 2012;20(5):416–26.Coelho et al. Journal of Eating Disorders            (2018) 6:40 Page 12 of 12


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items