Purpose of review Research in eating disorders in males has been active lately compared to the past. This review aims to provide an overview of the recently published studies of eating disorders in males.
Recent findings Publication of the Diagnostic and Statistical Manual of Mental Disorders, 5th edition has outlined more sexneutral diagnostic criteria for eating disorders. Data of socioeconomic factors, prenatal influences, clinical characteristics, assessment, and mortality for eating disorders have been reported independently for males. Unlike in females, higher parental education showed no association with eating disorders in males, but twin or triplet status and lower gestational age at birth had an independent association with anorexia nervosa in males. Contrary to earlier suggestions, no differences in eating disorder symptoms such as binging, vomiting, or laxative abuse were observed between the sexes. Yet, males tended to score lower on eating disorder symptom measures than females. High rates of premorbid overweight and higher BMIs at various stages of eating disorders have been confirmed repeatedly. Higher age and lower BMI at admission, and restrictive anorexia nervosa subtype predicted fatal outcome for anorexia nervosa in males.
Summary Contemporary research provides grounds for improved recognition, diagnosis, and treatment for males suffering from eating disorders. Introduction Eating disorders in males are no longer a niche topic. An important milestone was the revised diagnostic classification Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM5) published in 2013. The diagnostic criteria for anorexia nervosa and bulimia nervosa were broadened, and binge eating disorder (BED) was included as a new diagnostic entity. This has particular relevance for males, whose eating disorder symptoms tend not to fit into stringent categories. However, there is still much room for improvement in treatment, recognition, prevention, and reducing stigma of eating disorders among males.
This review aims to provide an overview of the recently published studies on eating disorders in males. To review the literature, we searched for articles using PubMed and Ovid MEDLINE(R), using the terms 'eating disorders', 'anorexia nervosa', 'bulimia nervosa', 'binge eating disorder', 'male', and 'men'.
Eating disorders have a highly skewed sex distribution. The lifetime prevalence of anorexia nervosa is 0.16–0.3%, of bulimia nervosa 0.1–0.5%, and of BED 1.1–3.1% among men. The rate ratio of lifetime prevalence of anorexia nervosa and bulimia nervosa in males vs. females is often reported to be equal or less than 1 : 10, but the figures likely reflect clinical underdetection among males. Epidemiological studies report higher variation: for anorexia nervosa, 1 : 3–1 : 12, and for bulimia nervosa, 1 : 3–1 : 18. In BED, the male to female rate ratio is more equal, 1 : 2–1 : 6.
Eating disorders in males differ somewhat from females in risk factors, clinical presentation, comorbidity, and consequences. Homosexual orientation associates with an increased risk of eating disorders in men.Premorbid overweight is common compared to females and when studying body dissatisfaction in men, it is important to distinguish weight and muscularity concerns as both are common in males, and share overlapping features related to vulnerability and behaviors.Athletic achievement and excessive exercise appear crucial features among men, and the rates of psychiatric comorbidity are higher among men than among women. Regarding consequences of eating disorders, males may be subjected to double stigma═¥ the stigma of a psychiatric disorder, and an additional stigma of suffering from what is commonly perceived to be a femalespecific disorder.
Below, we will discuss recent studies on eating disorders in males, emphasizing the most relevant articles published within the last 18 months.
Compared with the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSMIV), DSM5 allows more eating disorders in males to be identified by a specific diagnosis, instead of a residual category such as 'eating disorder not otherwise specified' EDNOS). As an example, the elimination of the amenorrhea requirement and more permissive wording of the weight criterion might facilitate anorexia nervosa diagnosis in males.
The proportion of EDNOS cases according to DSMIV criteria appears to be even higher among males than females: among adolescent males, 83.5% and among adult men, 83.3% of all eating disorders identified in a population study were classified as EDNOS. Lifetime prevalences of EDNOS in male adolescents and adults were 3.9 and 3.4%, respectively. Among adolescents, the prevalence of DSMIV EDNOS was 1.0% at 14 years, 0.9% at 17 years, and 2.2% at 20 years.
The prevalence of 'any DSM5 eating disorder' in males was recently reported to be 1.2% at 14 years, 2.6% at 17 years, and 2.9% at 20 years. The prevalence of DSM5 bulimia nervosa was 0.4%, 0.7%, and 1.6%, respectively. In a community cohort study of adolescents,the lifetime prevalence of any DSM5 eating disorder was 1.2% among male adolescents compared to 5.7% among female adolescents═¥ BED was the most common diagnosis in the male adolescents. Overall, males make up a substantial proportion of individuals with BED. Thus, its recognition as a specific eating disorder in DSM5 will increase the number of men diagnosed with an eating disorder. The lowered frequency for binge episodes required for a diagnosis of BED in DSM5 has been shown to have only a minimal effect on lifetime prevalence of BED in adult males.
The DSMIV residual category (EDNOS) has been renamed 'other specified feeding or eating disorder' (OSFED) in the DSM5. The OSFED specifies five disorders, including atypical anorexia nervosa, in which all criteria for anorexia nervosa are met except being underweight, and night eating syndrome. Males make up a substantial proportion of individuals with night eating syndrome. In DSM5, muscle dysmorphia is included as a specifier in body dysmorphic disorder. Its symptoms, such as body image disturbance, disordered eating, comorbid personality traits, and anxiety/affective disorders exhibit overlap of eating disorders, and nosological similarity with anorexia nervosa has been suggested.
In hospitalized males with severe anorexia nervosa, the age of onset was higher (20.8 vs. 18.1 years), and lifetime maximum and minimum BMIs, BMIs at admission, as well as desired BMIs were higher compared to females with the same condition. The duration of hospitalization for anorexia nervosa in males was shorter compared to females, and male patients had fewer suicide attempts. Males reported less somatization, obsessivecompulsivity, interpersonal sensitivity, and anxiety, but no differences in binging, vomiting, laxative and diet pill use, or substance abuse emerged. A cohort study of adolescent male patients with DSMIV anorexia nervosa, bulimia nervosa, or EDNOS reported a mean age of 14.7 years at diagnosis and a mean age of onset of 13.4 years═¥ these are largely in line with those found among females. The most commonly reported symptoms were restriction and excessive exercise═¥ only a minority experienced binging and purging.
Two recent studies have focused on the assessment of eating disorder symptoms in males. In the Eating Disorders Examination (EDE12), the global, shape, and weight concern scores in adolescent male patients were significantly lower compared to female patients, with effect sizes in the lowtomoderate range. The authors concluded that the use of Eating Disorders Examination is acceptable in adolescent males, but certain questions may reflect symptoms that are atypical for males═¥ current scoring ranges derived from female samples bore little resemblance to those of males with similar clinical presentations. A similar trend, men scoring lower, has been observed on other measures. The Eating Disorder Assessment for Men total score predicted correctly an eating disorder in 82.1% of male patients. The components derived from the factor analysis were binge eating, muscle dysmorphia, body dissatisfaction, and disordered eating.
A recent qualitative study of 10 young men with a history of anorexia nervosa and/or bulimia nervosa suggested that a perception of eating disorders as female problems led to an initial failure to recognize males' behaviors as symptoms of an eating disorder. Men iscussed the lack of genderappropriate information and resources for men with eating disorders as an additional impediment, and they felt that health and other professionals had been slow to recognize their symptoms because they were men. The authors concluded that raising awareness of eating disorders more widely in society is crucial to help men to recognize and seek help before their symptoms become intractable. They further suggested that it is important to decouple the experience and management of eating disorders from feminized cultural imagery, resources, and clinical practice, and prevent professionals from overlooking signs and symptoms in males that they may readily recognize as indicative of eating disorders in young women.
In a large Swedish registrybased study, higher maternal or paternal education was not associated with eating disorders in male patients, contrary to females. This was explained by potential lower parental weight concerns toward boys, which may target less easily to sons compared to daughters in families with high socioeconomic position. Another explanation relates to previously shown higher vulnerability load in males who develop eating disorders═¥ hence, their eating disorder symptoms are also more likely to reflect global psychological impairment. Twin or triplet status and lower gestational age at birth predicted anorexia nervosa in both sexes, but mother's higher prepregnancy weight did not predict lower risk of anorexia nervosa in males, as in females.
Among military veterans with trauma histories, the lifetime prevalence of DSMIV bulimia nervosa and DSMIV BED in males was 0.49% (N = 2) and 3.7% (N = 15). Posttraumatic stress disorder and depression severity were most consistently associated with bulimia nervosa and BED symptoms, and 45% of male veterans reported one or more current symptoms of bulimia nervosa or BED.
In patients with severe anorexia nervosa, longterm survival did not differ between the sexes, but males died sooner after discharge than females. Three deaths over the followup of 9.8 years occurred among males═¥ higher age and lower BMI at admission, and restrictive anorexia nervosa subtype predicted fatal outcome. The standardized mortality ratio for anorexia nervosa was 8.1 (95% CI 1.6–23.6) in men═¥ for the restrictive anorexia nervosa subgroup, it was 13.2 (95% CI 2.7–38.6). Another recent study of mortality reported an elevated risk ratio of 3.6 (95% CI 1.4–9.4) for allcause mortality for composite of anorexia nervosa, bulimia nervosa, and BED in male patients compared to matched population controls. There was no significant difference to females, although more males (7.1%═¥ N = 8, mortality rate of 9.10 per 1000 personyears) died than females (2.3%═¥ N = 53, mortality rate of 2.65 per 1000 personyears) during the followup═¥ this was explained by the higher overall mortality rate in the male general population.
For the first time to our knowledge, data from large samples covering mortality, socioeconomic factors, and prenatal influences in relation to eating disorders have been reported independently for males. In women, increased mortality in bulimia nervosa, BED, and EDNOS has been shown, and likewise, research should extend to examine mortality specifically in other eating disorders beyond anorexia nervosa also in males. It has been consistently shown that symptoms and risk factors for eating disorders vary by sex, and for example the lack of an effect for higher parental education on the risk of eating disorders in sons adds to this information.
So far, sexspecific differences have not been reflected in the eating disorders instruments, almost all of which have been developed and validated in females. This is a significant problem, which likely contributes to underdiagnosis and undertreatment in males. In the future, it is imperative to develop instruments for assessment of eating disorder symptoms in boys and men. DSM5 has recently outlined more sexneutral diagnostic criteria. Its publication provides opportunities for improved recognition and diagnosis for eating disorders among boys and men, but the implementation remains in the hands of clinicians diagnosing and treating male patients.