Male eating disorders in midlife—possible links between excessive sports and hormones

Abstract

Although eating disorders were long considered a typical female disorder, it is now clear that men are also affected. However, the literature on eating disorders in men is still very limited, and the actual extent is not known. Even less is known about the epidemiology of eating disorders in older individuals. In this focused review, we will present an update of the available data on disordered eating and eating disorders in middle-aged and older males. In addition, we will highlight the relationship of eating disorders with excessive sports as a purging method of choice for this age group and discuss the impact of age-related hormonal imbalances in aging men.

Keywords:

• Eating disorders
• male
• testosterone
• age
• excessive sports

Introduction

In recent decades, the field of eating disorder (ED) research has experienced a dramatic shift in their diagnostic orientation. While classically EDs were viewed as typical female-centric psychiatric disorders that only affected very few and atypical males [1], this view has been revised toward a wider spectrum of ED manifestations affecting both sexes. Still, males have been and continue to be underrepresented in the ED literature [2]. With this broader view on EDs, also the actual extent of societal costs based on ED-related treatments is better reflected. Not only do EDs constitute serious public health problems that besides strong impact on the affected individuals also burden families and social support systems, but they also cause tremendous health care costs [3].

Until now, only few studies have assessed the prevalence of EDs in the general population, with considerable differences between various geographic regions [4]. However, female : male ratios were reported to lie between 3–12 : 1 for anorexia nervosa (AN), 3–18 : 1 for bulimia nervosa (BN), and 2–6 : 1 for binge eating disorder (BED), whereas the proportion of EDs not otherwise specified (EDNOS) was even higher for males as compared to females [5].

In addition to this increased focus on EDs in males, emphasis has recently been put on older individuals. Besides geriatric patients, where due to the wide range of medical and pharmacological causes of weight loss EDs may go undetected [6–8], the majority of women between 30 and 74 years reported to be dissatisfied with their weight despite being at normal weight already in 1998 [9]. For women aged 40 years and above, prevalence rates between 3 and 4% were reported [10], and associations with menopausal transition [11–16] and symptoms have been found [17,18].

Only recently, we have conducted some of the first epidemiological studies of ED symptoms in older men, reporting ED symptoms in close to 7% of participants of a community sample or men attending a gym [19,20]. Further investigations of the underlying causes are only slowly gaining traction.

In this focused review article, we will provide an update on EDs in middle-aged and older males. We will further highlight excessive exercise as the purging method of choice and link hormonal changes in midlife to the high prevalence of ED symptoms in this age group.

Eating disorders in middle-aged men

Already in 2015, Reas and Stedal have reviewed data on male EDs in midlife and beyond [21]. They identified 16 clinical case reports that documented ED diagnoses of AN, BN, EDNOS, and BED in men aged 40 years and above, with the majority showing psychiatric comorbidities including depressive and anxiety disorders as well as suicidal ideation. Epidemiological data on this age group are still sparse, with the majority coming either from US [22] or UK national surveys [23] or military veteran discharge diagnoses [24,25]. Interestingly, the US National Comorbidity Replication study found that while prevalence rates of AN, BN and BED were still dramatically more frequent in women than in men, the lifetime prevalence of sub-threshold binge eating behavior was even higher in males than in females [22]. This lack of full ED diagnoses might be related to the diagnostic tools in use, of which the majority has been developed and validated in females, and which do not reflect sex-differences in the presentation of ED symptoms [5]. Although recent revisions of the diagnostic criteria for EDs try to adapt this catalog of symptoms used for diagnosis, like the removal of the occurrence of amenorrhea as one of the diagnostic hallmark criteria for AN [26], we are still far from sex or gender-specific diagnostics of EDs.

Since the systematic review by Reas and Stedal [21], only a small number of new publications on this topic were published. In a recently published 30-year longitudinal study, Brown et al. [27] found that while the percentage of women meeting the criteria for any full ED diagnosis decreased significantly from age 20 to 30 and stabilized form 30 to 50, no such difference was detected in male participants. Also, while, drive for thinness decreased for women, it increased for men. Interestingly, point prevalence of ED diagnoses at age 50 did not differ between genders.

While in a community study in Cyprus the age group of 46–60 years showed higher mean total scores on the Eating Attitude Test (EAT-26 questionnaire) than younger cohorts, suggesting disordered eating behavior [28], ED symptoms were found in 7% of participants in an Austrian community sample [20]. In a Dutch population based study, researchers found that dieting was most prevalent in 45- to 65-year-old men (31.7–31.9%), whereas fear of weight gain was highest in 25- to 55-year-old men (43.2–46.1%) [29]. For both measures, frequencies rose dramatically over the entire age range (no longitudinal study, based on different age groups).

Some additional new data exist from groups of military veterans. Slane et al. [30] showed that both men and women veterans who engaged in disordered eating behaviors had high rates of PTSD and major depressive disorder as compared to veterans without eating disorders (although here the average age was 30 ± 10); and homelessness might increase the likelihood for EDs [31].

Lastly, it was shown that EDs were associated with increased risk for fall injuries and fractures in a Swedish cohort study, which might be due to common ED side comorbidities like hypotension, arrhythmias, hypoglycemia, peripheral neuropathy, and osteoporosis [32].

All these findings are in line with the results of Hilbert et al. [33], who found that in males the peak total score on the Eating Disorder Examination-Questionnaire (EDE-Q) occurred at age 55–64, which means in midlife.

Taken together, the prevalence rates for EDs in middleaged men are not too far from those of women. The actual extent will become more apparent, the more research is performed in this direction. Although the rates of the classical eating disorders AN and BN are not the highest, EDNOS as well as BED are way more frequent and approach prevalence rates of females. A gender-biased propensity for EDs might therefore be only artificial, and by adapting criteria and assessment tools, prevalences might slowly even out between those gendered groups.

Excessive sports as purging method of choice in aging males

As we have briefly discussed, diagnostic criteria as well as assessment tools for EDs have so far mainly been based on clinical presentations and validations in female groups, leading to a gender-biased view of ED symptomatology. While in females the control of eating behavior is mainly influenced by a socially and media conveyed image of thinness, the ideal body image is different for males. For them it is directed toward a muscular Adonis-like physique, that requires a combination of losing weight in the sense of reducing fat content and gaining weight due to increasing muscle tissue [26]. This combination of leanness and muscularity is characterized by a “bulking and cutting” dietary practice, where “bulking” is a phase of excessive protein replenishment and “cutting” restrictive eating behavior, thereby closely resembling bulimic binges [26,34,35]. In addition, classical purging methods, like vomiting or laxative use are applied after binges.

Interestingly, Mitchison et al. [36] could show that male sex and age over 45 years was associated with increased prevalence rates in extreme dieting and purging, thereby marking this age group as particularly vulnerable to weight control measures. In addition, we could show in our community sample [20] that disordered eating was associated with increased scores on the Exercise Addiction Inventory [37], suggesting that excessive exercise is indeed used by our non-clinical sample of aging males as the purging method of choice. This connection between sports and disordered eating, although previously described [38], is further corroborated by reports that around one fifth of team sports members between 18 and 55 years showed an ED diagnosis [39], and that military veterans were more likely to engage in excessive exercise instead of other purging methods, like vomiting [40]. Also in the context of fitness centers, excessive exercise was found as the predominant compensation for binging in men with bulimic symptoms [19].

Excessive sports activity can therefore be seen as male-specific characteristics of eating disorders that may serve as a strategy for acute emotion regulation counteracting depressive symptoms [41,42].

Midlife hormonal changes—andropause and eating disorders

In recent years a body of literature has been produced linking the hormonal change of menopause to female eating disorders, therefore posing these hormonal changes as a critical window of susceptibility for the occurrence of eating disorder symptomatology [10]. The menopause, or climacteric, is an age dependent cessation of the menstrual period, which is characterized by a sharp decrease in the production of the hormones estradiol and progesterone by the ovaries. In contrast, changes of male hormone levels, most importantly testosterone, do not show such an abrupt decrease, but a more gradual decline over years. This age-dependent reduction in testosterone levels is termed late-onset hypogonadism (LOH) or testosterone deficiency [43,44]. To have a direct equivalent for menopause, some authors have also called it andropause, although this terminus has been criticized for inconsistency and not reflecting the actual symptoms [45].

The psychological relevance of LOH is made apparent by prevalence rates of 20–31% and the association with various psychiatric disorders [45]. With the stark increase of psychological disturbances during middle-age in men, the connection with aging symptomatology became apparent [46]. Although in an early study andropause has not been associated with specific psychological symptoms, but more with subthreshold depressive symptoms [47], it has been shown that testosterone replacement therapy can modestly improve depressive symptoms [48]. Interestingly, low plasma testosterone levels have been shown to be bidirectionally associated with obesity, with obese men being at particularly high risk for developing secondary hypogonadism and testosterone deficiency being a cause for the development of obesity [49,50]. This is of special interest as also obesity is associated with a substantial number of psychological comorbidities [51].

Data showing a direct link between disordered eating and andropause are sparse. In a community sample we could recently show that men that showed high values on the Aging Male's Symptom scale (AMS) also showed higher prevalence rates for eating disturbances [52], suggesting that also here the psychological manifestation of age-related decline of testosterone acts as a critical window for eating disorder susceptibility. Indeed, also in young adult men it was shown that lower testosterone levels significantly correlated with dysregulated eating symptoms, even after controlling for depressive symptoms, body mass index and age [53], whereas initial data on 8–9 year old children revealed a greater body dissatisfaction with higher testosterone levels, although this association vanished after adjustment for covariates [54]. Although testosterone replacement therapy is able to change body composition and decrease fat mass in obese individuals [49,55], its effect on the actual expression of disordered eating behaviors has so far not been investigated systematically. And while the application of low-dose testosterone in female AN patients did not affect disordered eating symptoms [56], a possible treatment effect on aging males still needs to be determined.

Conclusion

The increasing number of reported eating disorders in males suggests both greater awareness and acceptance in the public and professional perception. However, the systematic investigation of epidemiological differences and risk factors especially in older age groups is only slowly starting to gain traction. While prevalence rates of disordered eating in middle-aged and older males are comparable to females of the same age group, the strategies they use to control weight are different, with excessive exercise being their purging method of choice. Given the apparent societal trend toward healthy lifestyle behaviors, these problematic eating and restraint patterns often go undetected. In addition, the age-dependent decrease of testosterone levels in aging males seems to constitute a specific window of susceptibility for disordered eating, similar to that of menopausal hormone dysregulation in women. Further research is urgently needed to better understand the relationship between age and eating disorders, and to provide affected individuals with adequate therapeutic interventions.

Disclosure statement

No potential conflict of interest was reported by the author(s).