Introduction

The notion that eating disorders (EDs) did not afflict men has changed. Last century this asseveration gave way to the notion that men account for approximately one in ten ED cases, but recent evidence suggests that men may represent approximately one in four presentations of EDs [1]. In Spain, Sepulveda et al. [2] determined in man university students that the prevalence of risk of developing an ED was 14.9% while a proportion of 11.6% reported binge eating behavior. In Mexico, data from the National Health and Nutrition Survey (Encuesta Nacional de Salud y Nutrición [ENSANUT]) in 2012 [3] showed that 0.8% of adolescent men was at risk of developing an ED. This percentage of adolescent men in risk was twice (0.4%) that observed in ENSANUT 2006 [4]. There are multiple risk factors associated with the emergence, development, and maintenance of EDs [5,6,7]. One of these risk factors are disordered eating behaviors (DEB), defined as abnormal behaviors related to food and eating, such as dietary restraint, binge eating behavior, fasting and excessive/compulsive exercise to modify the body shape and lose weight [5, 8, 9]. In Spain, Pamies et al. [10] found a prevalence of 3.3% of DEB in adolescent men. They also found that subjects with body mass index (BMI) above normality (overweight and obesity) achieved the highest values to develop DEB [10]. The ENSANUT 2012 [3] reported that the most frequent DEB in adolescent men were exercising to lose weight (12.7%) and binge eating behavior (11.9%). Another Mexican studies conducted on men aged 15–20 have reported a prevalence of DEB from 2.8 to 15.6% [11,12,13,14,15,16,17].

On the other hand, being strong, athletic, and with a body muscular definition have been established as the ideal model for men [18]. Men suffer a social pressure to maintain or achieve this “ideal figure” [19], and probably the increase of exercise at gyms together with the use of anabolic steroids is regarded as a response to this pressure [20]. Men who think they cannot reach this ideal may develop the drive for muscularity (DM), term coined by McCreary and collaborators [21], characterized by excessive concern with musculature and body size as well as attempts to gain weight without fat [22]. The drive for muscularity has been suggested to be associated with muscle dysmorphia (MD), a subcategory of body dysmoprhic disorder characterized by an obsession with one’s body not being sufficiently lean and muscular [17, 22,23,24]. Recently, it has been suggested that DM may be part of the muscularity-oriented disordered eating, psychopathologies defined as an array of DEB that is driven by the pursuit of muscular ideal [25]. In men, higher DM has been associated with higher ranges of DEB and increased use of anabolic steroids and supplement use [15, 26,27,28].

With respect to driven by muscularity concerns, an overall point prevalence of 3–4% has been suggested in men [29]. In the USA, up to 60% of all boys report purposefully manipulating dietary practices in the pursuit of greater muscularity [30]. In Mexico, using the Drive for Muscularity Scale [DMS; 21] the prevalence of adolescents and youths in risk to develop DM has showed values from 8 to 10.6% [15, 31].

In addition, McCreary and colleagues [32] explored the relationship for anthropometric measurements of men’s muscularity and adiposity with the drive for muscularity. They worked in a sample of college-aged men and used the DMS [21]. Only flexed bicep circumference was significantly associated with the drive for muscularity. Another study designed to clarify the association for drive for muscularity and anthropometric indicators (percentage body fat, fat-free mass, and BMI) conducted on college-aged men found no correlation [33]. In Mexico State, a study was conducted on men university members who did not engage in physical activity (sedentary) aged 17 and 36 years and using the same scale [21]; BMI, body density, percentage body fat, fat-free mass, and waist/hip ratio were not significantly correlated with the overall drive for muscularity score [34].

At the same time, physical activity (PA) has been linked to overall health improvement. Regular PA has positive effects, for example, on the control and prevention of chronic non-communicable diseases such as obesity, hypertension, and diabetes mellitus type II [35, 36]. Likewise, PA can also provide benefits at the psychosocial level by improving self-esteem, self-satisfaction, overall mood, and increase socialization [37]. However, when a person’s attitude toward exercise, specifically strength training, becomes extreme or obsessive, it may be a sign of DM [20, 38] or a precursor of EDs [39, 40].

Based on the above, the aim of this study was to determine the risk of developing DEB and DM and its association with anthropometric indicators and PA in adolescent men at a private high school from Hidalgo State in Mexico.

Methods

Design and participants

A cross sectional, descriptive, and correlational field study was carried out in a private high school in Hidalgo, Mexico. The sample initially consisted of 285 men. Participants, who did not complete questionnaires in full and/or did not have anthropometric measurements or were ≥ 20 years old, were not included in the sample, leaving 267 Mexican adolescent men in the current analyses. The final sample ranged in age from 15 to 19 (M = 16.23, SD = 0.1). The present study was approved by the Ethics and Research Committee of the Health Sciences Institute of the Universidad Autónoma del Estado de Hidalgo (Folio Number 044), Mexico. Informed consent was obtained from each student who wanted to take part in the study as well as the parents’ or legal guardians’ consent. Participants completed questionnaires in the classrooms. Once the questionnaires had been completed, anthropometrical measurements were taken. The facilitators’ team (students enrolled in the bachelor’s degree program in nutrition) was trained to apply the questionnaires and take the anthropometric measurements in a standardized way.

Instruments and measurements

Drive for Muscularity Scale (DMS)

DMS [21] measures the attitudes and behaviors showing the degree of concern about becoming more muscular. In Mexico, Escoto et al. [41] conducted an exploratory factor analysis to evaluate the Spanish version of the scale, achieving an adequate level of internal consistency (Cronbach’s alpha coefficient = 0.86). DMS comprises 15 items with a Likert-type scale including six possible answers, ranging from “always” (1) to “never” (6). As there was no cutoff score for the scale, we obtained it from the mean plus one standard deviation [42]. The cutoff point was ≥ 52, with a Cronbach’s alpha coefficient of 0.85, indicating a risk of developing drive for muscularity.

Brief Questionnaire to Measure Disordered Eating Behaviors (BQDEB)

The BQDEB (CBCAR Spanish acronym) was developed and validated for the Mexican population, yielding a Cronbach’s alpha coefficient of 0.83 for women [43] and 0.63 for men [12]. The cutoff point, sensitivity (0.81), and specificity (0.78) and predictive values (positive predictive value = 0.38; negative predictive value = 0.96) of the questionnaire were determined using 2 × 2 tables. A discriminant analysis showed that almost 90% of cases were correctly classified. It consists of ten items to evaluate concern about gaining weight and disordered eating behaviors (i.e., dietary restraint, weight concerns, self-induced vomiting, laxative use, fasting and driven/compulsive exercise) in the 3 months before the questionnaire was applied. It is scored on a four point Likert-type scale with four answer choices ranging from never (0), to very often/more than twice a week (3). It has a cutoff score of > 10 to identify subjects at a high risk of developing disordered eating behaviors [43]. This questionnaire has been widely used in several studies to examine DEB in groups of adolescents and youth people [11,12,13,14,15,16,17] and very specifically in the ENSANUT surveys in 2006 [4] and 2012 [3].

Body mass index (BMI)

The anthropometric measurement of height was registered using a portable stadiometer (SECA-214, Hamburg, Germany); weight was measured using a digital scale (BIA; TANITA TBF-521, Illinois, USA).

Both measurements were carried out in a standardized way, with participants wearing light clothing, in bare feet. BMI cutoffs were obtained by age using the values recommended by the National Center of Health Statistics [44].

Body fat percentage (BFP)

This anthropometric measurement was done through Bioelectrical Impedance Analysis (BIA; TANITA TBF-521, Illinois, USA) following the protocol described by the manufacturer. This method was chosen because of its advantages such as low cost, ease of transportation and handling, and low variability between observers [45]. The classification proposed by Lohman [46] was used to categorize BFP. A percentage of ≥ 25% in men indicates obesity.

Fat-free mass index (FFMI)

The fat-free mass index (FFMI) is the weight of all body tissue, except fat. This calculation was made using the formula proposed by Kouri and Kyle [47, 48].

$$\left[ {{\text{Weight}} \times \left( {100 - \% {\text{ fat}}} \right)/\left( {{\text{size}}^{2} \times \, 100} \right)} \right] + \left[ {6.1 \, \times \left( {1.8 - {\text{size}}} \right)} \right] .$$

Subjects were classified using the cutoff score proposed by Gruber and colleagues [49] with five categories ranging from a value of 18 for low musculature to 25 as the upper limit of muscularity achieved without the use of steroids.

International Physical Activity Questionnaire (IPAQ)

IPAQ was developed to estimate the physical activity patterns of populations of different countries and may be used with adolescents, youths, and middle-aged adults (15–69 years) [50]. The short form comprising 8 items—the version validated for Mexico—[51] was used to assess PA in the past 7 days. For the present study, the self-report format was used. For the interpretation of the duration of PA (minutes per week), the number of minutes of vigorous physical activity was multiplied by two which was added to the total number of minutes of moderate activity (including walking) to obtain moderate physical activity (MPA). Vigorous physical activity was multiplied by two taking the WHO’s [52] recommendation for adolescents into account: adolescents should engage in 30 min of vigorous physical activity daily or 60 min (twice) of MPA. On the basis of these recommendations, three levels of MPA were obtained, multiplying 60 min × 7 days of the week: (1) < 420 min/week is equivalent to a low level of MPA; (2) ≥ 420 to < 840 min/week equals a MPA; (3) ≥ 840 min/week means high MPA level. Because different studies have observed an overestimation of PA reported by the IPAQ self-report format, these data were adjusted using the equation proposed by Medina et al. [53].

Statistical analysis

Data were analyzed using SPSS software version 20 for Windows (IBM Inc., New York, USA). Descriptive analyses of frequency, percentages, and measures of central tendency were conducted. Subsequently, as well all the variables were categorical, a Chi-square test (χ2) was performed to evaluate significant statistical differences between subjects without and with risk to develop DEB or DM and the anthropometric indicators and MPA levels. Fisher’s exact tests were used for testing association in tables 2 × 2 or when more than 50% of the cells had values lower than 5. Analysis of variance (ANOVA) with Bonferroni post hoc test was carried out to assess group differences. Pearson correlations were performed to identify the linear correlation between DEB, DM, MPA, and anthropometric measures. Binary logistic regression analyses were used to identify predictive factors for DEB and DM. Both dependent variables were coded as 0 (without risk) and 1 (with risk). Factors entered into the model included BMI, BFP, FMMI, and MPA, for both outcome variables. It should be noted that for the first model, DM was included as one more predictive factor for DEB; in the second model, DEB was another predictive factor for DM. The R2 of Nagelkerke was reported and odds ratios (OR) were calculated as well as the confidence intervals (95% CI); p < 0.05 was considered to be statistically significant. Data models were adjusted.

Results

Out of the final sample and according to the cutoff score of ≥ 52 for the DMS, the results showed that 14.2% of the subjects reported being at risk of developing DM, while taking into account the cutoff score for the BQDEB (> 10), 6.7% of the adolescents were at risk of developing DEBs; 23% of this last percentage were also at risk of developing DM. The most frequent DEBs engaged by participants were exercising to lose weight (21.2%) and binge eating behavior (15.6%).

With respect to the anthropometric indicators, BMI distribution for overweight and obesity was 32.6%; meanwhile to BFP, 54.7% of the respondents were classified as having overweight and obesity. In agreement with FMMI, 84.3% were categorized as having low musculature. Even though no statistically significant differences were found for BMI and the risk of developing DM, subjects with normal weight achieved the highest percentage (63.2%), followed by overweight participants (21.1%). Analyzing the distribution of the BFP categories by the risk of developing DM, but without significant differences, subjects classified with obesity reached the highest percentage (34.2%). The percentage of adolescents with low musculature (FFMI) was lower among those with high risk for developing DM with statistically significant difference (p = 0.04). These results are given in Table 1.

Table 1 Distribution by BMI, BFP, FFMI categories and MPA levels according to the cutoff point of the DMS and BQDEB
Full size table

According to the data, the MPA mean was 390.53 min/week (SD = 151.0 min/week), slightly below the daily recommendation indicated by WHO [52], of at least 60 min per day (420 min/week). Nevertheless we obtained three levels; the total sample was distributed in just two: low (57.7%) and moderate (42.3%) levels of PA. With respect to MPA, there was statistically significant differences according to Fisher’s exact test (p = 0.02); the highest percentage of those at risk of developing DM was identified in subjects classified as engaging in a moderate level of PA (60.5%); on the contrary, participants with a low level of MPA and not at risk of developing DM reached the highest value (60.7%).

In the comparative analysis of BQDEB, significant differences were found in the anthropometric indicators. Subjects classified with both BMI and BFP with obesity had the highest percentages of risk of developing DEB (38.9% and 67.7%, respectively), while participants with low musculature reported the highest risk value (50%). Finally, in keeping with the cutoff points established for MPA, we did not find any statistically significant differences by level; however, we must underline that participants performing low level of MPA achieved the highest percentage for the risk of developing DEB (61.1%). These findings are also shown in Table 1.

A slight but significant positive correlation was found between the DMS score and the BQDEB score, as well as between DMS and MPA (min/week) among the sample. Also with significant correlations, the BQDEB score increased, as did the three anthropometric indicators, particularly the BMI (r = 0.507; p < 0.01). DMS did not show correlations with the anthropometric indicators. These findings are displayed in Table 2.

Table 2 Correlations between DMS, BQDEB, anthropometric indicators, and MPA (min/week)
Full size table

Finally, binary logistic regression analyses were conducted, beginning with BMI, BFP, FFMI, MPA, and DM as independent variables (predictors) and the risk of developing DEB as a dependent variable (outcome variable). The findings indicated that two variables were associated with DEB: DM and BMI. Adolescents in the DM risk category (OR 8.28, CI 2.71–25.22) had significantly greater odds of developing DEB, while adolescents with higher BMI scores (OR 1.28, CI 1.14–1.45) tended to be more likely to this risk. The model correctly classified 92.9% of the cases and explained 30.0% of the DEB variance (R2 = 0.295). The other anthropometric variables and MPA were not significant. For the second model, BMI, BFP, FFMI, MPA, and DEB were taken into account as predictors of DM risk. The results showed that DEB and MPA were associated with DM; adolescents who engaged in DEB had an 8.64 times (CI 3.05–24.52) higher risk of developing DM. Meanwhile, participants classified in the category of highest MPA (OR 2.66, CI 1.26–5.59) had significantly greater odds of developing DM. The rest of the independent variables were not significant. The model correctly classified 86.9% of the cases and explained 14% of the DM variance (R2 = 0.137). Table 3 shows the coefficients of both models.

Table 3 Binary logistic analysis of anthropometric indicators and physical activity for DEB and DM
Full size table

Discussion

The present study provides information to expand the limited research on the association between DM and BQDEB scores and anthropometric indicators in adolescent men. The aim of this study was to determine the risk of developing DEB and DM and its relationship with anthropometric indicators and MPA in a sample of Mexican adolescent men at a private high school.

The prevalence of subjects at risk of developing DM is difficult to determine, since most studies have been conducted on gym users [5, 22, 38]. An examination of the results of DMS shows that the risk reported by our sample (14.2%) was close to the 10.6% rate found by Escoto [31] among men aged 15–41, who were gym users rather than bodybuilders, using the same instrument and cutoff point. A comparison of our outcomes with those of Villarreal [15], who in a sample of adolescent men from a public high school in Hidalgo, Mexico (using the same instrument and cutoff point), identified 8% at risk of developing DM, shows that the rates in our study were higher. This recalls the findings of Camacho et al. [34], who reported that 7.9% of a sample of man university students was at risk of developing DM, which is almost half the percentage shown in the current sample. However, it is important to note that, to identify the risk, they established a cutoff point of ≥ 45, which is lower than the one used in the sample under study. On the basis of these data, it is feasible to suggest that even though a higher cutoff point was used, a higher percentage of DM was found in our research.

With regard to the prevalence of adolescent men at risk of developing DEB in our sample, the value found was far higher (6.7%) than that reported in other studies [2,3,4, 10,11,12, 15,16,17], yet lower than that obtained by Chávez et al. (15.6%) [14], suggesting that this phenomenon is increasing in Mexican adolescent men. Although it has been reported that the risk of developing an ED may occur in any socioeconomic group [54], our findings, like those of other Mexican studies [11, 12, 15, 55] (which regard students at private schools as having high socioeconomic status), support the proposal that the prevalence of risk of developing symptoms of ED is higher in adolescents with a high socioeconomic status. The authors found the same trend for the risk of developing DM, which could be explained by the fact that higher socioeconomic status provides greater access to gyms and anabolic steroids and subjects those in that group to greater sociocultural pressure.

Subjects with normal BMI showed the highest percentage of risk of developing DM, while subjects in the obesity category had the highest risk of developing DEB (according to the cutoff score for the BQDEB). This contrast suggests that in subjects with normal BMI, their main concern would be to increase their muscle mass [29], while subjects with obesity might be more concerned with losing weight [40]. The association between the BQDEB score and BMI was similar to that reported in studies conducted on women and men [12, 13, 15,16,17]; the higher the BMI levels, the higher the BQDEB score. This suggests that the increase in BMI may be one of the factors that has influenced the increase in the risk of developing DEB, as adolescents and youth people who wish to be thin use strategies that are not always healthy to achieve this goal. The same trend was observed in relation to BFP and BQDEB scores, indicating that higher BFP levels drive subjects to engage in behaviors focusing on thinness (such as low body fat) [28].

In relation to FFMI, unlike the linear theoretical association identified by Davis and colleagues [56] and McCabe and Riccardelli [57], who found a positive correlation between muscle mass and DMS score, we observed that subjects with normal and low muscle mass showed the highest percentage of risk of developing DM. We posit that these subjects engage in behaviors associated with gaining muscle mass [21]. Moreover, despite the low correlation between FMMI and the BQDEB score, it should be noted that subjects with low and normal FMMI reported the highest percentages of risk of developing DEB. This data may indicate that high scores in BQDEB among adolescents with normal and low muscle mass are indicative of a desire to be thinner. These findings are supported by Olivardia et al. [23], who posit that men wish to gain substantial amounts of muscle and lose a significant amount of body fat at the same time.

Interestingly, participants who engaged in moderate physical activity reported the highest risk of developing DM. We also observed a low positive correlation between the amount of MPA and the DMS score, suggesting that when exercise becomes excessive, it may be a sign of DM [21] or a precursor of MD [38]. Moreover, participants classified with a low level of MPA achieved the highest value for the risk of developing DEB. In our opinion, these data suggest that subjects who engage in MPA are more concerned with their muscle mass, while subjects with low levels of PA prefer to adopt disordered eating behaviors (with the aim of controlling their body weight) rather than engage in physical activity.

Like Bratland-Sanda and Sundgo-Borgen [40], we failed to find an association between the amount of MPA and BQDEB scores, even though authors such as Waaddegaard, Davidsen and Kjoller [39] and León-Vázquez et al. [17] have reported that excessive amounts of exercise are associated with an increased risk of developing DEB. This lack of association was probably a result of the sample distribution (in just two categories: low and moderate). Another reason is the fact that MPA was not categorized by type of exercise or the underlying reasons for doing exercise [40, 58]. Moreover, we did not identify the type of physical activity performed (aerobic or strength building). Some authors have associated strength-building PA with the development of DM and symptoms of ED as DEB [40, 58].

In keeping with other studies [33, 34], there was no significant correlation between the anthropometric indicators measured and the overall DMS score in our sample. McCreary and colleagues [32] have suggested that men’s drive for muscularity is not associated with the actual level of muscularity or their anthropometric index of adiposity. However, they report that flexed bicep circumference is predictive of muscularity-oriented behaviors. Other authors [59, 60] consider that DM is more closely related to the body shape of an inverted triangle (pec-waist index; muscular shoulders and chest combined with a narrow waist) than body weight.

A low positive correlation between DMS and BQDEB was identified [5, 40], probably as a result of the difference between the two scales. While DMS measures attitudes and behaviors associated with increasing muscular mass, BQDEB measures behaviors focused on losing weight. It has been said that men usually do not have much motivation to be thin and are more likely to want to gain weight rather than lose it [17, 24]. Using methods and conceptions based on feminine ideals is, therefore, an unsuitable approach to ED in men [61]. Likewise, in a sociocultural model of DEB among French adolescent boys, Rodgers et al. [62] showed that the pursuit of muscularity is one of the pathways associated with DEB.

Further analysis of the results reveals that the small positive correlations between MPA, BQDEB, and DMS found among adolescents in the study may account for the small percentage of variance explained by the models in the regression analyses, mainly in the second model (14%). The data indicate that DM increases 8.28 times the risk of developing DEB, which is higher than rate reported by Unikel-Santoncini [16] of 4.4 times and of Leon-Vazquez et al. of 5.3 times [17]. These findings are supported by Lavender and colleagues [28], who found that muscularity-related behaviors were positively associated with eating disorder symptoms among men, as well as Cafri and colleagues [20] and Rodgers et al. [62], who have posited DM as a factor for increasing the frequency of DEB among men. Although some studies have failed to observe any association between BMI and DEB in populations of men [12, 13, 15,16,17], we found that this anthropometric indicator significantly increases 1.29 times the risk of developing DEB. This slight OR data are similar to that reported by Eisenberg et al. [30] in adolescent men. Moreover, the association between BMI and DEB in men is lower than that reported in women [10, 12, 16], suggesting that individual factors, such as weight history [24] and sociocultural variables [5, 18, 20] could probably be more important in explaining DEB in men.

Even though we found that DEB increases 8.64 times the risk of developing DM, we were unable to find studies proposing DEB as a predictor of DM. Only Rodgers and colleagues [62], who studied French adolescent boys, identified a relationship between the pressure to lose weight and the drive for muscularity; however, this relationship was mediated by internalization and appearance comparison. Regarding physical activity, the highest category of MPA increases 2.66 times the risk of developing DM, supporting the idea that when PA becomes extreme or obsessive, it could be linked to DM [28, 30], MD [5, 21], and EDs [39, 40]. According to Eisenberg and colleagues [30], adolescent men who participated in sports teams were 2.05 times more likely to report more muscle-improving behaviors than those who do not play sports. Studies show that men are less likely to participate in compensatory behaviors, such as vomiting or laxative abuse; and more inclined to use excessive exercise as a compensatory method for controlling weight and body shape [24]. It is important to underline the fact that the most frequent DEB engaged in by participants was exercising to lose weight (21.2%).

Limitations and directions for future research

It should be noted that the limitations of the present study are associated with its cross-sectional design, since causality cannot be inferred from the findings. Moreover, because the participants were not randomly selected, the data cannot be generalized to the entire Mexican adolescent population, even though validated, reliable instruments were used. However, these limitations are partly offset by the fact that anthropometric indicators were measured directly, rather than using self-report questionnaires. Future studies should include questions to identify the type of physical activity (aerobic or strength building). Moreover, the motivation behind PA should be explored [40], since it is important to examine the qualitative rather than just the quantitative dimensions of PA, which will provide a clearer view of the association between PA and the development of DEB or DM in men.

In our opinion, and in line with several authors [5, 6, 25, 28], to improve the analysis of the multivariate effects and the models’ predictive capability for the risk of developing DEB and DM, future studies should include other multiple mediating variables such as body image, since authors such as Jones and Crawford [63], who explored the relationship through a dual pathway model for body dissatisfaction among adolescent men, found weight concern to be associated with high BMI while muscularity concern was significantly higher among boys with lower BMI. Other important mediating variables could be muscle appearance satisfaction [64], physical self-concept, self-esteem [27, 31], depression [16, 27], body thin ideal internalization [17], and sociocultural factors [62].

Another limitation could be associated with the inability to separate the assessment of muscularity-related attitudes and muscularity-related behaviors which, according to Parent and Bradstreet [27], is vital for achieving a better explanation of the relationship between anthropometric indicators such as BMI, PA, DEBs, and DM. The existence of a three-factor structure for the Spanish version of the DMS: attitudes, supplement consumption, and training adherence [41] as opposed to other studies with a two-factor structure of DMS: attitudes to muscularity and behaviors oriented toward gaining muscle mass [21, 65, 66], limited the detection of these differential relationships.

Conclusion

Despite the limitations, to our knowledge, this is the first study to examine the association between DM, DEB, anthropometric indicators, and PA in Mexican adolescents. We have also observed a high prevalence of adolescent men at risk of developing DM and DEB in the research sample. Moreover, we suggest that higher socioeconomic status, measured by type of school (private or public), increases the risk of developing both pathologies. Although we did not work with a representative sample of Mexican adolescent men and are, therefore, unable to generalize the findings, our data could be interpreted as a warning sign for researchers who design and implement prevention programs.

Nevertheless, we observed a slight correlation between BQDEB and DMS scores; subjects who engage in DM tend to be at risk of developing DEB, particularly as a function of BMI. We also found that subjects who engage in DEB were more likely to develop DM when subjects were classified in the highest category of MPA.

As has been reported in the literature, we were able to identify non-significant correlations between the anthropometric indicators and DM. However, future longitudinal and cohort studies focusing on mediators’ and moderators’ variables may be required to clarify causality.

To explore the link between PA, DM, and DEB, it is necessary to analyze the type of PA, specifically those focused on increasing muscularity or losing body fat. These findings are also evidence that instruments to measure DEB in men originally designed for women based on the female model may have difficulty identifying DEB and establishing significant correlations. This is an area where valid, reliable instruments should specifically be designed for men in future studies.

Given the association observed between PA and DM as well as between BMI and DEB, prevention programs for adolescent men should target healthy eating behaviors together with appropriate physical activity emphasizing health, nutrition, and body composition rather than muscle mass development.