Bidirectional Relations Between Disordered Eating,Internalization of Beauty Ideals,and Self-Esteem: A Longitudinal Study With Adolescents

Abstract

Associations between disordered eating, internalization of beauty ideals, and self-esteem have been mainly studied in a unidirectional way. Therefore, this study explored the potential bidirectional associations between these three aspects, in a sample of 303 adolescents aged 12 to 15 (140 females and 163 males), in a 16-month longitudinal prospective study. Participants self-reported disordered eating, internalization of beauty ideals, and self-esteem. Autoregressive cross-lagged analyses showed bidirectional associations between self-esteem and disordered eating and unidirectional links between internalization of beauty ideals and disordered eating and lower self-esteem. All cross-lagged effects were equivalent across genders, except that the effect of internalization of beauty ideals at baseline on disordered eating 4 months later was higher for females. Temporal stability was moderate for all measures. These results contribute toward completing explanatory models of these problems and suggest that interventions based on the critique of esthetic ideals and on strengthening self-esteem should be implemented in early adolescence.

Keywords

longitudinal disordered eating reciprocal associations self-esteem internalization

Introduction and Aims

Eating disorders tend to appear in late adolescence and early adulthood (Rohde, Stice, & Marti, 2015). However, the biological, psychological, and sociocultural risk factors for disordered eating can appear in childhood and early adolescence, in particular perceived pressure to be thin, thin-ideal internalization, body dissatisfaction, weight concerns, weight importance, dieting, negative affectivity, and self-esteem (Loth, MacLehose, Bucchianeri, Crow, & Neumark-Sztainer, 2014; Ricciardelli, McCabe, Holt, & Finemore, 2003; Rohde et al., 2015).

Recent researchers have examined a biopsychosocial model of body image concerns and disordered eating in early adolescence (Rodgers, Paxton, & McLean, 2014) that expands the Tripartite Influence Model (Rodgers, Chabrol, & Paxton, 2011; Thompson, Heinberg, Altabe, & Tantleff-Dunn, 1999), incorporating psychological (negative affect) and biological (body mass index) factors as moderating variables between sociocultural influences, internalization, and social comparison (Rodgers et al., 2014). In this model, negative affect, which includes depressive symptoms and low self-esteem, was associated with internalization and social comparison, as well as with bulimic symptoms. Thus, in the biopsychosocial model, negative affect is posited to be a precursor of body image and eating concerns (Rodgers et al., 2014), rather than an outcome, as hypothesized by the Tripartite Influence Model (Rodgers et al., 2011).

Low self-esteem has been identified as a predictor factor of the beginning of diets and disordered eating (Loth et al., 2014) or as a result of bulimic symptoms (Rodgers et al., 2011). High levels of disordered eating have been associated with lower levels of self-esteem in adolescents of both genders (Hansson, Daukantaitė, & Johnsson, 2016). This background suggests a bidirectional association between self-esteem and disordered eating; however, we are not aware of any studies that analyze these associations.

As regards internalization, in addition to its mediating role in the development of disordered eating, a direct predictive association has also been described in disordered eating in adolescents of both genders (e.g., Espinoza, Penelo, & Raich, 2010; Francisco et al., 2015; Rodgers et al., 2011). Some studies suggest that in males, the internalization of appearance ideals predicts the start and continuation of disordered eating (Dakanalis, Pla-Sanjuanelo, et al., 2016). However, females seem to respond to the internalization of beauty ideals with greater intensity and would be more vulnerable to developing body dissatisfaction and disordered eating (Knauss, Paxton, & Alsaker, 2007).

The interrelation between internalization and self-esteem has received more attention in females than in males. Some authors have reported that females with low self-esteem are concerned about adjusting to the sociocultural norms of appearance to be socially accepted, internalize the thin ideal with more intensity, pay more attention to advertisements related to appearance, compare themselves more to media models, and perceive greater social pressure toward thinness than females with higher self-esteem (Mischner, van Schie, & Engels, 2013). In males, we are not aware of any studies on the association between these variables. However, it has been described that in males, self-esteem is a moderating variable of the impact of sociocultural pressures on body change strategies and that males with low self-esteem were more affected by sociocultural pressures (Ricciardelli & McCabe, 2001). One gap in the extent research is the examination of the bidirectional relationship between internalization and self-esteem. The internalization of strict and unattainable body ideals could lead to lower self-esteem in adolescents, by using these social standards as a parameter of self-worth (Crocker, Sommers, & Luhtanen, 2002).

Recent studies in this research field indicate that during adolescence the predictors of disordered eating and of body image dissatisfaction vary with age (Loth et al., 2014; Rohde et al., 2015) and that bidirectional associations can appear between them which are dynamic and sensitive to developmental changes (Sharpe et al., 2017; Wichstrøm & von Soest, 2016). Early adolescence is a relatively short period during which intense and rapid physical and psychosocial changes take place, and as a result the potential reciprocal relations between the risk factors for disordered eating need to be studied with short-interval longitudinal monitoring, to be able to pinpoint these changes and contribute toward identifying the appropriate age for prevention programs. In addition, this knowledge is relevant to widen the etiological models of eating problems and to refine preventive programs in this field of study.

Therefore, the aim of this study is to explore the potential bidirectional longitudinal associations between disordered eating, internalization of beauty ideals, and self-esteem in early adolescence and to examine the equivalence of these associations between genders. We hypothesize that the three aspects considered will reinforce each other reciprocally over time and that the magnitudes of these associations will be greater in females.

Method

Participants

Participants were recruited from the second year of compulsory secondary education in the cities of Manresa and Terrassa (Catalonia, Spain), selected using incidental sampling. Participants were 303 adolescents (140 females and 163 males) aged 12 to 15 (M = 13.4 years; SD = 0.49) at baseline (T1) from five schools. Most of the participants (48.6%) were of medium-high/high socioeconomic status (SES; Hollingshead, 2011), while 32.5% were of medium-low and 18.9% of low SES. Mean body mass index (BMI), based on in situ measurements of height and weight, was 20.5 (SD = 3.4) for females and 19.9 (SD = 3.3) for males at baseline, and 21.8 (SD = 3.5) for females and 23.3 (SD = 3.9) for males 16 months later. Weight status at baseline, according to international criteria that consider gender and age (Cole, Bellizzi, Flegal, & Dietz, 2000; Cole, Flegal, Nicholls, & Jackson, 2007), was 6.7% underweight, 71.6% normal weight, 18.1% overweight, and the remaining 3.6% being classified as obese. The distribution of participants in terms of origin was 81.2% from Spain or Europe, 9.6% from Morocco, 7.9% from Central and South America, and 1.3% from other (Asia or Africa). These participants were followed up and completed the measures 4 (T2; n = 270), 9 (T3; n = 265), and 16 (T4; n = 256) months later. There were no differences regarding gender or weight status between participants who dropped out at some follow-up moment compared with those who did not drop out (p ⩾ .360). Regarding SES, there were no differences between participants who dropped out during follow-up compared with those who did not drop out at T2 (p = .588) and T3 (p = .189). However, we found differences at T4, since low SES participants presented a higher dropout rate compared with medium and high SES participants (p = .001). Eighty percent of the participants reported having pubertal changes (menarche in females or changes in voice or some other physical change in males) at the baseline. Table 1 presents the descriptive data for each measure considered over the four assessment points, and internal consistency reliability for self-reported scores (see below) in the study sample, and bivariate correlations between measures.

Table 1.

Descriptive Statistics, Internal Consistency, and Intercorrelations by Gender at the Four Time-Assessment Points (N = 303).

Time (n)	Measure (minimum ÷ maximum)	Cronbach’s alpha		M (SD)		Pearson’s correlations
Time (n)	Measure (minimum ÷ maximum)	Males	Females	Males	Females	1	2	3	4	5	6	7	8	9	10	11	12
T1 (285)	1. EAT-26 Eating attitudes (26 ÷ 156)	.88	.91	45.87 (16.41)	53.76 (19.87)		.54**	−.57**	.76**	.50**	−.44**	.60**	.32**	−.26**	.60**	.33**	−.34**
	2. SATAQ-R Internalization (1 ÷ 5)	.67	.86	2.65 (0.71)	2.85 (0.88)	.45**		−.47**	.61**	.65**	−.44**	.55**	.55**	−.35**	.53**	.54**	−.42**
	3. RSES Self-esteem (10 ÷ 40)	.71	.84	32.12 (4.65)	29.59 (5.58)	−.59**	−.45**		−.53**	−.47**	.73**	−.50**	−.36**	.68**	−.40**	−.33**	.61**
T2 (270)	4. EAT-26 Eating attitudes (26 ÷ 156)	.90	.91	43.94 (17.54)	51.72 (19.53)	.71**	.31**	−.56**		.64**	−.56**	.82**	.48**	−.40**	.79**	.42**	−.49**
	5. SATAQ-R Internalization (1 ÷ 5)	.80	.83	2.58 (0.85)	2.71 (0.84)	.35**	.65**	−.31**	.32**		−.45**	.63**	.68**	−.45**	.56**	.59**	−.50**
	6. RSES Self-esteem (10 ÷ 40)	.79	.82	32.69 (5.26)	29.81 (5.46)	−.47**	−.38**	.72**	−.49**	−.46**		−.54**	−.38**	.72**	−.44**	−.32**	.63**
T3 (265)	7. EAT-26 Eating attitudes (26 ÷ 156)	.90	.93	42.98 (17.65)	53.71 (22.26)	.59**	.34**	−.40**	.62**	.35**	−.41**		.62**	−.54**	.84**	.51**	−.51**
	8. SATAQ-R Internalization (1 ÷ 5)	.78	.87	2.59 (0.81)	2.88 (0.91)	.27**	.54**	−.29**	.18	.64**	−.37**	.36**		−.50**	.46**	.73**	−.43**
	9. RSES Self-esteem (10 ÷ 40)	.80	.83	32.37 (5.32)	29.41 (5.59)	−.38**	−.30**	.62**	−.29**	−.40**	.70**	−.32**	−.40**		−.43**	−.40**	.74**
T4 (256)	10. EAT-26 Eating attitudes (26 ÷ 156)	.89	.91	42.49 (18.44)	51.88 (20.30)	.55**	.41**	−.38**	.51**	.31**	−.36**	.42**	.29**	−.34**		.53**	−.50**
	11. SATAQ-R Internalization (1 ÷ 5)	.79	.87	2.59 (0.81)	2.86 (0.87)	.19*	.54**	−.22*	.24*	.59**	−.24*	.33**	.63**	−.22*	.31**		−.49**
	12. RSES Self-esteem (10 ÷ 40)	.78	.85	32.95 (5.07)	30.07 (5.77)	−.43**	−.42**	.59**	−.42**	−.36**	.54**	−.40**	−.36**	.69**	−.53**	−.24**

Note. For Pearson’s correlations, values for males are below the diagonal and values for females are above the diagonal. EAT = Eating Attitudes Test; SATAQ = Sociocultural Attitudes Towards Appearance Questionnaire; RSES = Rosenberg Self-Esteem Scale.

p < .05. **p < .01.

Measures

As mentioned previously, internal consistency coefficients for each measure at each time point layered by females and males are presented in Table 1.

Eating Attitudes Test (EAT-26)

The EAT-26 (Garner, Olmsted, Bohr, & Garfinkel, 1982; Spanish adaptation by Gandarillas, Zorrilla, Sepúlveda, & Muñoz, 2003) contains 26 questions that measure attitudes, feelings, and behaviors that are characteristic of individuals with disordered eating (e.g., in Spanish and English, respectively, Procuro no comer aunque tenga hambre/I avoid eating when I am hungry). The items provide six response options ranging from 1 (never) to 6 (always), which are usually later recoded as 0-1-2-3. In our case, and following Barr, Petit, Vigna, and Prior (2001), the total score using a continuous scoring system of 1 to 6 for never, rarely, sometimes, often, very often, and always was obtained from the sum of these original answers, reversed when necessary. In doing so, the total score can range between 26 and 156, and higher scores indicate more disordered eating.

Sociocultural Attitudes Towards Appearance Questionnaire-R (SATAQ-R)

SATAQ-R (Cusumano & Thompson, 1997; Spanish adaptation by Calado, 2008) contains 21 items adapted for each gender, which provide five response options ranging from 1 (strongly disagree) to 5 (strongly agree). This self-report questionnaire assesses awareness and internalization to strive toward social standards of appearance. In this study, we used the internalization scale (10 items) which has been defined as deep incorporation or acceptance of these values affects one’s attitudes or personal behavior (e.g., in Spanish and English, respectively, Tiendo a comparar mi cuerpo con el de las estrellas que aparecen en TV y en las películas/I tend to compare my body to TV and movie stars). Scale scores are obtained from the average of the corresponding items, and higher scores indicate more (worse) internalization.

Rosenberg Self-Esteem Scale (RSES)

RSES (Rosenberg, 1965; Spanish adaptation by Guillén, Quero, Baños, & Botella, 1999) contains 10 items assessing global self-esteem (e.g., in Spanish and English, respectively, Creo tener varias cualidades buenas/I feel that I have a number of good qualities). The items provide four response options ranging from 1 (strongly disagree) to 4 (strongly agree). The total score is obtained from the sum of the items, and higher scores indicate better self-esteem.

Body mass index

Measurements of weight and height were taken in situ at baseline (T1) and 16 months later (T4), and BMI for both time points was calculated using the standard method (weight/height in kg/m²).

Procedure

The Ethics Committee of our institution and the Municipal Institute of Health and Social Welfare of the cities of Manresa and Terrassa approved the study. The participants of this sample were students who came from all control schools of a wider study assessing the effectiveness of an eating disorder prevention program. Informed written consent from parents and oral consent from adolescents were obtained. Participants answered the battery assessment individually, anonymously, and confidentially during normal lesson time at school.

Statistical Analysis

The statistical analyses were carried out with Mplus7.11 (Muthén & Muthén, 1998-2013). Path analysis with continuous manifest variables was utilized to test the bidirectional models, by fitting an autoregressive cross-lagged model, between disordered eating attitudes (EAT), internalization (SATAQ), and self-esteem (RSES), adjusted by BMI with its inclusion as time-varying covariates at T1 and T4. Overall, this model accounts for three types of effects: (a) EAT, SATAQ, and RSES scores were allowed to correlate cross-sectionally (contemporaneous associations) at each time point; and each construct measured during the last three waves of data collection was regressed (b) on the same measure one (short-autoregressive paths) and when necessary two (long-autoregressive paths) waves earlier; and (c) on the other two measures one wave earlier (cross-lagged paths). Thus, these cross-lags allow simultaneous examination of paths between constructs while controlling for effects in the opposite direction (Sharpe et al., 2017).

A multigroup approach was conducted using the Robust Maximum Likelihood (MLR) method of estimation, which is robust to nonnormality for continuous variables and is also a full-information method (Enders & Bandalos, 2001; Graham, 2009). Therefore, all participants with information for at least one assessment point were included. Goodness of fit was assessed using the common fit indices: scaled χ², comparative fit index (CFI), Tucker-Lewis index (TLI), and root mean square error of approximation (RMSEA). Reasonable and adequate model fit was considered for CFI and TLI above .90 or .95, and RMSEA below .08 or .06, respectively (Hu & Bentler, 1999). First, a baseline model was established with all parameters freely estimated between genders, and then invariance for paths across males and females was tested using the scaled chi-square difference (Bryant & Satorra, 2012) for nested models (α level set at .05).

Results

The initial model tested in both genders jointly (Figure 1) included the three types of paths detailed in the statistical analysis section: (a) contemporaneous associations between all three measures at each time point, (b) short-autoregressive paths for each construct, and (b) cross-lagged paths between pairs of measures. This model had rather poor fit to the data, χ²(96) = 207.362, CFI = .938, TLI = .877, RMSEA = .093. After establishing long-autoregressive paths by regressing each construct at the last two waves of data collection on the same measure two waves earlier, goodness-of-fit indices for the baseline multigroup model (configural model) were acceptable, χ²(78) = 133.125, CFI = .971, TLI = .933, RMSEA = .068. When fixing all parameters to be equivalent between genders, complete equivalence was not achieved, Δχ²(45) = 72.115, p = .002. After freeing three parameter based on modification indices (starting with the largest sensitive modification; i.e., that with the greatest decrement in chi-square value), partial equivalence was then attained, Δχ²(42) = 55.138, p = .084, and this final model showed a satisfactory fit, χ²(120) = 187.129, CFI = .965, TLI = .947, RMSEA = .061. Figure 2 shows separated diagrams for each type of path of the final model obtained. The nonequivalent paths between genders were (a) the contemporaneous association between SATAQ and RSES at T2, which was null for females (−.11) and negative for males (−.40) (Figure 2a); (b) the short-autoregressive path from EAT at T3 to EAT at T4, which was higher for females (.54) than for males (.28) (Figure 2b); and (c) the cross-lag path from SATAQ at T1 to EAT at T2, which was positive for females (.24) and null for males (−.01) (Figure 2c).

Figure 1.

Initial model including contemporaneous associations and autoregressive and cross-lagged paths.

Figure 2.

Final multigroup model for males/females between disordered eating attitudes (EAT), internalization (SATAQ) and self-esteem (RSES): Standardized coefficients for contemporaneous associations (a), autoregressive (b) and cross-lagged (c) paths.

Contemporaneous effects between measures at each time point were all statistically significant, ranging from .19 to .60 in absolute value, except for the aforementioned path (SATAQ between RSES at T2 in females being −.11, p = .222), with positive associations between EAT and SATAQ and negative between both EAT and SATAQ with RSES (Figure 2a). Short-autoregressive effects in the .48 to .72 range, except for the aforementioned path (EAT at T3 to EAT at T4 in males being .28), and long-autoregressive effects in the .14 to .33 range were observed for all measures (Figure 2b). After controlling for the stability of each measure over time and the contemporaneous relationships between measures at each time point, five statistically significant cross-lagged effects were observed; four were equivalent between genders: RSES at T1 predicted lower EAT at T2, SATAQ at T2 predicted both higher EAT at T3 and lower RSES at T3, and EAT at T3 predicted lower RSES at T4; finally, SATAQ at T1 predicted higher EAT at T2 only in females (Figure 2c).

Discussion

The aim of the present study was to examine the potential bidirectional longitudinal associations between disordered eating, the internalization of beauty ideals and self-esteem in Spanish adolescents, and the equivalence of these relationships across both genders.

We found equivalent bidirectional associations between self-esteem and disordered eating in different periods of the follow-up in both genders. Lower self-esteem at T1 predicted higher levels of disordered eating at T2. This association fades in subsequent follow-ups. It is possible that self-esteem is a relevant risk factor for disordered eating only at the beginning of adolescence, where the earlier physical changes of puberty and body and shape concerns start. These results partially support our hypothesis and support the incorporation of self-esteem in the explanatory models of disordered eating and body image dissatisfaction as a precursor factor of these disorders in early adolescence. Other variables such as depressive symptoms may participate in this relation, as pointed out by Rodgers et al. (2014) in their biopsychosocial model. Low self-esteem can lead directly to a negative evaluation of the body and to negative feelings about body image (Ricciardelli & Yager, 2016) and can increase the risk of developing disordered eating. In addition, low self-esteem can increase adolescents’ vulnerability to social pressure to attain the ideal body and increase internalization and social comparison (Rodgers et al., 2014).

Disordered eating at T3 predicted lower self-esteem at T4 in accordance with previous studies (Rodgers et al., 2011). It is possible that in this period the attitudes and behaviors that characterize disordered eating (dieting, binge eating, exercise to lose weight, body dissatisfaction, weight concerns and weight importance) can lead to a decrease in self-esteem. This association was not observed from T2 to T3. Disordered eating emerges with greater intensity from 15 years of age onward (Stice, Shaw, & Marti, 2007), and it is very likely that at the age of our community sample at T2 and T3 the disordered eating was less intense and interfered less with the general psychological functioning and with self-esteem in particular.

Contrary to our hypothesis, self-esteem predicted disordered eating similarly in males and females. It is possible that in males the current pressure to reach the muscular ideal contributes to increasing the importance of the physical appearance and that this becomes an important parameter of their self-worth (Crocker et al., 2002), fostering disordered eating.

The internalization of beauty ideals predicted disordered eating in a unidirectional way in two different periods of the follow-up in accordance with previous studies (Espinoza et al., 2010; Francisco et al., 2015; Rohde et al., 2015). The internalization of beauty ideals at T1 predicted disordered eating at T2 only in females and at T2 predicted disordered eating at T3 in both genders. These results suggest that in the beginning of adolescence, females seem to respond to the internalization of beauty ideals with greater intensity and may be more vulnerable to develop disordered eating (Knauss et al., 2007). Subsequently around the age of 14, the adolescents of both genders who have internalized a demanding and unrealistic beauty ideal would base their self-worth on their physical appearance, leading to a decrease in their self-esteem, when there is a discrepancy between their real and ideal body. Further research is needed to validate these associations in males and to clarify how self-esteem and the internalization of the muscular beauty ideal are related with the development of concerns about body image and disordered eating aimed at losing weight and building muscle in males (Lavender, Brown, & Murray, 2017).

Low self-esteem did not predict an increase in internalization in the follow-ups, contrary to our hypothesis and previous findings (Francisco et al., 2015; Mischner et al., 2013). It is possible that in this age group other variables modulate this relation. According to the biopsychosocial model of disordered eating (Rodgers et al., 2014), low self-esteem and high levels of depressive symptoms (negative affect) can increase early adolescents’ vulnerability to internalizing the ideal of thinness. Longitudinal studies are required to assess if these variables predict the internalization of the muscular ideal in males.

The cross-lagged effects were unstable between disordered eating, internalization of beauty ideals, and self-esteem during the follow-up period, which suggests that these predictive associations may be influenced by age and by pubertal changes. The cross-lagged effects were small, which indicates that other variables may play an important role in the development of disordered eating in both genders, such as social body comparison, concern about body image (e.g., Puccio et al., 2017), and especially body dissatisfaction (Sharpe et al., 2017). Body dissatisfaction was identified as a robust early predictor (at 13 years of age) of future eating disorders (Rohde et al., 2015) and is associated reciprocally with depressive symptoms (Sharpe et al., 2017) and with self-esteem from the beginning of adolescence in both genders (Wichstrøm & von Soest, 2016).

In our study, the age of 14 (T3) was a critical age in both genders for interrelations between the internalization of beauty ideals, self-esteem, and disordered eating. At this age, pubertal, cognitive, and interpersonal changes increase adolescents’ ability to reach a more abstract characterization of themselves, the influence of their peers increases (Rohde et al., 2015), and they may become more aware of and vulnerable to pressures to attain sociocultural beauty ideals. The age of 14 has been suggested as a potential age to carry out preventive programs with vulnerable girls (Rohde et al., 2015). However, in our study we observed a predictive association at age 13 between self-esteem and disordered eating in both genders and between internalization of beauty ideals and disordered eating only in females. These findings can help understand the appearance of gender differences in the development of disordered eating and suggest the need to prevent the internalization of thin ideal in females at the beginning of early adolescence. Studies confirming these findings are required, which contribute toward identifying the appropriate age for prevention programs, considering gender differences (González et al., 2015).

Furthermore, a moderate temporal stability was observed in self-esteem and the internalization of beauty ideals between 12 and 15 years of age in both genders, pointing to the need to implement prevention programs based on the critique of aesthetic ideals (Espinoza, Penelo, & Raich, 2013; González, Penelo, Gutiérrez, & Raich, 2011; McLean, Paxton, & Wertheim, 2016; Mora et al., 2015) and on strengthening self-esteem for males and females, during early adolescence (Rohde et al., 2015), before the stabilization process of these predictors occurs. Disordered eating shows a moderate temporal stability between 12 and 15 years of age, in both genders. We observe a nonequivalent magnitude of the association between genders in the short-autoregressive effect for disordered eating from T3 to T4 (around 14 years of age), in which males showed a much weaker association. It is possible that at this age, pubertal changes bring males closer to beauty ideals than females (Bearman, Presnell, Martinez, & Stice, 2006). Another possible explanation is that the EAT was not sensitive to the disordered eating attitudes and behaviors that adolescents develop to reach the male ideal that currently emphasizes both muscles and thinness (Eisenberg, Wall, & Neumark-Sztainer, 2012; Lavender et al., 2017).

The limitations of our study are related with the intended sampling technique and a limited follow-up that prevented us from determining the course of the bidirectional associations during late adolescence. Another limitation of this study was the impossibility of including biological variables in the model assessed, such as the pubertal age (which was only asked in T1), and an evaluation of unhealthy muscle-enhancing behaviors (Eisenberg et al., 2012). There is a need for studies which assess the bidirectional associations between other important predictive factors which trigger and maintain the symptomatology of disordered eating, such as depressive symptoms, anxiety, and concern about body image (Puccio et al., 2017).

This study provides relevant information referring to the potential course of risk factors, because it provides information about males and females and it was obtained in short time intervals during a stage of adolescence characterized by intense and fast physical and psychological changes. Our results revealed unidirectional and bidirectional associations between important predictive factors of the “onset” of clinically significant disordered eating symptomatology (Dakanalis, Timko, et al., 2016) with more similarities than differences between genders, and they provide information about the appropriate age to carry out prevention programs. This information contributes toward completing the explanatory models of disordered eating for both genders.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was partially supported by grants PSI2013-47212-P from the Spanish Ministry of Economy and Competitiveness and 2014-SGR-1673 from the Autonomous Government of Catalonia.

ORCID iD

Paola Espinoza

Author Biographies

Paola Espinoza, PhD, is a researcher collaborator at the Department of Assessment and Intervention of Body Image at the Universitat Autònoma de Barcelona. Her main research is on risk factors of eating disorders and evaluation of the effectiveness of programs to prevent eating disturbances and body dissatisfaction.

Eva Penelo, PhD, is an associate professor of the unit of methodology at the Universitat Autònoma de Barcelona (Spain). Her research interests include applied psychometrics, instrument validation, and structural equation modeling in health sciences.

Marisol Mora, PhD, is professor at the Departament de Psicologia Clínica i de la Salut, Universitat Autònoma de Barcelona, Spain. She has developed research on eating disorders and body image. She has published in these areas in peer reviewed journals.

Rita Francisco, PhD, is an assistant professor of the School of Human Sciences at the Universidade Católica Portuguesa. Her research interests include eating disorders, well-being, and adjustment of adolescents in different social situations. She has clinical experience in the contexts of individual and family therapy with adolescents and adults.

Marcela L. González, PhD, is psychotherapist in Institut de Trastorns Alimentaris and a researcher collaborator at the Department of Assessment and Intervention of Body Image at the Universitat Autònoma de Barcelona. Her main research is evaluation of the effectiveness of programs to prevent eating disturbances and body dissatisfaction.

Rosa M. Raich is professor of cognitive/behavioral psychology. She has developed research on eating disorders and body image during more than 30 years and is author of more than 200 scientific publications.

References

Barr

S. I.

Petit

M. A.

Vigna

Y. M.

Prior

J. C.

(2001). Eating attitudes and habitual calcium intake in peripubertal girls are associated with initial bone mineral content and its change over 2 years. Journal of Bone and Mineral Research, 16, 940-947. doi:10.1359/jbmr.2001.16.5.940

Bearman

S. K.

Presnell

Martinez

Stice

(2006). The skinny on body dissatisfaction: A longitudinal study of adolescent girls and boys. Journal of Youth and Adolescence, 35, 229-241. doi:10.1007/s10964-005-9010-9

Bryant

F. B.

Satorra

(2012). Principles and practice of scaled difference chi-square testing. Structural Equation Modeling, 19, 372-398. doi:10.1080/10705511.2012.687671

Calado

(2008). Influence of the mass media on body image and disordered eating in secondary school students (Unpublished doctoral thesis). University of Vigo, Ourense, Spain.

Cole

T. J.

Bellizzi

M. C.

Flegal

K. M.

Dietz

W. H.

(2000). Establishing a standard definition for child overweight and obesity worldwide: International survey. British Medical Journal, 320, 1240-1243.

Cole

T. J.

Flegal

K. M.

Nicholls

Jackson

A. A.

(2007). Body mass index cut offs to define thinness in children and adolescents: International survey. British Medical Journal, 335, 194-197.

Crocker

Sommers

Luhtanen

(2002). Hopes dashed and dreams fulfilled: Contingencies of self-worth in the graduate school admissions process. Personality and Social Psychology Bulletin, 28, 1275-1286. doi: 10.1177/01461672022812012

Cusumano

D. L.

Thompson

J. K.

(1997). Body image and body shape ideals in magazines: Exposure, awareness, and internalization. Sex Roles, 37, 701-721. doi:10.1007/BF02936336

Dakanalis

Pla-Sanjuanelo

Caslini

Volpato

Riva

Clerici

Carrà

(2016). Predicting onset and maintenance of men’s eating disorders. International Journal of Clinical and Health Psychology, 16, 247-255. doi:10.1016/j.ijchp.2016.05.002

10.

Dakanalis

Timko

Serino

Riva

Clerici

Carrà

(2016). Prospective psychosocial predictors of onset and cessation of eating pathology amongst college women. European Eating Disorders Review, 24, 251-256. doi:10.1002/erv.2433

11.

Eisenberg

M. E.

Wall

Neumark-Sztainer

(2012). Muscle-enhancing behaviors among adolescent girls and boys. Pediatrics, 130, 1019-1026. doi:10.1542/peds.2012-0095

12.

Enders

C. K.

Bandalos

D. L.

(2001). The relative performance of full information maximum likelihood estimation for missing data in structural equation models. Structural Equation Modeling, 8, 430-457. doi:10.1207/S15328007SEM0803_5

13.

Espinoza

Penelo

Raich

R. M.

(2010). Disordered eating behaviors and body image in a longitudinal pilot study of adolescent girls: What happens 2 years later? Body Image, 7, 70-73. doi:10.1016/j.bodyim.2009.09.002

14.

Espinoza

Penelo

Raich

R. M.

(2013). Prevention programme for eating disturbances in adolescents. Is their effect on body image maintained at 30 months later? Body Image, 10, 175-181. doi:10.1016/j.bodyim.2012.11.004

15.

Francisco

Espinoza

González

M. L.

Penelo

Mora

Rosés

Raich

R. M.

(2015). Body dissatisfaction and disordered eating among Portuguese and Spanish adolescents: The role of individual characteristics and internalization of sociocultural ideals. Journal of Adolescence, 41, 7-16. doi:10.1016/j.adolescence.2015.02.004

16.

Gandarillas

Zorrilla

Sepúlveda

A. R.

Muñoz

P. E.

(2003). Trastornos del comportamiento alimentario [Eating disorders] (Documentos Técnico de Salud Pública, Nº 85). Madrid, Spain: Instituto de Salud Pública. Retrieved from http://www.publicaciones-isp.org/productos/d085.pdf

17.

Garner

D. M.

Olmsted

M. P.

Bohr

Garfinkel

P. E.

(1982). The Eating Attitudes Test: Psychometric features and clinical correlates. Psychological Medicine, 12, 871-878. doi:10.1017/S0033291700049163

18.

González

Mora

Penelo

Goddard

Treasure

Raich

R. M.

(2015). Gender differences found in a qualitative study of a disordered eating prevention programme: What do boys have to say? Journal of Health Psychology, 20, 858-874. doi:10.1177/1359105315573426

19.

González

Penelo

Gutiérrez

Raich

R. M.

(2011). Disordered eating prevention programme in schools: A 30-month follow-up. European Eating Disorders Review, 19, 349-356. doi:10.1002/erv.1102

20.

Graham

J. W.

(2009). Missing data analysis: Making it work in the real world. Annual Review of Psychology, 60, 549-576. doi:10.1146/annurev.psych.58.110405.085530

21.

Guillén

Quero

Baños

Botella

(1999, November). Datos psicométricos y normativos del cuestionario de autoestima de Rosenberg en una muestra española [Psychometric and normative data of the Rosenberg self-esteem questionnaire in a Spanish sample]. Paper presented at the I Congreso Iberoamericano de Psicología Clínica y de la Salud, Granada, Spain.

22.

Hansson

Daukantaitė

Johnsson

(2016). Typical patterns of disordered eating among Swedish adolescents: Associations with emotion dysregulation, depression, and self-esteem. Journal of Eating Disorders, 4, Article 28. doi:10.1186/s40337-016-0122-2

23.

Hollingshead

A. B.

(2011). Four factor index of social status. Yale Journal of Sociology, 8, 21-51.

24.

L. T.

Bentler

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55. doi:10.1080/10705519909540118

25.

Knauss

Paxton

S. J.

Alsaker

F. D.

(2007). Relationships amongst body dissatisfaction, internalization of the media body ideal and perceived pressure from media in adolescent girls and boys. Body Image, 4, 353-360. doi:10.1016/j.bodyim.2007.06.007

26.

Lavender

J. M.

Brown

T. A.

Murray

S. B.

(2017). Men, muscles, and eating disorders: An overview of traditional and muscularity-oriented disordered eating. Current Psychiatry Reports, 19, Article 32. doi:10.1007/s11920-017-0787-5

27.

Loth

K. A.

MacLehose

Bucchianeri

Crow

Neumark-Sztainer

(2014). Predictors of dieting and disordered eating behaviors from adolescence to young adulthood. Journal of Adolescent Health, 55, 705-712. doi:10.1016/j.jadohealth.2014.04.016

28.

McLean

S. A.

Paxton

S. J.

Wertheim

E. H.

(2016). The role of media literacy in body dissatisfaction and disordered eating: A systematic review. Body Image, 19, 9-23. doi:10.1016/j.bodyim.2016.08.002

29.

Mischner

I. H.

van Schie

H. T.

Engels

R. C.

(2013). Breaking the circle: Challenging Western sociocultural norms for appearance influences young women’s attention to appearance-related media. Body Image, 10, 316-325. doi:10.1016/j.bodyim.2013.02.005

30.

Mora

Penelo

Gutiérrez

Espinoza

González

Raich

R. M.

(2015). Assessment of two school-based programs to prevent universal eating disorders: Media literacy and theatre-based methodology in Spanish adolescent boys and girls. The Scientific World Journal, 2015, Article 328753. doi:10.1155/2015/328753

31.

Muthén

L. K.

Muthén

B. O.

(1998–2013). Mplus user’s guide (7th ed.). Los Angeles, CA: Author.

32.

Puccio

Fuller-Tyszkiewicz

Youssef

Mitchell

Byrne

Allen

Krug

(2017). Longitudinal bi-directional effects of disordered eating, depression and anxiety. European Eating Disorders Review, 25, 351-358. doi:10.1002/erv.2525

33.

Ricciardelli

L. A.

McCabe

M. P.

(2001). Self-esteem and negative affect as moderators of sociocultural influences on body dissatisfaction, strategies to decrease weight, and strategies to increase muscles among adolescent boys and girls. Sex Roles, 44, 3189-3207. doi:10.1023/A:1010955120359

34.

Ricciardelli

L. A.

McCabe

M. P.

Holt

K. E.

Finemore

(2003). A biopsychosocial model for understanding body image and body change strategies among children. Journal of Applied Developmental Psychology, 24, 475-495. doi:10.1016/S0193-3973(03)00070-4

35.

Ricciardelli

L. A.

Yager

(2016). Adolescence and body image: From development to preventing dissatisfaction. Abingdon, UK: Routledge.

36.

Rodgers

Chabrol

Paxton

S. J.

(2011). An exploration of the tripartite influence model of body dissatisfaction and disordered eating among Australian and French college women. Body Image, 8, 208-215. doi:10.1016/j.bodyim.2011.04.009

37.

Rodgers

Paxton

S. J.

McLean

S. A.

(2014). A biopsychosocial model of body image concerns and disordered eating in early adolescent girls. Journal of Youth and Adolescence, 43, 814-823. doi:10.1007/s10964-013-0013-7

38.

Rohde

Stice

Marti

C. N.

(2015). Development and predictive effects of eating disorder risk factors during adolescence: Implications for prevention efforts. The International Journal of Eating Disorders, 48, 187-198. doi:10.1002/eat.22270

39.

Rosenberg

(1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

40.

Sharpe

Patalay

Choo

Wall

Mason

Goldschmidt

Neumark-Sztainer

(2017). Bidirectional associations between body dissatisfaction and depressive symptoms from adolescence through early adulthood. Development and Psychopathology, 16, 1-12. doi:10.1017/S0954579417001663

41.

Stice

Shaw

Marti

C. N.

(2007). A meta-analytic review of eating disorder prevention programs: Encouraging findings. Annual Review of Psychology, 3, 207-231. doi:10.1146/annurev.clinpsy.3.022806.091447

42.

Thompson

J. K.

Heinberg

L. J.

Altabe

Tantleff-Dunn

(1999). Exacting beauty: Theory, assessment, and treatment of body image disturbance. Washington, DC: American Psychological Association.

43.

Wichstrøm

von Soest

(2016). Reciprocal relations between body satisfaction and self-esteem: A large 13-year prospective study of adolescents. Journal of Adolescence, 47, 16-27. doi:10.1016/j.adolescence.2015.12.003