Eating Behaviors 18 (2015) 1–6

Contents lists available at ScienceDirect

Eating Behaviors

Comparison of disordered eating symptoms and emotion regulation difficulties between female college athletes and non-athletes Gena Wollenberg a,⁎, Lenka H. Shriver b, Gail E. Gates a a b

Department of Nutritional Sciences, Oklahoma State University, 301 Human Sciences, Stillwater, OK 74078, USA Department of Nutrition, University of North Carolina Greensboro, 319 College Avenue, 318 Stone Building, Greensboro, NC 27412, USA

a r t i c l e

i n f o

Article history: Received 5 August 2014 Received in revised form 18 January 2015 Accepted 19 March 2015 Available online 28 March 2015 Keywords: Disordered eating Emotion regulation Female college athletes College non-athletes

a b s t r a c t The purpose of the study was to compare the prevalence of disordered eating between female college athletes and non-athletes and explore emotion regulation as a potential mediator of the link between participation in athletics and disordered eating symptoms. Data for this cross-sectional study came from 527 college students in a mid-western state of the USA in fall of 2013 (376 non-athletes and 151 athletes). Disordered eating symptoms and emotion regulation were assessed utilizing the Eating Attitudes Test and the Difficulties with Emotion Regulation Scale in a survey-based format. The prevalence of disordered eating was higher in nonathletes (16.5%, vs. 6.6%; X2 = 62.8; p b .05). Non-athletes reported more signs and symptoms of disordered eating than athletes (p b .01). A linear regression approach indicated a statistically significant indirect effect (0.63, CI95 = 0.18, 1.20) of athletic-status on disordered eating via emotion regulation; however, this effect did not reach practical significance. Our findings show that female athletes in our sample were somewhat protected from disordered eating compared to non-athletes, but the mechanism of this relationship is unclear. A further in-depth examination of other factors, such as self-esteem and body satisfaction, that may have contributed to this finding is warranted utilizing a large sample of female college students and athletes representing a variety of sports. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Dieting, binge eating and preoccupation with food are examples of pathological eating behaviors and attitudes known as disordered eating (DE) (DePalma et al., 2002; Lowry et al., 2000; Torstveit, Rosenvinge, & Sundgot-Borgen, 2008). DE may in some, but not all, progress into an eating disorder over time (Anderson & Petrie, 2012; Neumark-Sztainer, Wall, Larson, Eisenberg, & Louth, 2011). Clinically diagnosed eating disorders are complex psychiatric conditions (i.e., anorexia and bulimia) that require a multidisciplinary and long-term treatment approach (American Psychiatric Association (APA), 2013). Because of the complex nature of eating disorders, efforts to promote healthy eating behaviors and attitudes are critical to optimize individuals' physical and psychological wellbeing before a clinical eating disorder develops (Ozier & Henry, 2011). Young females are at a substantially higher risk for eating disturbances compared to males (Fortes, Kakeshita, Almeida, Gomes, & Ferreira, 2014; Martinsen, Brantland-Sandra, Eriksson, & Sundgot-Borgen, 2010; Sira & Pawlak, 2010). Previous studies have reported that between 11% and 56% of females in late adolescence and young adulthood engage in ⁎ Corresponding author. Tel.: +1 405 385 2335. E-mail addresses: [email protected] (G. Wollenberg), [email protected] (L.H. Shriver), [email protected] (G.E. Gates).

http://dx.doi.org/10.1016/j.eatbeh.2015.03.008 1471-0153/© 2015 Elsevier Ltd. All rights reserved.

some type of dysregulated eating behaviors (Croll, Neumark-Sztainer, Story, & Ireland, 2002; Hoerr, Bokram, Lugo, Bivins, & Keast, 2002; Sira & Pawlak, 2010). Previous research has suggested several attributes and/or behaviors as DE risk factors, including a family history of eating disorders, low self-esteem, weigh/appearance concerns, certain personality traits (i.e., being a perfectionist or extroverted), negative body image, poor emotional well being, and high stress (Croll et al., 2002; Jacobi, Hayward, de Zwaan, Kraemer, & Agras, 2004; Striegel-Moore & Bulik, 2007). The societal emphasis on thinness for females, strongly perpetrated through media, has been identified as an underlying contributor to weight concerns, poor body image and desire to lose weight that are strongly associated with DE behaviors among many girls and women (Bratland-Sanda & Sundgot-Borgen, 2013; Polivy & Herman, 2002). Overall, strong evidence suggests that eating disturbances are multifactorial, with unique interactions between personal, environmental and genetic factors (Ghaderi & Scott, 2001; Striegel-Moore & Bulik, 2007). Female college students represent a particularly vulnerable population for engaging in unhealthy eating patterns (Fortes et al., 2014; Krahn, Kurth, Gomberg, & Drewnowski, 2005). The period between ages 18 and 21, a typical age of attending college, has been identified as the time of peak onset of clinical eating disorders (Berg, Frazier, & Sherr, 2009). Recent studies indicated that college females report engaging in dysregulated eating frequently and also report using a wide range of pathological behaviors coupled with negative attitudes related

2

G. Wollenberg et al. / Eating Behaviors 18 (2015) 1–6

to either eating or weight (Bratland-Sanda & Sundgot-Borgen, 2013; Fitzsimmons-Craft, Harney, Brownstone, Higgins, & Bardone-Cone, 2012). These trends may be potentially explained by college students facing a variety of stressors as they transition from adolescence to adulthood, such as dealing with college-level academic expectations, creating new work and social relationships and being away from home (Cooley & Toray, 2001; Fitzsimmons-Craft et al., 2012; French & Jeffery, 1994). Exercise offers multiple benefits to individuals across the age and gender groups, including young females (Costarelli, Demerzi, & Stamou, 2009; Hausenblas & Downs, 2001; Varnes et al., 2013). In addition to improved physical health and fitness, a recent systematic review by Varnes et al. (2013) indicated that girls and women involved in sports had higher body satisfaction and more positive body image than those who did not participate in athletics. The study, however, found that the benefit of being involved in athletics might be reduced in athletes based on their level of competition and/or type of sport (Varnes et al., 2013). In fact, some studies found that female athletes were at a greater risk of DE than general population of females and thus research in this area remains inconclusive (Sundgot-Borgen & Torstveit, 2004; Torstveit & Sundgot-Borgen, 2005). Excessive training, frequent food restriction and extreme dieting are examples of dysregulated behaviors that have been reported by female athletes in previous research (De Bruin, Oudejans, & Bakker, 2007; Monthuy-Blanc, Maiano, & Therme, 2010). Given the common belief within the athletic environment that low body weight and body fat enhance performance, athletes may engage in unhealthy patterns to achieve lower weight or body fat under pressures created by coaches, parents, and/or female athletes themselves (Barrack, Ackerman, & Gibbs, 2013; Holm-Denoma, Scaringi, Gordon, Van Orden, & Joiner, 2009). In the area of dysregulated eating behaviors, emotion regulation represents an emerging construct (Costarelli et al., 2009; Han & Pistole, 2014). Sim and Zeman (2006) were among the first to publish data identifying emotional status as a potential predictor of DE in a sample of young females. Difficulties with emotion regulation have also been linked to DE, specifically to binge eating, in a study by Whiteside et al. (2007). In a sample of 695 college students (both females and males), those with poor access to emotion regulation strategies and greater difficulty identifying emotional states were more likely to engage in binge-eating behaviors (Whiteside et al., 2007). This association was stronger than the contributions of gender, food restriction and weight/shape concerns to the overall variance in the binge eating behaviors in this sample. A few studies have found similar associations between DE and emotion regulation among men. For instance, Lavender and Anderson (2010) indicated that DE behaviors and body dissatisfaction in college male students were predicted by difficulties with emotion regulation. In their sample, young men with lower ability to accept their emotions and those without adequate emotion regulation strategies, reported greater DE scores. All together, findings of these studies point to the potentially important influence of emotion regulation on DE patterns among young college-age individuals. Despite the proposed associations between sports participation, emotion regulation, and DE patterns in previous studies, research examining these constructs has been limited in the at-risk population of female college students and none of the studies have assessed emotion regulation of athletes compared to non-athletes (Fortes & Ferreire, 2011; Haase, 2011; Holm-Denoma et al., 2009; Reinking & Alexander, 2005). The main purpose of this study was to examine athletic participation and emotion regulation as potential predictors of DE in a sample of female college students. First, we hypothesized that participation in athletics and greater emotion regulation difficulties will predict greater DE symptoms in our sample of young females. Second, we hypothesized that emotion regulation mediates the link between athletic status (athlete vs. non-athlete) and DE symptoms.

2. Materials and methods 2.1. Participants and recruitment procedures Data for this study were collected from a sample of female college students in a NCAA Division I university in a mid-western state of the U.S. The university's Institutional Review Board approved the study protocol prior to any data collection. Additionally, official approval was obtained from the head sports physician who was responsible for the medical care of all athletes at the university. Non-athletes (students who were not members of any of the Division I athletic teams at the university) were recruited from six undergraduate classes across campus with the instructors' prior approval. The Primary Investigator (PI) informed interested students about the study during previously scheduled classroom visits. Written informed consents were collected from participants before data collection began. There was no penalty or incentive to participate in the study. The inclusion criteria for nonathletes included: 1) not being a member of an official university women's athletic team; 2) being recreationally active only (i.e., no training for a significant athletic event such as a marathon or half-marathon; and 3) being 18 or older. Female athletes were recruited through certified and trained athletic trainers working with the individual teams. Participants were informed about the purpose, nature and details of the study by the athletic trainers during a team meeting, without the presence of the coaching staff to minimize potential perceived pressure for athletes to participate in the study. A written informed consent form was obtained from interested participants prior to any data collection. Participants were recruited from all women's athletic teams at the university and included the following sports: soccer, cross-country, track and field, basketball, cheer/pom/dance, equestrian, tennis, golf and softball. The inclusion criteria for this subsample included: 1) being a member of an official university women's athletic team; 2) being 18 or older; and 3) participating in regular team practices and activities at the time of the study (i.e., no recent injuries or illnesses). 2.2. Study procedures and research instruments The data related to disordered eating behaviors and emotion regulation were collected in a survey format utilizing the Eating Attitudes Test (EAT-26) and the Difficulties in Emotion Regulation Scale (DERS). Participants also answered demographic questions and were asked to self-report their actual weight, height, and desired weight. Additional questions included topics about the participants' menstrual cycle, previous ED diagnosis, family history of ED, and exercise-related questions to determine whether participants met inclusion criteria for the study. The PI and/or athletic trainers provided an envelope containing the survey and participants were given approximately 30 min to complete it. No personal identifiers linking individual participants to their responses were included in the survey. Participants were instructed to place the completed survey in a sealed envelope and return it to the PI or the team athletic trainers. If the PI could not be present during data collection the sealed envelopes were collected by the athletic trainers or instructors and given to the PI within 24 h of data collection. Disordered eating symptoms were assessed utilizing the EAT-26, which is one of the most commonly used instruments for examining disordered eating symptoms in college-aged populations (Fortes et al., 2014; Shriver, Betts, & Payton, 2009). The EAT-26 contains three subscales and yields a global score and three individual subscale scores: 1) Dieting; 2) Bulimia and Food Preoccupation, and 3) Oral Control (Garner & Garfinkel, 1979). This test contains 4 supplemental behavioral questions that refer to an individual's eating habits in the past 6 months. One of the 4 questions asks if the individual has ever been treated for an eating disorder. EAT-26 is scored using six answer options (0 = never, rarely, or sometimes, 1 = often, 2 = usually, and 3 = always). Item number 26 (i.e., “I enjoy trying new foods.”) was analyzed using reversed

G. Wollenberg et al. / Eating Behaviors 18 (2015) 1–6

scoring. A total score of 20 or greater or a “yes” answer to 1 supplemental question identifies the individual at risk for an eating disorder. No risk for an eating disorder was identified by a score of less than 20 or a “no” answers to all of the supplemental questions (Fortes et al., 2014; Haase, 2011). The EAT-26 has been widely used and validated with good internal consistency in previous research with a Cronbach's alpha of .90 (Garner, Olmsted, Bohr, & Farfinkel, 1982). In the current study, the EAT-26 also indicated good to acceptable internal consistency, similar to previous studies (Hund & Espelage, 2006; Voelker, Gould, & Reel, 2014) with α = .88 (total scale), α = .87 for the Dieting subscale, α = .68 for the Bulimia subscale, and α = .60 for the Oral subscale. Emotion regulation was assessed using the DERS. It is a 36-item selfreported questionnaire developed to assess different dimensions of emotion regulation (Gratz & Roemer, 2004). The DERS includes 6 subscales: 1) Non-acceptance of emotion responses (i.e., “When I'm upset, I feel guilty for feeling that way;” score range of 6–30; 2) Difficulties engaging in goal directed behavior (i.e., “When I'm upset, I have difficulty concentrating;” score range of 5–25; 3) Impulse control difficulties (i.e., “When I'm upset, I lose control over my behaviors;” score range of 6–30; 4) Lack of emotion awareness (i.e., “I am attentive to my feelings;” score range of 6–30; 5) Limited access to emotion regulation strategies (i.e., “When I'm upset, I believe that I'll end up feeling very depressed,” score range of 8–40; and 6) Lack of emotion clarity (i.e., “I have difficulty making sense out of my feelings,” score range of 5–25. The DERS is scored using five answer options with the following scores (1 = almost, 2 = sometimes, 3 = about half the time, 4 = most of the time, 5 = almost always). Reversed-scoring is used for eleven items as appropriate (Gratz & Roemer, 2004). The total possible score ranges from 36 to 180, with higher scores indicating poorer emotion regulation (Han & Pistole, 2014). Because currently there is no standardized cutoff score for emotion regulation, the DERS score is used as a continuous variable or may be compared to mean scores of females from previous research studies (Gratz & Roemer, 2004). In our sample, the Cronbach's alpha was .91 for the total DERS, .89 for the Non-accept subscale, .84 for the Goals subscale, .82 for the Impulse subscale, .48 for the Clarity subscale (note: not included in the final analysis due to low alpha), .81 for the Aware subscale and .80 for the Strategies subscale. 2.3. Statistical analysis Descriptive statistics, including means and standard deviations, frequencies and ranges were computed to describe the athletic and non-athletic sub-sample in terms of their demographic and anthropometric characteristics. Furthermore, descriptive statistics served to summarize the prevalence of disorder eating behaviors and symptoms characteristics in the entire sample and within each subsample (athletes vs. non-athletes). The proportion of individuals at-risk for eating disordered (using the EAT-26 cutoff score of 20 and answer of “yes” to the supplemental questions) was compared between athletes and non-athletes using Chi-square statistics for dichotomous variables. Differences in DE and emotion regulation were examined by the type of sport among athletes, classifying athletes into two categories: 1) lean (n = 74; endurance/esthetic sports-cross country running, cheerleading, pom, equestrian); and 2) non-lean (n = 77; basketball, softball, soccer, tennis, golf, track). In preliminary analyses, Pearson's bivariate correlations were utilized to explore associations between disordered eating scores (i.e., total and subscale EAT scores) and emotion regulation (total and subscale DERS scores). Analysis of covariance (ANCOVA) was conducted to identify potential differences in disordered eating between athletes and non-athletes, while controlling for a family history of eating disorders. A linear regression model was used to examine potential predictors of disordered eating in the entire sample of female college students. The independent variables in the regression model included a family history of eating disorders, the 6 individual DERS subscales,

3

and the female athletic status (athlete vs. non-athlete). The athletic status was dummy-coded for inclusion in the regression model (0 = athletes; 1 = non-athletes). The overall score on EAT-26 served as the dependent variable in this regression analysis. Additionally, emotion regulation (total DERS score) was tested as a potential mediator of indirect effects of athletic status on DE. The EAT-26 total score, as well as Dieting, Bulimia and Food Preoccupation, and Oral Control subscale scores were tested as potential mediators in four separate models. A linear regression approach was used to test for direct effects of athletic status on the mediator, and for direct effects of athletic status on DE. Family history of DE was entered as a covariate at each step of model testing. Bootstrapping (1000 samples per model) was used to estimate indirect effects and to generate bias-corrected 95% confidence intervals in order to interpret statistical significance. All analyses were performed using the Statistical Package for Social Sciences (SPSS for PC; 20.0) and the level of significance was set at p b .05 for all tests. Mediation analyses were conducted using the PROCESS macro for SPSS created by Dr. Andrew Hayes (Hayes, 2013).

3. Results A total of 540 female college students volunteered to participate in the study and completed the survey (n = 389 non-athletes; n = 151 athletes). At the time of the study, 183 female athletes were officially members of the university teams, thus their participation rate was 83%. Thirteen participants (all non-athletes) returned incomplete surveys and were excluded from the final analyses (97% completion rate). The final analyses were conducted utilizing complete data from 527 participants. Female athletes were significantly taller but weighed less than the non-athletes (Table 1). The mean Body Mass Index (BMI) was within a healthy weight range of 18.5–24.99 in both groups, but athletes had a significantly lower mean BMI than non-athletes. The detailed demographic and anthropometric characteristics of the participants are presented in Table 1. Non-athletes scored significantly higher on the EAT-26 as well as on its three subscales (Table 2). A significantly greater proportion of nonathletes were categorized as “at-risk” for eating disorders compared to athletes (X2 (1, N = 527) = 8.89, p b .05) (Table 3). ANCOVA revealed that non-athletes scored significantly higher on EAT-26 than athletes even after controlling for a family history of eating disorders (p b .001). Non-athletes had greater overall difficulties regulating their emotions than athletes, scoring higher on the total DERS as well as on the Goals and Strategies subscales. In the overall sample, greater scores on the total DERS were positively associated with higher scores on total EAT-26 and all of its subscales (total EAT-26; r = 0.29; Dieting; r = 0.26; Bulimia; r = 0.31; Oral; r = 0.16; p b 0.001). The correlation was also significant with all of the DERS subscales except for the Aware subscale which was non-significant. No significant differences in the variables

Table 1 Characteristics of the female college students in the sample (n = 527). Variable

Age (year) Height (in.) Weight (lb) BMIa (kg/m2) Ideal body weight (IBW)b (lb) BMI based on IBW (kg/m2)

Athletes

Non-athletes

n = 151

n = 376

Mean ± SD

Mean ± SD

19.50 ± 1 65.76 ± 3* 136.35 ± 22* 22.11 ± 3*** 131.62 ± 19 21.35 ± 3

19.83 ± 3 65.11 ± 3 141.39 ± 31 23.39 ± 5 129.14 ± 19 21.39 ± 3

Significant difference between athletes and non-athletes (***p ≤ 0.001, *p ≤ 0.05). a BMI: body mass index measured by self reported height and weight. b IBW: perceived ideal body weight (weight participant felt that would be ideal for them).

4

G. Wollenberg et al. / Eating Behaviors 18 (2015) 1–6

Table 2 Scores on the EAT-26 (including 3 subscales) and distribution of participants in at-risk for disordered eating by athlete- vs. non-athlete status. Variable

EAT-26a Dietingb Bulimiac Oral controld EAT score of b20e EAT score of ≥20

Table 4 Linear regression analysis predicting disordered eating as a function of family history of eating disorders, emotion regulation difficulties, and athletic status (n = 527).

Athletes

Non-athletes

Variable

B

SE B

β

n = 151

n = 376

Athletic statusa Family history of EDb Non-acceptc Goalsd Impulsee Awaref Strategiesg

3.81 −3.90 .422 .084 −.031 −.078 .206

.924 2.24 .103 .109 .145 .109 .113

.172⁎⁎ .071 .223⁎⁎⁎ .041 −.013 −.034 .126

Mean ± SD

Mean ± SD

6.9 ± 7.7*** 5.0 ± 6.0*** 0.6 ± 1.3*** 1.4 ± 1.8*** 141 (93.4%) 10 (6.6%)*

11.8 ± 10.5 8.2 ± 7.7 1.1 ± 2.2 2.5 ± 2.9 314 (83.5%) 62 (16.5%)

*p ≤ 0.05; ***p ≤ 0.001; Chi-square statistics. a Range of scores for total EAT-26: 0–78. b Range of scores for EAT subscale of dieting: 0–39. c Range of scores for EAT subscale of bulimia: 0–6. d Range of scores for EAT subscale of oral control: 0–7. e Score of b20 = no risk for disordered eating; score of ≥20 = at risk for disordered eating.

⁎ p b 0.05. ⁎⁎⁎ p b 0.001. a Athletic status: 1) athlete; 0) non-athlete. b ED: eating disorders. c Non-accept subscale. d Goals subscale. e Impulse subscale. f Aware subscale. g Strategies subscale.

4. Discussion of interests, including emotion regulation and EAT-26, scores were detected among athletes by type of sport (lean versus non-lean). Family history of eating disorders, emotion regulation (including the individual subscales) and athletic status (athlete vs. non-athlete) were tested as potential predictors of DE in a linear regression model. The regression model was significant and predicted 15.4% of total variance in participants' EAT-26 scores (R2 = .15; F = 11.75; p b .001). Athletic status (athletes or non-athletes) and the emotion regulation subscale of Non-accept were found to be significant individual predictors of DE in the sample (p b .001) (Table 4). Full results of the mediation analyses are presented in Table 5. Athletic status was a significant predictor of the proposed mediator (β = 4.84, SE = 1.20, p b .05). Athletic status also remained a significant independent predictor of DE when the proposed mediator also was entered as an independent predictor (p b 0.05). Family history of eating disorders did not achieve significance as an independent predictor at any stage of the models. Estimates of indirect effects were positive and achieved statistical significance in all four models, suggesting higher DE total scores and subscale scores for non-athletes.

Table 3 Differences in Difficulties in Emotion Regulation Scores (DERS) between athletes (n = 151) and non-athletes (n = 376). Variable

Difficulties in emotion Regulation (DERS)a Non-acceptb Goals

c

Impulse

d

Awaree Strategies

f

Athletes

Non-athletes

Mean ± SD %

Mean ± SD %

70.9 ± 17.9 39%

75.8 ± 21.8⁎ 42%

11.5 ± 4.8 38% 12.0 ± 4.4 48% 9.8 ± 3.5 33% 14.7 ± 4.4 50% 13.3 ± 5.1 33%

12.8 ± 5.5 43% 13.7 ± 5.1⁎⁎⁎ 55% 10.4 ± 4.3 35% 13.6 ± 4.4 45% 15.2 ± 6.5⁎⁎⁎ 38%

⁎ p ≤ 0.05. ⁎⁎⁎ p ≤ 0.001. a Range of scores for total DERS: 36–180. b Range of scores for the DERS subscale of non-accept: 6–30. c Range of scores for the DERS subscale of goals; 5–25. d Range of scores for the DERS subscale of impulse; 6–30. e Range of scores for the DERS subscale of aware, 6–30. f Range of scores for the DERS subscale of strategies: 8–40.

The purpose of this study was to examine the prevalence of disordered eating among female college students and to explore sports participation and emotion regulation as potential predictors of DE in this population. Although a few studies have suggested that female athletes may be at an increased risk of DE compared to general population (Fortes et al., 2014; Sundgot-Borgen & Torstveit, 2004), this hypothesis was not supported in our study. This finding was upheld even after DE symptoms were compared between athletes in lean vs. non-lean sports, with no significant differences found by type of sport. Non-athletes in our sample had a higher prevalence of DE than athletes and they also reported greater difficulties with regulating their emotions, skills that have been implicated in both DE and body dissatisfaction in previous studies with college-aged populations (Lavender & Anderson, 2010; Sim & Zeman, 2006). Although the prevalence of DE in our overall sample was lower than in some previous studies (Thompson & Gabriel, 2004), a relatively high proportion of female non-athletes engaged in dysregulated eating and reported a variety of unhealthy behaviors that ranged from avoiding carbohydrates to taking laxatives. For instance, well over half of the participants (64%) thought about burning calories while exercising and as many as 62% reported they were terrified about being overweight. Also, nearly half of them

Table 5 Mediation models and estimates of indirect effects for predicting disordered eating (EAT-26) as a function of emotion regulation difficulties, athletic status and family history of eating disorders (n = 527). Mediator

Predictors

β

SE

t

p

IE

LLCI

ULCI

Total EAT-26

Constant Emotion reg.a Athletic statusb Family historyc Constant Emotion reg. Athletic status Family history Constant Emotion reg. Athletic status Family history Constant Emotion reg. Athletic status Family history

4.76 0.13 4.27 −3.66 4.34 0.08 2.86 −2.74 −0.58 0.03 0.40 −0.44 1.00 0.02 1.02 −0.47

4.76 0.02 0.91 2.26 3.56 0.01 0.68 1.69 0.95 0.00 0.18 0.45 1.30 0.01 0.25 0.62

1.00 6.60 4.68 −1.62 1.22 5.71 4.19 −1.62 −0.61 7.10 2.17 −0.98 0.77 3.34 4.08 −0.76

0.32 0.00 0.00 0.11 0.22 0.00 0.00 0.10 0.54 0.00 0.03 0.33 0.44 0.00 0.00 0.44

0.63

0.17

1.20

0.41

0.11

0.78

0.14

0.04

0.28

0.09

0.02

0.21

Dieting

Bulimia

Oral control

p b 0.05; ***p b 0.001. a Athletic status: 1) athlete; 0) non-athlete. b Emotion reg.: DERS total score. c Family history: family history of eating disorders.

G. Wollenberg et al. / Eating Behaviors 18 (2015) 1–6

reported that they were aware of the calorie content of the foods they consumed. Our findings suggest that female college athletes in our sample, regardless of type of sport, were somewhat protected from DE compared to female students who were not involved in athletics. The current study is consistent with some previous comparative research that was conducted with adolescent or adult athletes (Martinsen et al., 2010; Michou & Costarelli, 2011; Rosendahl, Bormann, Aschenbrenner, Aschenbrenner, & Strauss, 2009). In a study of elite adolescent athletes, Martinsen et al. (2010) found that controls (adolescent non-athletes) reported greater DE symptoms, greater body dissatisfaction and a stronger drive for thinness than their athletic counterparts. These findings are similar to the trends found in our sample where non-athletes reported greater dissatisfaction with their current body weight and scored higher not only on the total EAT-26, but also on all of the subscales that assess a variety of unhealthy behaviors/attitudes, ranging from dieting and binge eating to pre-occupation with food or weight. Our findings thus support the idea that participation in athletics may be associated with other attributes that are protective of DE but were not assessed in the current study. For example, female athletes may have greater body satisfaction or self-esteem, both of which represent cognitive attitudes that are linked to lower risk of eating disorders (Armstrong & Oomen-Early, 2009; Hausenblas & Downs, 2001). Emotion regulation difficulty was positively associated with disordered eating patterns in our study. The “non-accept” dimension of emotion regulation, which signals an individual's inability to accept emotion responses (e.g., When I am upset, I feel ashamed for feeling this way) was identified as a significant predictor of EAT-26 in the overall sample. A handful of previous studies have found similar results when examining eating patterns in relation to emotion regulation (Han & Pistole, 2014; Lavender & Anderson, 2010). Han and Pistole (2014) found that females who experienced troubles identifying and accepting emotions were more likely to engage in disordered eating behaviors, and specifically binge eating. Despite the associations between emotion regulation and DE in previous research and the current study, the findings should be interpreted with caution. While emotion regulation difficulties did partially mediate the link between athletic status and DE in our sample, the mediation effect was relatively small. Furthermore, it did not reach practical significance in terms of the EAT-26 scores. Because athletic status was a dichotomous variable, the indirect effect estimates also represented the estimated total potential magnitude of the mediational effect. In all cases, this represented a very small (less than one point) difference in predicted disordered eating outcome scores. Thus, the mechanism of the relationship between athletic status and DE is likely influenced by additional factors that need to be examined in further research (e.g., self-esteem, body satisfaction). The current study adds to the limited knowledge on the complex associations between disordered eating, sports participation and emotion regulation in young female college students. Since athletic status predicted lower DE in the sample, independent of family history of eating disorders and emotion regulation difficulties, characteristics of female college athletes should be examined further to identify potential mediators through which athletic status serves as a protective measure from DE risks. For instance, body image, body satisfaction as well as selfesteem should be examined in future studies with this population. Given the inconsistent findings on the prevalence of DE among female athletes and general population of females, there is also a need to further investigate potential correlates and predictors of DE with large and representative samples of female college athletes that would allow for comparison by types of sport (Bonci et al., 2008; Smolak, Murnen, & Ruble, 2000). The current study has several strengths that include utilization of a large sample of female college students from a Division I university, and a high participation rate of eligible participants. Furthermore, utilization of validated assessment tools for disordered eating symptoms

5

and emotion regulation also serves as one of the strengths of the study. An anonymous survey format allowed participants to disclose behaviors and feelings they may have considered uncomfortable otherwise and may not have fully disclosed in a one-on-one interview format. The current study also has several limitations that should be noted. First, we utilized a convenience sample of female college students. Although we successfully recruited a large sample, the participant pool should not be assumed to be representative of the national population of female college students as geographical, ethnic and other differences were not accounted for in the study. Second, data were self-reported and some participants may not have disclosed truthful information for different reasons (i.e., denial of eating disturbance, social bias). However, this limitation of self-reported data is well acknowledged in research on sensitive topics such as disordered eating (Bratland-Sanda & Sundgot-Borgen, 2013; Sundgot-Borgen & Torstveit, 2004). Third, athletes with relatively high reported prevalence of DE (e.g., gymnastics, diving) were not represented in our sample because the university did not have such teams at the time of the study. Thus, future studies in this area should include a sample that allows for comparisons between athletes and non-athletes by type of sport. 5. Conclusions Female college athletes in our sample reported lower prevalence of DE and fewer difficulties with regulating their emotions than nonathletes. Emotion regulation was not a powerful mediator of the link between sports participation and DE, thus it appears that female athletes were more protected from dysregulated eating due to other attributes related to their athletic status. Given the devastating effects of clinical eating disorders, routine screenings for early signs of dysregulated eating should be implemented with young female university students wherever feasible. Since sports participation was a significant predictor of lower DE score regardless of family history of eating disorders and emotion regulation, health professionals working with college females should emphasize the importance of physical activity for both physiological and psychological benefits. Future studies should examine physiological, psychological and/or cognitive characteristics of female college athletes in order to identify factors that mediate the link between athletic-status and DE in this population. Role of funding sources The study and the manuscript preparation were not supported by any internal or external funding sources. Contributors All authors contributed substantially and meaningfully to this research study and the final manuscript. Dr. Wollenberg along with Dr. Shriver designed the study and its methodology, with Dr. Gates serving as a methodological and statistical consultant through the study design process. Dr. Wollenberg led the recruitment and data collection phases of the study with the guidance from Drs. Shriver and Gates. All authors participated in final data analysis. All authors participated in the preparation of this manuscript and they have approved the final manuscript. Conflict of interest The authors declare no conflicts of interest associated with this research study. Acknowledgments Authors wish to thank the athletes who volunteered for the study and the university athletic trainers who assisted with the recruitment and data collection from athletes in this study. We would like to thank Dr. Jeffrey Labban, a statistical analyst in the School of Health and Human Sciences Office of Research at the University of North Carolina Greensboro, for his consulting services in terms of data analyses and interpretation of the findings in this study.

References American Psychiatric Association (2013). The Diagnostic and Statistical Manual of Mental Disorders: DSM 5. bookpointUS.

6

G. Wollenberg et al. / Eating Behaviors 18 (2015) 1–6

Anderson, C., & Petrie, T.A. (2012). Prevalence of disordered eating and pathogenic weight control behaviors among NCAA division I female collegiate gymnasts and swimmers. Research Quarterly for Exercise and Sport, 83(1), 120. Armstrong, S., & Oomen-Early, J. (2009). Social connectedness, self-esteem, and depression symptomatology among collegiate athletes versus nonathletes. Journal of American College Health, 57(5), 521–526. Barrack, M.T., Ackerman, K.E., & Gibbs, J.C. (2013). Update on the female athlete triad. Current Reviews in Musculoskeletal Medicine, 6, 195–204. Berg, K.B., Frazier, P., & Sherr, L. (2009). Change in eating disorder attitudes and behaviors in college women: Prevalence and predictors. Eating Behaviors, 10, 137–142. Bonci, C.M., Bonci, L.J., Granger, L.R., Johnson, C.L., Maline, R.M., Miline, L.W., et al. (2008). National Athletic Trainers' Association position statement: Preventing, detecting, and managing disordered eating in athletes. Journal of Athletic Training, 43(1), 80–108. Bratland-Sanda, S., & Sundgot-Borgen, J. (2013). Eating disorders in athletes: Overview of prevalence, risk factors and recommendations for prevention and treatment. European Journal of Sport Science, 13(5), 499–508. Cooley, E., & Toray, T. (2001). Body image and personality predictors of eating disorder symptoms during the college years. International Journal of Eating Disorders, 30(1), 28–36. Costarelli, V., Demerzi, M., & Stamou, D. (2009). Disordered eating attitudes in relation to body image and emotion intelligence in young women. Journal of Human Nutrition and Dietetics, 22(3), 239–245. Croll, J., Neumark-Sztainer, D., Story, M., & Ireland, M. (2002). Prevalence and risk and protective factors related to disordered eating behaviors among adolescents: Relationship to gender and ethnicity. Journal of Adolescent Health, 31, 166–175. De Bruin, A.P., Oudejans, R.R., & Bakker, F.C. (2007). Dieting and body image in aesthetic sports: A comparison of Dutch female gymnasts and non-aesthetic sport participants. Psychology of Sport and Exercise, 8(4), 507–520. DePalma, M., Koszewski, W., Romani, W., Case, J.G., Zuiderhof, N.J., & McCoy, P.M. (2002). Identifying college athletes at risk for pathogenic eating. British Journal of Sports Medicine, 36(1), 45–50. Fitzsimmons-Craft, E.E., Harney, M.B., Brownstone, L.M., Higgins, M.K., & Bardone-Cone, A.M. (2012). Examining social physique anxiety and disordered eating in college women. The roles of social comparison and body surveillance. Appetite, 59(3), 796–805. Fortes, L.S., & Ferreire, M.E.C. (2011). Comparison of body dissatisfaction and inappropriate eating behavior in adolescent athletes of different sports. Brazilian Journal of Physical Education and Sport, 25(4), 707–716. Fortes, L.S., Kakeshita, I.S., Almeida, S.S., Gomes, A.R., & Ferreira, M.C. (2014). Eating behaviors in youths: A comparison between female and male athletes and non-athletes. Scandinavian Journal of Medicine & Science in Sports, 24(1), e62–e68. French, S.A., & Jeffery, R.W. (1994). Consequence of dieting to lose wither: Effect on physical and metal health. Health Psychology, 13, 195–212. Garner, D.M., & Garfinkel, P.E. (1979). The Eating Attitudes Test: An index of the symptoms of anorexia nervosa. Psychological Medicine, 9(02), 273–279. Garner, D.M., Olmsted, M.P., Bohr, Y., & Farfinkel, P.E. (1982). The Eating Attitudes Test: Psychometric features and clinical correlates. Psychological Medicine, 12(4), 871–878. Ghaderi, A., & Scott, B. (2001). Prevalence, incidence and prospective risk factors for eating disorders. Acta Psychiatrica Scandinavica, 104(2), 122–130. Gratz, K., & Roemer, L. (2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the Difficulties in Emotion Regulation Scale. Journal of Psychopathology and Behavioral Assessment, 26(1), 41–54. Haase, A.M. (2011). Weight perception in female athletes: Associations with disordered eating correlates and behavior. Eating Behaviors, 12(1), 64–67. Han, S., & Pistole, M.C. (2014). College student binge eating: Insecure attachment and emotion regulation. Journal of College Student Development, 55(1), 16–29. Hausenblas, H.A., & Downs, D.S. (2001). Comparison of body image between athletes and nonathletes: A meta-analytic review. Journal of Applied Sport Psychology, 13(3), 323–339. Hayes, A.F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York: The Guilford Press. Hoerr, S.L., Bokram, R., Lugo, B., Bivins, T., & Keast, D.R. (2002). Risk for disordered eating relates to both gender and ethnicity for college students. Journal of the American College of Nutrition, 21(4), 307–314. Holm-Denoma, J.M., Scaringi, V., Gordon, K.H., Van Orden, K.A., & Joiner, T.E. (2009). Eating disorder symptoms among undergraduate varsity athletes, club athletes, independent exercisers and non-exercisers. International Journal of Eating Disorders, 12(1), 47–53. Hund, A.R., & Espelage, D.L. (2006). Childhood emotional abuse and disordered eating among undergraduate females: Mediating influence of alexithymia and distress. Child Abuse & Neglect, 30(4), 393–407.

Jacobi, C., Hayward, C., de Zwaan, M., Kraemer, H.C., & Agras, W.S. (2004). Coming to terms with risk factors for eating disorders: Application of risk terminology and suggestions for a general taxonomy. Psychological Bulletin, 130(1), 19–23. Krahn, D.D., Kurth, C.L., Gomberg, E., & Drewnowski, A. (2005). Pathological dieting and alcohol use in college women — A continuum of behaviors. Eating Behaviors, 6, 43–52. Lavender, J.M., & Anderson, D.A. (2010). Contribution of emotion regulation difficulties to disordered eating and body dissatisfaction in college men. International Journal of Eating Disorders, 43(4), 352–357. Lowry, R., Galuska, D.A., Fulton, J.E., Wechsler, H., Kann, L., & Collins, J.L. (2000). Physical activity, food choices and weight management goals and practices among U.S. college students. American Journal of Preventive Medicine, 18(1), 11–27. Martinsen, M., Brantland-Sandra, S., Eriksson, A.K., & Sundgot-Borgen, J. (2010). Dieting to win or to be thin? A study of dieting and disordered eating among adolescent elite athletes and non-athletes. British Journal of Sports Medicine, 44(1), 70–76. Michou, M., & Costarelli, V. (2011). Disordered Eating Attitudes in relation to anxiety levels, self-esteem and body image in female basketball players. Journal of Exercise Science & Fitness, 9(2), 109–115. Monthuy-Blanc, J., Maiano, C., & Therme, P. (2010). Prevalence of eating disorders symptoms in nonelite ballet dancers and basketball players: An exploratory and controlled study among French adolescent girls. Revue d'Épidémiologie et de Santé Publique, 58(6), 415–424. Neumark-Sztainer, D., Wall, M., Larson, N.I., Eisenberg, M.E., & Louth, K. (2011). Dieting and disordered eating behaviors from adolescence to young adulthood: Findings from a 10-year longitudinal study. Journal of the American Dietetic Association, 111(7), 1004–1011. Ozier, A.D., & Henry, B.W. (2011). Position of the American Dietetic Association: Nutrition intervention in the treatment of eating disorders. Journal of the American Dietetic Association, 111(8), 1236–1241. Polivy, J., & Herman, C.P. (2002). Causes of eating disorders. Annual Review of Psychology, 53, 187–213. Reinking, M.F., & Alexander, L.E. (2005). Prevalence of disordered-eating behaviors in undergraduate female collegiate athletes and nonathletes. Journal of Athletic Training, 40(1), 47–51. Rosendahl, J., Bormann, B., Aschenbrenner, K., Aschenbrenner, F., & Strauss, B. (2009). Dieting and disordered eating in German high school athletes and non-athletes. Scandinavian Journal of Medicine & Science in Sports, 19(5), 731–739. Shriver, L.H., Betts, N.M., & Payton, M.E. (2009). Changes in body weight, body composition, and eating attitudes in high school wrestlers. International Journal of Sport Nutrition & Exercise Metabolism, 19(4), 424–432. Sim, L., & Zeman, J. (2006). The contribution of emotion regulation to body dissatisfaction and disordered eating in early adolescent girls. Journal of Youth and Adolescence, 35(2), 207–216. Sira, N., & Pawlak, R. (2010). Prevalence of overweight and obesity, and dieting attitudes among Caucasian and African American college students in eastern North Carolina: A cross-sectional survey. Journal of Nutrition Research and Practice, 4(1), 36–42. Smolak, L.O., Murnen, S.K., & Ruble, A.E. (2000). Female athletes and eating problems; a meta-analysis. International Journal of Eating Disorders, 27, 371–380. Striegel-Moore, R.H., & Bulik, C.M. (2007). Risk factors for eating disorders. American Psychologist, 62(3), 181. Sundgot-Borgen, J., & Torstveit, M.K. (2004). Prevalence of eating disorders in elite athletes is higher than in the general population. Clinical Journal of Sport Medicine, 14(1), 25–32. Thompson, S.H., & Gabriel, M. (2004). Risk factors for the female athlete triad among female collegiate and noncollegiate athletes. The Physical Educator, 61(4), 200. Torstveit, M. K., & Sundgot-Borgen, J. (2005). The female athlete triad exists in both elite athletes and controls. Medicine and Science in Sports and Exercise, 37(9), 1449. Torstveit, M.K., Rosenvinge, J.H., & Sundgot-Borgen, J. (2008). Prevalence of eating disorders and the predictive power of risk models in female elite athletes: A controlled study. Scandinavian Journal of Medicine & Science in Sports, 18(1), 108–118. Varnes, J.R., Stellefson, M.L., Janelle, C.M., Dorman, S.M., Dodd, V., & Miller, M.D. (2013). A systematic review of studies comparing body image concerns among female college athletes and non-athletes, 1997–2012. Body Image, 10(4), 421–432. Voelker, D.K., Gould, D., & Reel, J.J. (2014). Prevalence and correlates of disordered eating in female figure skaters. Psychology of Sport and Exercise. Whiteside, U., Chen, E., Neighbors, C., Hunter, D., Lo, T., & Larimer, M. (2007). Difficulties regulating emotions: Do binge eaters have fewer strategies to modulate and tolerate negative affect? Eating Behaviors, 8(2), 162–169.