Eating Behaviors 14 (2013) 529–532
Contents lists available at ScienceDirect
Eating Behaviors
Dieting, exercise, and intuitive eating among early adolescents Jordan Moy a, Trent A. Petrie a,⁎, Sally Dockendorff a, Christy Greenleaf b, Scott Martin b a b
Department of Psychology, University of North Texas, Denton, TX 76203, United States Department of Kinesiology, Health Promotion, & Recreation, University of North Texas, Denton, TX 76203, United States
a r t i c l e
i n f o
Article history: Received 15 December 2012 Received in revised form 8 May 2013 Accepted 25 June 2013 Available online 2 July 2013 Keywords: Intuitive eating Dieting Exercising Body mass index Adolescence
a b s t r a c t Dieting to lose weight, with its focus on restriction of caloric intake, may disrupt intuitive eating processes, though other forms of weight loss, such as exercising, which do not emphasize food may not. In a sample of 669 middle school boys and 708 girls, regardless of sex or exercising, dieting was related to feeling less free to eat what was wanted and to eating more to soothe emotions than to satisfy actual physical hunger. Exercising, independent of dieting, was associated with feeling less permission to eat what was wanted, but also eating to satisfy physical hunger as opposed to coping with emotional distress. Overall, girls were more aware and trusting of their bodily hunger and satiety cues than boys, but when boys were exercising, they scored similarly to girls on this dimension. These findings suggest that different weight loss approaches – dieting vs. exercising – have unique relationships to young adolescents' intuitive eating and these associations tend to be stable across sex. Longitudinal studies now are needed to examine how dieting that begins in childhood or early adolescence might have long-term effects on the progression of intuitive eating. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Intuitive eating is based on physiological hunger and satiety cues rather than situational and emotional cues, thus its central premise is that individuals innately are able to stay in tune with their bodies' natural signals about hunger and fullness (Tylka, 2006). From this perspective, individuals are thought to be able to regulate their food intake in a healthy manner, eating what they want, when they want it, beginning when they become hungry, and stopping when they feel full. When individuals eat intuitively, they generally consume more nutritious foods and are able to maintain a weight that is healthy for their height and sex (Eneli, Crum, & Tylka, 2008). Dieting, which may include unhealthy eating practices such as an extreme restriction of overall caloric intake and/or eating only certain types of food (such as those deemed “good” foods by the dieter), may disrupt intuitive eating processes and has been associated with long-term weight gain (Neumark-Sztainer, Wall, Story, & Standish, 2012). When individuals “diet” they generally have to (a) exert cognitive control over their internal signals about hunger and satiety, often overriding their physiology, and (b) limit what, when, and how much they eat. That is, dieters may forbid consumption of certain “bad” (nondiet) foods and allow themselves to eat only those foods that are considered “good” and align with their dietary restrictions (Polivy & Herman, 1985). Over time, such rigid cognitive and behavioral controls may undermine individuals' understanding of their ⁎ Corresponding author at: Department of Psychology, 1155 Union Circle #311280, University of North Texas, Denton, 76203. Tel.: +1 940 565 4718. E-mail address: [email protected] (T.A. Petrie). 1471-0153/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.eatbeh.2013.06.014
bodies' signals, disrupt their ability to know what they want to eat (and the permission they give themselves to eat it), and lead them to eat more in response to emotions than physical hunger, which often encompasses a binge. The controls that are associated with dieting can cause problems for all age groups, but may be particularly difficult for children and early adolescents because of food- and eating-related pressures within the family environment and their relationship to emotional eating (e.g., Galloway, Farrow, & Martz, 2010; Kroon Van Diest & Tylka, 2010). Two types of controlling feeding behaviors have been identified – food restriction and pressure-to-eat (Eneli et al., 2008). Parents and other caregivers may restrict children's eating, ranging from what to how much they are allowed to consume, particularly when it comes to what are considered to be unhealthy (or “bad”) foods. Parents/caregivers also may pressure children to eat what they consider to be healthy (or “good”) foods, such as fruits and vegetables. Both of these controlling approaches to food and eating require constant monitoring on the part of parents/caregivers, and can interfere with children's intuitive awareness of and trust in their physical responses to hunger and fullness (Eneli et al., 2008). In many cases parents/caregivers may be unaware of their controlling behaviors and how their involvement may lead to dysfunctional eating in their children. For example, Carper, Fisher, and Birch (2000) found that (a) only 26% of parents believed they pressured their daughters about feeding, whereas 61% of the girls felt pressured, and (b) 51% of parents reported restricting certain foods from their daughters, though 63% of the girls thought this occurred. Girls who reported that their parents restricted what they could eat were three times more likely to experience high levels of disinhibited emotional eating
530
J. Moy et al. / Eating Behaviors 14 (2013) 529–532
(i.e., lack of restraint in relation to food) than girls who were able to eat with fewer dietary restrictions. Although parents/caregivers can play a positive role in helping children develop healthy eating patterns, many (intentionally or unintentionally) pressure their children about their food choices, often restricting access to food considered to be unhealthy and pressuring them to eat foods that they believe “should” be eaten (e.g., fruits, vegetables, whole grains). Such pressures and behaviors can interfere with adaptive eating processes (Eneli et al., 2008), and ultimately lead to unhealthy beliefs about food and eating that are carried forward and transformed into self-imposed rules about what foods are acceptable, how much one should eat, and whether or not food can serve as an emotional salve. In fact, it is the consistent use of dieting behaviors (e.g., eating very little, skipping meals, diet pills), as opposed to other unhealthy weight control behaviors (e.g., vomiting, laxatives), which are associated with long-term weight gain (Neumark-Sztainer et al., 2012). And, healthy forms of weight management, specifically exercising, would not be expected to lead to disruptions in intuitive eating because increased levels of physical activity are not related directly to food, do not involve pressures about what (or how much) to eat, and actually can increase individuals' satiety responses to meals (King et al., 2009). Thus, in the current study, the relations between dieting (i.e., decreasing caloric intake), exercising (i.e., increasing cardiorespiratory activity levels), and intuitive eating were examined. Specifically, we expected that early adolescents who were dieting to manage their weight would report more disruptions in their intuitive eating than those who were not. Even though the purpose of increasing physical activity levels also is to manage weight, we expected that there would be no differences in level of intuitive eating between those early adolescents who were exercising compared to those who were not. Because participants' sex has been associated with intuitive eating in adolescents (Dockendorff, Petrie, Greenleaf, & Martin, 2012) and girls have been found to diet more frequently than boys (Neumark-Sztainer, Wall, Larson, & Eisenberg, 2011), we considered it in relation to dieting and exercise as well. 2. Method 2.1. Participants Participants were 669 boys (Mage = 12.31 years, SD = ± 0.98) and 708 girls (Mage = 12.09 years, SD = ± 0.92) drawn from six middle schools in a suburban school district located in the Southwestern U.S. Although race/ethnicity data were not provided on 448 students in the sample, of the remaining, 65.3% were White/ Non-Hispanic, 22.6% White/Hispanic, 9.5% Black/NonHispanic, 3.7% Asian American/Pacific Islander, and 0.8% Native American, which was consistent with the overall racial/ethnic make-up of the school district; 44.1% were in sixth grade, 32% in seventh grade, and 23.8% in eighth grade. 2.2. Measures 2.2.1. Demographics During each school's annual physical fitness testing, height and weight were objectively measured by physical education teachers and converted to body mass index (BMI; kg/m2). BMI, as well as the participants' age, grade, and race/ethnicity, were provided by the school district following approval from their internal research committee. All data provided by the school district were coded solely by the students' school identification numbers; neither names nor other identifying information were used. 2.2.2. Intuitive eating The 16-item Intuitive Eating Scale for Adolescents-Revised (IESA-R) is a revision of the 17-item Intuitive Eating Scale for Adolescent
(Dockendorff et al., 2012; Tylka, 2006) and comprised of three factors: Unconditional Permission to Eat (5 items; i.e., eating when hungry and eating what food is desired; “I let myself eat ‘junk food’ when I want to”); Eating for Physical Rather Than Emotional Reasons (5 items, i.e., using food to satisfy physiological hunger drives rather than as a coping mechanism for emotional distress; “I eat when I am feeling sad or stressed, even when I am not physically hungry” – reversed scored item); and Reliance on Cues of Hunger and Satiety (6 items, i.e., awareness of physiological hunger and satiety cues and trusting these cues to manage eating; “I trust my body to let me know how much to eat”). For each item, participants respond using a 5-point scale that ranges from 1, strongly disagree, to 5, strongly agree. Total factor scores are the mean; higher scores indicate more positive intuitive eating. Internal consistency reliabilities (i.e., Cronbach's alphas) for the three factors were: 0.78 (Unconditional Permission), 0.89 (Eating for Physical Reasons), and 0.88 (Reliance on Physiological Cues). Dockendorff et al. provided extensive information about the factors' validity. 2.2.3. Disordered eating The 26-item Children's Eating Attitudes Test (ChEAT; Garner, Olmsted, Bohr, & Garfinkel, 1982; Smolak & Levine, 1994) is the children's version of the EAT and assesses a variety of attitudes and behaviors associated with anorexia nervosa and bulimia nervosa. On items such as “I think a lot about having fat on my body,” the adolescents responded on a 6-point scale, ranging from 0, always, to 6, never. For each item, the most symptomatic response is scored as a 3, the next most symptomatic as a 2, the next most symptomatic as a 1, and the remaining three responses are scored as 0. Total score is the sum of the items, and can range from 0, no symptoms, to 78, high level of symptoms. Smolak and Levine reported internal consistency reliability (Cronbach's alpha) of .87; alpha for the current study was .73. Smolak and Levine provided extensive information about the scale's validity. In the current study, the ChEAT was used screen out adolescents who might have an eating disorder because dieting and intuitive eating have been strongly related to disordered eating as measured by the EAT (Tylka & Wilcox, 2006). Based on a score of 20 or highter on the ChEAT, 4.1% of the boys and 6% of the girls from the initial sample were eliminated from all analyses. 2.2.4. Dieting Consistent with how dieting has been assessed and classified in adolescents (e.g., Neumark-Sztainer et al., 2011; Neumark-Sztainer et al., 2012), participants responded to a single item question, indicating whether they had, during the last three months, dieted (i.e., “Eaten fewer calories [example – eat less food”]) specifically to manage their weight. Response options were never, seldom, sometimes, often, and always, and were dichotomized into nondieters (never or seldom, n = 862) and dieters (other responses, n = 515). 2.2.5. Exercising Consistent with how exercising has been measured and classified in adolescents (e.g., Neumark-Sztainer, Paxton, Hannan, Haines, & Story, 2006), the boys and girls indicated, through a single item question, whether they had, during the last three months, exercised (i.e., “Increased cardiorespiratory physical activity, such as running”) specifically to manage their weight. Response options were never, seldom, sometimes, often, and always, and were dichotomized into nonexercisers (never or seldom, n = 186) and exercisers (other responses, n = 1191). 2.3. Procedure The university's Institutional Review Board for Human Subjects Research granted approval for the study as did the school district's internal research committee and the principals at each school. Parental consent forms were completed during fall registration (just prior to
J. Moy et al. / Eating Behaviors 14 (2013) 529–532
school starting) and sent home to parents again during the school year via physical education (PE) classes. All children for whom we had parental consent provided their own assent. As part of the statemandated FITNESSGRAM (The Cooper Institute, 2007) testing that was being conducted at each school, the children anonymously completed the IESA-R and ChEAT in small groups during their required PE classes. The authors, and other members of the larger research team, supervised the adolescents as they completed the questionnaires. Participation was voluntary and scores from the questionnaires were matched with BMI and other data provided by the school district based on the students school identification number only. Students who completed the questionnaires were entered into a drawing to win one of several $10 cash prizes that were given away at each school.
531
scored lower on Unconditional Permission to Eat (F [1, 1368] = 6.24, p b .05, partial η2 = .005), but higher on Eating for Physical Reasons (F [1, 1368] = 4.97, p b .05, partial η2 = .004); there were no significant differences on Reliance on Hunger and Satiety Cues (F [1, 1443] = 3.32, p = .07, partial η2 = .002 (See Table 1 for adjusted means and standard errors associated with all main effects). In terms of the Exercise by Sex interaction, boys (Madj = 3.62, SE = ±0.04) and girls (Madj = 3.80, SE = ± 0.04) who were exercising to manage their weight and girls who were not exercising for that purpose (Madj = 3.81, SE = + 0.12) did not differ significantly from each other, but all three groups scored higher than boys who were not exercising to manage their weight (Madj = 3.26, SE = ±0.13) in terms of their awareness of and trust in their bodies' signals about being hungry or full.
2.4. Data analysis 4. Discussion Data were found to be missing either at random or completely at random. Missing values, which ranged from 0.6% to 3.0%, were replaced using expectation maximization procedures (Schlomer, Bauman, & Card, 2010), which is an iterative, two-step process based on maximum likelihood procedures. Because body mass index (BMI) has been related to intuitive eating scores (Tylka, 2006), we used it as a covariate in our analyses to control for the potential influence of actual physical size (e.g., overweight, obesity). Multivariate analysis of covariance (MANCOVA) was used to examine the relation among having dieted (i.e., yes vs. no), having exercised (i.e., yes vs. no), and sex (boy or girl) to the IESA-R factor scores. Partial η2 was used as the estimate of the effect size for all analyses (Tabachnick & Fidell, 2013). 3. Results Because data were available from all students in the district who completed the FITNESSGRAM testing, we compared those students whose parents had consented and who completed the questionnaires (“responders”) with those who did not (“nonresponders”). Within the responders, there was a slightly higher percentage of girls (53.2%) than boys (46.8%); more boys (56.2%) than girls (43.8%) were in the nonresponder group (χ2 [N = 4521] = .36.54, p b .0001). Responders had slightly lower BMIs (M = 20.63 kg/m2, SD = ±4.71) than nonresponders (p b .0001, partial η2 = .005; M = 21.43 kg/m2, SD = ±5.38), and had slightly higher levels of cardiorespiratory fitness (i.e., VO2max) than those who did not respond (p b .0001, partial η2 = .007; Responders – M = 46.58 mL/kg/min, SD = ±5.32; Nonresponders – M = 45.63 mL/kg/min, SD = ±5.58). The MANCOVA for dieting, exercise, and sex revealed no significant three- or two-way interactions (p's N .05), with one exception – Exercise by Sex for Reliance on Hunger and Satiety Cues (F [1, 1368] = 3.81, p = .05, partial η2 = .003. The covariate BMI was related significantly to the three IESA-R factors (Wilk's Lambda = .955, F [3, 1366] = 22.246, p b .0001, partial η2 = .045). The main effects for dieting (Wilk's Lambda = .978, F [3, 1366] = 10.27, p b .0001, partial η2 = .022), exercise (Wilk's Lambda = .989, F [3, 1366] = 5.26, p b .001, partial η2 = .011), and sex (Wilk's Lambda = .988, F [3, 1366] = 5.38, p b .0001, partial η2 = .012) also were significant. For dieting, the ANCOVAs were significant on two of the three factors – dieters felt they had less permission to eat what they wanted, when they wanted (F [1, 1368] = 22.00, p b .0001, partial η2 = .016) and were more likely to eat for emotional, as opposed to physical, reasons (F [1, 1368] = 9.78, p b .005, partial η2 = .01). The two diet groups did not differ significantly on the Reliance on Hunger and Satiety Cues factor (F [1, 1368] = 2.85, p = .092, partial η2 = .002. Further, girls were more aware and trusting of their internal physical cues concerning hunger/satiety than boys (F [1, 1368] = 13.85, p b .0001, partial η2 = .011); no significant sex differences were found for Unconditional Permission to Eat or for Eating for Physical Reasons. Students who were exercising to manage their weight
As hypothesized, regardless of participant sex or exercise status, and controlling for BMI, dieting to manage weight was related inversely to the students' intuitive eating (IE) factor scores – specifically Unconditional Permission to Eat and Eating for Physical Reasons. Students who dieted were more constrained in what they allowed themselves to eat and were more likely to eat for emotional (rather than physical) reasons; they did not, however, differ significantly in their awareness and trust of their bodies' signals regarding hunger/satiety compared to those who were not dieting. When individuals “diet” they may (a) ignore internal signals about hunger and being full, (b) restrict when they allow themselves to eat, and (c) only eat certain foods (e.g., have “good” but not “bad” foods). Over time, such cognitive restraint may put individuals in a state of actual caloric deficit, which would increase the likelihood that they react to stressors by eating and using food to soothe themselves (Heatherton & Baumeister, 1991). Young adolescents who are developing their identities, becoming aware of their bodies, and being influenced by external messages (e.g., parental pressures about food) may be particularly vulnerable to the effects of dieting on intuitive eating. Given the connection between dieting and long-term weight gain (Neumark-Sztainer et al., 2012), longitudinal research that examines the potential mediating effects of intuitive eating in this relationship and ascertains when (and how) intuitive eating processes may be disrupted is needed. There were significant effects for exercise across two IE factors – Unconditional Permission to Eat and Eating for Physical Reasons. Regardless of sex and dieting status, students who were exercising to manage their weight reported feeling less free to eat what (and when and how much) they wanted, but were more likely to eat for
Table 1 Adjusted Means and standard errors for IESA-R factors by dieting and exercising. Factor 1 M Dieting to lose weight – Yes 3.52 – No 3.89⁎⁎ Sex – Boys 3.71 – Girls 3.71 Exercising to lose weight – Yes 3.61 – No 3.81⁎
Factor 2
Factor 3
SE
M
SE
M
SE
0.07 0.04
4.06 4.32⁎⁎
0.07 0.04
3.54 3.70
0.08 0.04
0.06 0.05
4.20 4.18
0.06 0.05
3.44 3.80⁎⁎
0.07 0.06
0.02 0.07
4.28⁎ 4.10
0.02 0.08
3.71 3.54
0.03 0.09
Note: Factor 1 – Unconditional Permission to Eat (eating when hungry and eating what food is desired); Factor 2 – Eat for Physical Reasons (using food to satisfy physiological hunger needs rather than as a coping mechanism for emotional distress); Factor 3 – Reliance on Hunger and Satiety Cues (awareness of physicall hunger and satiety cues and trust in these cues to manage eating). For all three factors, scores ranged from 1, low, to 5, high. BMI served as the covariate. ⁎ p b .05. ⁎⁎ p b .005
532
J. Moy et al. / Eating Behaviors 14 (2013) 529–532
physical reasons (i.e., when hungry, as opposed to when emotionally distressed) compared to those who were not exercising to manage weight. These findings suggest that there are some positive associations between exercising for weight loss and intuitive eating. Exercising can help young adolescents become more aware of their bodies, and gives them a mechanism to manage stress and negative mood other than eating. Thus, adolescents who exercise regularly may eat simply because they are hungry, not because they are upset, and may have improved satiety effects from what they consumed (King et al., 2009). Interestingly, students who were exercising indicated being a bit more restrictive about what they allowed themselves to eat. Because we did not collect food logs or other dietary intake data, we cannot tell if this restriction also was associated with healthier eating (e.g., eating more vegetables, fruits). Additional research is needed to examine how exercising, over time, might be related to changes in how early adolescents think about food and eating, how they manage their mood, and what they allow themselves to eat. There was one significant main effect for sex on Reliance on Hunger and Satiety Cues. In general, girls were more aware and trusting than boys, though when boys were exercising they reported levels of awareness of and trust in their internal physical hunger cues that were similar to the girls (regardless of whether or not they were exercising). Only one other study has examined adolescent sex differences for intuitive eating, reporting that girls were more likely than boys to manage their emotions through eating (Dockendorff et al., 2012). In a study of college undergraduates, women scored lower on Eating for Physical Reasons (i.e., they ate more for emotional reasons), but also were less attuned to their bodies' signals about hunger and fullness than their male counterparts (Tylka & Kroon Van Diest, 2013). Thus, girls learn at a young age to associate eating with relief from negative emotions, which may be caused by the pressures they experience from parents regarding what, when, and how much to consume (Galloway et al., 2010), but remain more attuned with their bodies regarding hunger and fullness than boys. Over the course of adolescence, though, as girls diet at higher rates than boys (Neumark-Sztainer et al., 2011) and are subject to increasing societally-based pressures to achieve a thin body ideal, they may experience more disruptions in their adaptive eating processes, such as becoming less aware and trusting of their internal signals about hunger and fullness. More research is needed on potential sex differences, particularly the extent to which they exist, when such differences emerge, how the differences may change over the course of adolescence, and why boys and girls may have different relationships with food and eating. 5. Conclusion Limitations existed in this study. First, all constructs, with the exception of BMI, were measured through self-report, though our approach was consistent with past research (e.g., Neumark-Sztainer et al., 2012). In future studies, researchers might obtain more objective measures of dieting (e.g., food logs) or physical activity (e.g., accelerometers). Second, data were cross-sectional, therefore no determination of temporal sequencing could be determined. However, it would be expected that dieting, and the restrictions and pressures associated with it, would lead to disruptions in healthy eating behaviors over time (Eneli et al., 2008). Third, across all groups, intuitive eating scores were generally high, suggesting that future studies might examine it throughout adolescence to see when further disruptions occur. Fourth, because of the lack of race/ethnicity information available for all the adolescents, we could not examine that variable in our analyses. Future research would benefit by considering the potential influences of race/ ethnicity on intuitive eating. Despite these limitations, the sample was large and diverse and the findings robust in terms of effect size. These
results provide important information of how different approaches to weight loss may have differential effects on young adolescents' intuitive eating – dieting being more disruptive than exercising – and form the basis for conducting future longitudinal and experimental studies to examine how (and when) dieting has its effects. Role of funding sources There was no external funding provided for the collection of the data associated with this study nor for the design, analysis and interpretation of the data, writing the manuscript, or the submission of this paper. Contributors All authors contributed materially in this research project. Drs. Martin and Greenleaf assisted in organizing the study and data collection, implementing the data collection, and reviewing the paper. Dr. Petrie assisted in the study's organization, design, statistical analysis, and writing of the manuscript. Ms. Dockendorff assisted in the data collection and design of the research question, and provided guidance and summaries of the intuitive eating literature. Ms. Moy assisted in the data collection, statistical analysis, and writing of the paper. Conflict of interest All authors declare that they have no conflicts of interest.
References Carper, J. L., Fisher, J. O., & Birch, L. L. (2000). Young girls' emerging dietary restraint and disinhibition are related to parental control in child feeding. Appetite, 35, 121–129. http://dx.doi.org/10.1006/appe.2000.0343. Dockendorff, S. A., Petrie, T. A., Greenleaf, C. A., & Martin, S. (2012). Intuitive eating scale: An examination among early adolescents. Journal of Counseling Psychology, 1–8. http://dx.doi.org/10.1037/a0029962. Eneli, I. U., Crum, P., & Tylka, T. L. (2008). The trust model: A different feeding paradigm for managing childhood obesity. Obesity, 16, 2197–2204. http://dx.doi.org/10.1038/ oby.2008.378. Galloway, A. T., Farrow, C. V., & Martz, D. M. (2010). Retrospective reports of child feeding practices, current eating behaviors, and BMI in college students. Obesity, 18, 1330–1335. http://dx.doi.org/10.1038/oby.2009.393. Garner, D., Olmsted, M., Bohr, Y., & Garfinkel, P. (1982). The Eating Attitudes Test: Psychometric features and clinical correlates. Psychological Medicine, 12, 871–878. Heatherton, T., & Baumeister, R. (1991). Binge eating as escape from self-awareness. Psychological Bulletin, 110, 86–108. King, N., Caudwell, P., Hopkins, M., Stubbs, J., Naslund, E., & Blundell, J. (2009). Dual-process action of exercise on appetite control: Increase in orexigenic drive but improvement in meal-induced satiety. American Journal of Clinical Nutrition, 90, 921–927. Kroon Van Diest, A. M., & Tylka, T. L. (2010). The caregiver eating messages scale: Development and psychometric investigation. Body Image, 7, 317–326. http://dx.doi.org/10.1016/j.bodyim.2010.06.02. Neumark-Sztainer, D., Paxton, S., Hannan, P., Haines, J., & Story, M. (2006). Does body satisfaction matter? Five-year longitudinal association between body satisfaction and healthy behaviors in adolescent females and males. Journal of Adolescent Health, 39, 244–251. http://dx.doi.org/10.1016/j.jadohealth.2005.12.001. Neumark-Sztainer, D., Wall, M., Larson, & Eisenberg, M. (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, 1004–1011. http://dx.doi.org/10.1016/j.jada.2011.04.012. Neumark-Sztainer, D., Wall, M., Story, M., & Standish, A. (2012). Dieting and unhealthy weight control behaviors during adolescence: Associations with 10-year changes in body mass index. Journal of Adolescent Health, 50, 80–86. http://dx.doi.org/ 10.1016/j.jadohealth.2011.05.010. Polivy, J., & Herman, C. P. (1985). Dieting and binging: A causal analysis. American Psychologist, 20, 193–201. Schlomer, G. L., Bauman, S., & Card, N. A. (2010). Best practices for missing data management in counseling psychology. Journal of Counseling Psychology, 57, 1–10. http://dx.doi.org/10.1037/a0018082. Smolak, L., & Levine, M. (1994). Psychometric properties of the Children's Eating Attitudes Test. International Journal of Eating Disorders, 16, 275–282. Tabachnick, B., & Fidell, L. (2013). Using multivariate statistics (6th ed.)Upper Saddle River, NJ: Pearson. The Cooper Institute (2007). FITNESSGRAM/ACTIVITYGRAM: Test Administration Manual. Champaign, IL: Human Kinetics. Tylka, T. (2006). Development and psychometric evaluation of a measure of intuitive eating. Journal of Counseling Psychology, 53, 226–240. http://dx.doi.org/10.1037/ 0022-0167.53.2.226. Tylka, T., & Kroon Van Diest, A. (2013). The Intuitive Eating Scale-2: Item refinement and psychometric evaluation with college women and men. Journal of Counseling Psychology, 60, 137–153. http://dx.doi.org/10.1037/a0030893. Tylka, T., & Wilcox, J. (2006). Are intuitive eating and eating disorder symptomatology opposite poles of the same construct? Journal of Counseling Psychology, 53, 474–485. http://dx.doi.org/10.1037/0022-0167.53.4.474.
Contents lists available at ScienceDirect
Eating Behaviors
Dieting, exercise, and intuitive eating among early adolescents Jordan Moy a, Trent A. Petrie a,⁎, Sally Dockendorff a, Christy Greenleaf b, Scott Martin b a b
Department of Psychology, University of North Texas, Denton, TX 76203, United States Department of Kinesiology, Health Promotion, & Recreation, University of North Texas, Denton, TX 76203, United States
a r t i c l e
i n f o
Article history: Received 15 December 2012 Received in revised form 8 May 2013 Accepted 25 June 2013 Available online 2 July 2013 Keywords: Intuitive eating Dieting Exercising Body mass index Adolescence
a b s t r a c t Dieting to lose weight, with its focus on restriction of caloric intake, may disrupt intuitive eating processes, though other forms of weight loss, such as exercising, which do not emphasize food may not. In a sample of 669 middle school boys and 708 girls, regardless of sex or exercising, dieting was related to feeling less free to eat what was wanted and to eating more to soothe emotions than to satisfy actual physical hunger. Exercising, independent of dieting, was associated with feeling less permission to eat what was wanted, but also eating to satisfy physical hunger as opposed to coping with emotional distress. Overall, girls were more aware and trusting of their bodily hunger and satiety cues than boys, but when boys were exercising, they scored similarly to girls on this dimension. These findings suggest that different weight loss approaches – dieting vs. exercising – have unique relationships to young adolescents' intuitive eating and these associations tend to be stable across sex. Longitudinal studies now are needed to examine how dieting that begins in childhood or early adolescence might have long-term effects on the progression of intuitive eating. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Intuitive eating is based on physiological hunger and satiety cues rather than situational and emotional cues, thus its central premise is that individuals innately are able to stay in tune with their bodies' natural signals about hunger and fullness (Tylka, 2006). From this perspective, individuals are thought to be able to regulate their food intake in a healthy manner, eating what they want, when they want it, beginning when they become hungry, and stopping when they feel full. When individuals eat intuitively, they generally consume more nutritious foods and are able to maintain a weight that is healthy for their height and sex (Eneli, Crum, & Tylka, 2008). Dieting, which may include unhealthy eating practices such as an extreme restriction of overall caloric intake and/or eating only certain types of food (such as those deemed “good” foods by the dieter), may disrupt intuitive eating processes and has been associated with long-term weight gain (Neumark-Sztainer, Wall, Story, & Standish, 2012). When individuals “diet” they generally have to (a) exert cognitive control over their internal signals about hunger and satiety, often overriding their physiology, and (b) limit what, when, and how much they eat. That is, dieters may forbid consumption of certain “bad” (nondiet) foods and allow themselves to eat only those foods that are considered “good” and align with their dietary restrictions (Polivy & Herman, 1985). Over time, such rigid cognitive and behavioral controls may undermine individuals' understanding of their ⁎ Corresponding author at: Department of Psychology, 1155 Union Circle #311280, University of North Texas, Denton, 76203. Tel.: +1 940 565 4718. E-mail address: [email protected] (T.A. Petrie). 1471-0153/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.eatbeh.2013.06.014
bodies' signals, disrupt their ability to know what they want to eat (and the permission they give themselves to eat it), and lead them to eat more in response to emotions than physical hunger, which often encompasses a binge. The controls that are associated with dieting can cause problems for all age groups, but may be particularly difficult for children and early adolescents because of food- and eating-related pressures within the family environment and their relationship to emotional eating (e.g., Galloway, Farrow, & Martz, 2010; Kroon Van Diest & Tylka, 2010). Two types of controlling feeding behaviors have been identified – food restriction and pressure-to-eat (Eneli et al., 2008). Parents and other caregivers may restrict children's eating, ranging from what to how much they are allowed to consume, particularly when it comes to what are considered to be unhealthy (or “bad”) foods. Parents/caregivers also may pressure children to eat what they consider to be healthy (or “good”) foods, such as fruits and vegetables. Both of these controlling approaches to food and eating require constant monitoring on the part of parents/caregivers, and can interfere with children's intuitive awareness of and trust in their physical responses to hunger and fullness (Eneli et al., 2008). In many cases parents/caregivers may be unaware of their controlling behaviors and how their involvement may lead to dysfunctional eating in their children. For example, Carper, Fisher, and Birch (2000) found that (a) only 26% of parents believed they pressured their daughters about feeding, whereas 61% of the girls felt pressured, and (b) 51% of parents reported restricting certain foods from their daughters, though 63% of the girls thought this occurred. Girls who reported that their parents restricted what they could eat were three times more likely to experience high levels of disinhibited emotional eating
530
J. Moy et al. / Eating Behaviors 14 (2013) 529–532
(i.e., lack of restraint in relation to food) than girls who were able to eat with fewer dietary restrictions. Although parents/caregivers can play a positive role in helping children develop healthy eating patterns, many (intentionally or unintentionally) pressure their children about their food choices, often restricting access to food considered to be unhealthy and pressuring them to eat foods that they believe “should” be eaten (e.g., fruits, vegetables, whole grains). Such pressures and behaviors can interfere with adaptive eating processes (Eneli et al., 2008), and ultimately lead to unhealthy beliefs about food and eating that are carried forward and transformed into self-imposed rules about what foods are acceptable, how much one should eat, and whether or not food can serve as an emotional salve. In fact, it is the consistent use of dieting behaviors (e.g., eating very little, skipping meals, diet pills), as opposed to other unhealthy weight control behaviors (e.g., vomiting, laxatives), which are associated with long-term weight gain (Neumark-Sztainer et al., 2012). And, healthy forms of weight management, specifically exercising, would not be expected to lead to disruptions in intuitive eating because increased levels of physical activity are not related directly to food, do not involve pressures about what (or how much) to eat, and actually can increase individuals' satiety responses to meals (King et al., 2009). Thus, in the current study, the relations between dieting (i.e., decreasing caloric intake), exercising (i.e., increasing cardiorespiratory activity levels), and intuitive eating were examined. Specifically, we expected that early adolescents who were dieting to manage their weight would report more disruptions in their intuitive eating than those who were not. Even though the purpose of increasing physical activity levels also is to manage weight, we expected that there would be no differences in level of intuitive eating between those early adolescents who were exercising compared to those who were not. Because participants' sex has been associated with intuitive eating in adolescents (Dockendorff, Petrie, Greenleaf, & Martin, 2012) and girls have been found to diet more frequently than boys (Neumark-Sztainer, Wall, Larson, & Eisenberg, 2011), we considered it in relation to dieting and exercise as well. 2. Method 2.1. Participants Participants were 669 boys (Mage = 12.31 years, SD = ± 0.98) and 708 girls (Mage = 12.09 years, SD = ± 0.92) drawn from six middle schools in a suburban school district located in the Southwestern U.S. Although race/ethnicity data were not provided on 448 students in the sample, of the remaining, 65.3% were White/ Non-Hispanic, 22.6% White/Hispanic, 9.5% Black/NonHispanic, 3.7% Asian American/Pacific Islander, and 0.8% Native American, which was consistent with the overall racial/ethnic make-up of the school district; 44.1% were in sixth grade, 32% in seventh grade, and 23.8% in eighth grade. 2.2. Measures 2.2.1. Demographics During each school's annual physical fitness testing, height and weight were objectively measured by physical education teachers and converted to body mass index (BMI; kg/m2). BMI, as well as the participants' age, grade, and race/ethnicity, were provided by the school district following approval from their internal research committee. All data provided by the school district were coded solely by the students' school identification numbers; neither names nor other identifying information were used. 2.2.2. Intuitive eating The 16-item Intuitive Eating Scale for Adolescents-Revised (IESA-R) is a revision of the 17-item Intuitive Eating Scale for Adolescent
(Dockendorff et al., 2012; Tylka, 2006) and comprised of three factors: Unconditional Permission to Eat (5 items; i.e., eating when hungry and eating what food is desired; “I let myself eat ‘junk food’ when I want to”); Eating for Physical Rather Than Emotional Reasons (5 items, i.e., using food to satisfy physiological hunger drives rather than as a coping mechanism for emotional distress; “I eat when I am feeling sad or stressed, even when I am not physically hungry” – reversed scored item); and Reliance on Cues of Hunger and Satiety (6 items, i.e., awareness of physiological hunger and satiety cues and trusting these cues to manage eating; “I trust my body to let me know how much to eat”). For each item, participants respond using a 5-point scale that ranges from 1, strongly disagree, to 5, strongly agree. Total factor scores are the mean; higher scores indicate more positive intuitive eating. Internal consistency reliabilities (i.e., Cronbach's alphas) for the three factors were: 0.78 (Unconditional Permission), 0.89 (Eating for Physical Reasons), and 0.88 (Reliance on Physiological Cues). Dockendorff et al. provided extensive information about the factors' validity. 2.2.3. Disordered eating The 26-item Children's Eating Attitudes Test (ChEAT; Garner, Olmsted, Bohr, & Garfinkel, 1982; Smolak & Levine, 1994) is the children's version of the EAT and assesses a variety of attitudes and behaviors associated with anorexia nervosa and bulimia nervosa. On items such as “I think a lot about having fat on my body,” the adolescents responded on a 6-point scale, ranging from 0, always, to 6, never. For each item, the most symptomatic response is scored as a 3, the next most symptomatic as a 2, the next most symptomatic as a 1, and the remaining three responses are scored as 0. Total score is the sum of the items, and can range from 0, no symptoms, to 78, high level of symptoms. Smolak and Levine reported internal consistency reliability (Cronbach's alpha) of .87; alpha for the current study was .73. Smolak and Levine provided extensive information about the scale's validity. In the current study, the ChEAT was used screen out adolescents who might have an eating disorder because dieting and intuitive eating have been strongly related to disordered eating as measured by the EAT (Tylka & Wilcox, 2006). Based on a score of 20 or highter on the ChEAT, 4.1% of the boys and 6% of the girls from the initial sample were eliminated from all analyses. 2.2.4. Dieting Consistent with how dieting has been assessed and classified in adolescents (e.g., Neumark-Sztainer et al., 2011; Neumark-Sztainer et al., 2012), participants responded to a single item question, indicating whether they had, during the last three months, dieted (i.e., “Eaten fewer calories [example – eat less food”]) specifically to manage their weight. Response options were never, seldom, sometimes, often, and always, and were dichotomized into nondieters (never or seldom, n = 862) and dieters (other responses, n = 515). 2.2.5. Exercising Consistent with how exercising has been measured and classified in adolescents (e.g., Neumark-Sztainer, Paxton, Hannan, Haines, & Story, 2006), the boys and girls indicated, through a single item question, whether they had, during the last three months, exercised (i.e., “Increased cardiorespiratory physical activity, such as running”) specifically to manage their weight. Response options were never, seldom, sometimes, often, and always, and were dichotomized into nonexercisers (never or seldom, n = 186) and exercisers (other responses, n = 1191). 2.3. Procedure The university's Institutional Review Board for Human Subjects Research granted approval for the study as did the school district's internal research committee and the principals at each school. Parental consent forms were completed during fall registration (just prior to
J. Moy et al. / Eating Behaviors 14 (2013) 529–532
school starting) and sent home to parents again during the school year via physical education (PE) classes. All children for whom we had parental consent provided their own assent. As part of the statemandated FITNESSGRAM (The Cooper Institute, 2007) testing that was being conducted at each school, the children anonymously completed the IESA-R and ChEAT in small groups during their required PE classes. The authors, and other members of the larger research team, supervised the adolescents as they completed the questionnaires. Participation was voluntary and scores from the questionnaires were matched with BMI and other data provided by the school district based on the students school identification number only. Students who completed the questionnaires were entered into a drawing to win one of several $10 cash prizes that were given away at each school.
531
scored lower on Unconditional Permission to Eat (F [1, 1368] = 6.24, p b .05, partial η2 = .005), but higher on Eating for Physical Reasons (F [1, 1368] = 4.97, p b .05, partial η2 = .004); there were no significant differences on Reliance on Hunger and Satiety Cues (F [1, 1443] = 3.32, p = .07, partial η2 = .002 (See Table 1 for adjusted means and standard errors associated with all main effects). In terms of the Exercise by Sex interaction, boys (Madj = 3.62, SE = ±0.04) and girls (Madj = 3.80, SE = ± 0.04) who were exercising to manage their weight and girls who were not exercising for that purpose (Madj = 3.81, SE = + 0.12) did not differ significantly from each other, but all three groups scored higher than boys who were not exercising to manage their weight (Madj = 3.26, SE = ±0.13) in terms of their awareness of and trust in their bodies' signals about being hungry or full.
2.4. Data analysis 4. Discussion Data were found to be missing either at random or completely at random. Missing values, which ranged from 0.6% to 3.0%, were replaced using expectation maximization procedures (Schlomer, Bauman, & Card, 2010), which is an iterative, two-step process based on maximum likelihood procedures. Because body mass index (BMI) has been related to intuitive eating scores (Tylka, 2006), we used it as a covariate in our analyses to control for the potential influence of actual physical size (e.g., overweight, obesity). Multivariate analysis of covariance (MANCOVA) was used to examine the relation among having dieted (i.e., yes vs. no), having exercised (i.e., yes vs. no), and sex (boy or girl) to the IESA-R factor scores. Partial η2 was used as the estimate of the effect size for all analyses (Tabachnick & Fidell, 2013). 3. Results Because data were available from all students in the district who completed the FITNESSGRAM testing, we compared those students whose parents had consented and who completed the questionnaires (“responders”) with those who did not (“nonresponders”). Within the responders, there was a slightly higher percentage of girls (53.2%) than boys (46.8%); more boys (56.2%) than girls (43.8%) were in the nonresponder group (χ2 [N = 4521] = .36.54, p b .0001). Responders had slightly lower BMIs (M = 20.63 kg/m2, SD = ±4.71) than nonresponders (p b .0001, partial η2 = .005; M = 21.43 kg/m2, SD = ±5.38), and had slightly higher levels of cardiorespiratory fitness (i.e., VO2max) than those who did not respond (p b .0001, partial η2 = .007; Responders – M = 46.58 mL/kg/min, SD = ±5.32; Nonresponders – M = 45.63 mL/kg/min, SD = ±5.58). The MANCOVA for dieting, exercise, and sex revealed no significant three- or two-way interactions (p's N .05), with one exception – Exercise by Sex for Reliance on Hunger and Satiety Cues (F [1, 1368] = 3.81, p = .05, partial η2 = .003. The covariate BMI was related significantly to the three IESA-R factors (Wilk's Lambda = .955, F [3, 1366] = 22.246, p b .0001, partial η2 = .045). The main effects for dieting (Wilk's Lambda = .978, F [3, 1366] = 10.27, p b .0001, partial η2 = .022), exercise (Wilk's Lambda = .989, F [3, 1366] = 5.26, p b .001, partial η2 = .011), and sex (Wilk's Lambda = .988, F [3, 1366] = 5.38, p b .0001, partial η2 = .012) also were significant. For dieting, the ANCOVAs were significant on two of the three factors – dieters felt they had less permission to eat what they wanted, when they wanted (F [1, 1368] = 22.00, p b .0001, partial η2 = .016) and were more likely to eat for emotional, as opposed to physical, reasons (F [1, 1368] = 9.78, p b .005, partial η2 = .01). The two diet groups did not differ significantly on the Reliance on Hunger and Satiety Cues factor (F [1, 1368] = 2.85, p = .092, partial η2 = .002. Further, girls were more aware and trusting of their internal physical cues concerning hunger/satiety than boys (F [1, 1368] = 13.85, p b .0001, partial η2 = .011); no significant sex differences were found for Unconditional Permission to Eat or for Eating for Physical Reasons. Students who were exercising to manage their weight
As hypothesized, regardless of participant sex or exercise status, and controlling for BMI, dieting to manage weight was related inversely to the students' intuitive eating (IE) factor scores – specifically Unconditional Permission to Eat and Eating for Physical Reasons. Students who dieted were more constrained in what they allowed themselves to eat and were more likely to eat for emotional (rather than physical) reasons; they did not, however, differ significantly in their awareness and trust of their bodies' signals regarding hunger/satiety compared to those who were not dieting. When individuals “diet” they may (a) ignore internal signals about hunger and being full, (b) restrict when they allow themselves to eat, and (c) only eat certain foods (e.g., have “good” but not “bad” foods). Over time, such cognitive restraint may put individuals in a state of actual caloric deficit, which would increase the likelihood that they react to stressors by eating and using food to soothe themselves (Heatherton & Baumeister, 1991). Young adolescents who are developing their identities, becoming aware of their bodies, and being influenced by external messages (e.g., parental pressures about food) may be particularly vulnerable to the effects of dieting on intuitive eating. Given the connection between dieting and long-term weight gain (Neumark-Sztainer et al., 2012), longitudinal research that examines the potential mediating effects of intuitive eating in this relationship and ascertains when (and how) intuitive eating processes may be disrupted is needed. There were significant effects for exercise across two IE factors – Unconditional Permission to Eat and Eating for Physical Reasons. Regardless of sex and dieting status, students who were exercising to manage their weight reported feeling less free to eat what (and when and how much) they wanted, but were more likely to eat for
Table 1 Adjusted Means and standard errors for IESA-R factors by dieting and exercising. Factor 1 M Dieting to lose weight – Yes 3.52 – No 3.89⁎⁎ Sex – Boys 3.71 – Girls 3.71 Exercising to lose weight – Yes 3.61 – No 3.81⁎
Factor 2
Factor 3
SE
M
SE
M
SE
0.07 0.04
4.06 4.32⁎⁎
0.07 0.04
3.54 3.70
0.08 0.04
0.06 0.05
4.20 4.18
0.06 0.05
3.44 3.80⁎⁎
0.07 0.06
0.02 0.07
4.28⁎ 4.10
0.02 0.08
3.71 3.54
0.03 0.09
Note: Factor 1 – Unconditional Permission to Eat (eating when hungry and eating what food is desired); Factor 2 – Eat for Physical Reasons (using food to satisfy physiological hunger needs rather than as a coping mechanism for emotional distress); Factor 3 – Reliance on Hunger and Satiety Cues (awareness of physicall hunger and satiety cues and trust in these cues to manage eating). For all three factors, scores ranged from 1, low, to 5, high. BMI served as the covariate. ⁎ p b .05. ⁎⁎ p b .005
532
J. Moy et al. / Eating Behaviors 14 (2013) 529–532
physical reasons (i.e., when hungry, as opposed to when emotionally distressed) compared to those who were not exercising to manage weight. These findings suggest that there are some positive associations between exercising for weight loss and intuitive eating. Exercising can help young adolescents become more aware of their bodies, and gives them a mechanism to manage stress and negative mood other than eating. Thus, adolescents who exercise regularly may eat simply because they are hungry, not because they are upset, and may have improved satiety effects from what they consumed (King et al., 2009). Interestingly, students who were exercising indicated being a bit more restrictive about what they allowed themselves to eat. Because we did not collect food logs or other dietary intake data, we cannot tell if this restriction also was associated with healthier eating (e.g., eating more vegetables, fruits). Additional research is needed to examine how exercising, over time, might be related to changes in how early adolescents think about food and eating, how they manage their mood, and what they allow themselves to eat. There was one significant main effect for sex on Reliance on Hunger and Satiety Cues. In general, girls were more aware and trusting than boys, though when boys were exercising they reported levels of awareness of and trust in their internal physical hunger cues that were similar to the girls (regardless of whether or not they were exercising). Only one other study has examined adolescent sex differences for intuitive eating, reporting that girls were more likely than boys to manage their emotions through eating (Dockendorff et al., 2012). In a study of college undergraduates, women scored lower on Eating for Physical Reasons (i.e., they ate more for emotional reasons), but also were less attuned to their bodies' signals about hunger and fullness than their male counterparts (Tylka & Kroon Van Diest, 2013). Thus, girls learn at a young age to associate eating with relief from negative emotions, which may be caused by the pressures they experience from parents regarding what, when, and how much to consume (Galloway et al., 2010), but remain more attuned with their bodies regarding hunger and fullness than boys. Over the course of adolescence, though, as girls diet at higher rates than boys (Neumark-Sztainer et al., 2011) and are subject to increasing societally-based pressures to achieve a thin body ideal, they may experience more disruptions in their adaptive eating processes, such as becoming less aware and trusting of their internal signals about hunger and fullness. More research is needed on potential sex differences, particularly the extent to which they exist, when such differences emerge, how the differences may change over the course of adolescence, and why boys and girls may have different relationships with food and eating. 5. Conclusion Limitations existed in this study. First, all constructs, with the exception of BMI, were measured through self-report, though our approach was consistent with past research (e.g., Neumark-Sztainer et al., 2012). In future studies, researchers might obtain more objective measures of dieting (e.g., food logs) or physical activity (e.g., accelerometers). Second, data were cross-sectional, therefore no determination of temporal sequencing could be determined. However, it would be expected that dieting, and the restrictions and pressures associated with it, would lead to disruptions in healthy eating behaviors over time (Eneli et al., 2008). Third, across all groups, intuitive eating scores were generally high, suggesting that future studies might examine it throughout adolescence to see when further disruptions occur. Fourth, because of the lack of race/ethnicity information available for all the adolescents, we could not examine that variable in our analyses. Future research would benefit by considering the potential influences of race/ ethnicity on intuitive eating. Despite these limitations, the sample was large and diverse and the findings robust in terms of effect size. These
results provide important information of how different approaches to weight loss may have differential effects on young adolescents' intuitive eating – dieting being more disruptive than exercising – and form the basis for conducting future longitudinal and experimental studies to examine how (and when) dieting has its effects. Role of funding sources There was no external funding provided for the collection of the data associated with this study nor for the design, analysis and interpretation of the data, writing the manuscript, or the submission of this paper. Contributors All authors contributed materially in this research project. Drs. Martin and Greenleaf assisted in organizing the study and data collection, implementing the data collection, and reviewing the paper. Dr. Petrie assisted in the study's organization, design, statistical analysis, and writing of the manuscript. Ms. Dockendorff assisted in the data collection and design of the research question, and provided guidance and summaries of the intuitive eating literature. Ms. Moy assisted in the data collection, statistical analysis, and writing of the paper. Conflict of interest All authors declare that they have no conflicts of interest.
References Carper, J. L., Fisher, J. O., & Birch, L. L. (2000). Young girls' emerging dietary restraint and disinhibition are related to parental control in child feeding. Appetite, 35, 121–129. http://dx.doi.org/10.1006/appe.2000.0343. Dockendorff, S. A., Petrie, T. A., Greenleaf, C. A., & Martin, S. (2012). Intuitive eating scale: An examination among early adolescents. Journal of Counseling Psychology, 1–8. http://dx.doi.org/10.1037/a0029962. Eneli, I. U., Crum, P., & Tylka, T. L. (2008). The trust model: A different feeding paradigm for managing childhood obesity. Obesity, 16, 2197–2204. http://dx.doi.org/10.1038/ oby.2008.378. Galloway, A. T., Farrow, C. V., & Martz, D. M. (2010). Retrospective reports of child feeding practices, current eating behaviors, and BMI in college students. Obesity, 18, 1330–1335. http://dx.doi.org/10.1038/oby.2009.393. Garner, D., Olmsted, M., Bohr, Y., & Garfinkel, P. (1982). The Eating Attitudes Test: Psychometric features and clinical correlates. Psychological Medicine, 12, 871–878. Heatherton, T., & Baumeister, R. (1991). Binge eating as escape from self-awareness. Psychological Bulletin, 110, 86–108. King, N., Caudwell, P., Hopkins, M., Stubbs, J., Naslund, E., & Blundell, J. (2009). Dual-process action of exercise on appetite control: Increase in orexigenic drive but improvement in meal-induced satiety. American Journal of Clinical Nutrition, 90, 921–927. Kroon Van Diest, A. M., & Tylka, T. L. (2010). The caregiver eating messages scale: Development and psychometric investigation. Body Image, 7, 317–326. http://dx.doi.org/10.1016/j.bodyim.2010.06.02. Neumark-Sztainer, D., Paxton, S., Hannan, P., Haines, J., & Story, M. (2006). Does body satisfaction matter? Five-year longitudinal association between body satisfaction and healthy behaviors in adolescent females and males. Journal of Adolescent Health, 39, 244–251. http://dx.doi.org/10.1016/j.jadohealth.2005.12.001. Neumark-Sztainer, D., Wall, M., Larson, & Eisenberg, M. (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, 1004–1011. http://dx.doi.org/10.1016/j.jada.2011.04.012. Neumark-Sztainer, D., Wall, M., Story, M., & Standish, A. (2012). Dieting and unhealthy weight control behaviors during adolescence: Associations with 10-year changes in body mass index. Journal of Adolescent Health, 50, 80–86. http://dx.doi.org/ 10.1016/j.jadohealth.2011.05.010. Polivy, J., & Herman, C. P. (1985). Dieting and binging: A causal analysis. American Psychologist, 20, 193–201. Schlomer, G. L., Bauman, S., & Card, N. A. (2010). Best practices for missing data management in counseling psychology. Journal of Counseling Psychology, 57, 1–10. http://dx.doi.org/10.1037/a0018082. Smolak, L., & Levine, M. (1994). Psychometric properties of the Children's Eating Attitudes Test. International Journal of Eating Disorders, 16, 275–282. Tabachnick, B., & Fidell, L. (2013). Using multivariate statistics (6th ed.)Upper Saddle River, NJ: Pearson. The Cooper Institute (2007). FITNESSGRAM/ACTIVITYGRAM: Test Administration Manual. Champaign, IL: Human Kinetics. Tylka, T. (2006). Development and psychometric evaluation of a measure of intuitive eating. Journal of Counseling Psychology, 53, 226–240. http://dx.doi.org/10.1037/ 0022-0167.53.2.226. Tylka, T., & Kroon Van Diest, A. (2013). The Intuitive Eating Scale-2: Item refinement and psychometric evaluation with college women and men. Journal of Counseling Psychology, 60, 137–153. http://dx.doi.org/10.1037/a0030893. Tylka, T., & Wilcox, J. (2006). Are intuitive eating and eating disorder symptomatology opposite poles of the same construct? Journal of Counseling Psychology, 53, 474–485. http://dx.doi.org/10.1037/0022-0167.53.4.474.
