- Email: [email protected]
Reciprocal prospective associations between disordered eating and other psychological problems in a community sample of Swedish adolescent girls
Eating Behaviors 15 (2014) 159–163
Contents lists available at ScienceDirect
Eating Behaviors
Reciprocal prospective associations between disordered eating and other psychological problems in a community sample of Swedish adolescent girls Njördur Viborg ⁎, Margit Wångby-Lundh 1, Lars-Gunnar Lundh 2 Department of Psychology, Lund University, Box 213, SE-221 00 Lund, Sweden
a r t i c l e
i n f o
Article history: Received 24 July 2013 Received in revised form 26 October 2013 Accepted 20 November 2013 Available online 28 November 2013 Keywords: Eating disorders Disordered eating Adolescents Prospective Risk factors
a b s t r a c t Disordered eating and its associations with psychological difficulties and body satisfaction were prospectively studied in a community sample of 13–15 year old adolescent girls (N = 428). General psychological difficulties (including hyperactivity–inattention) and lower levels of body satisfaction at T1 were found to predict disordered eating at follow-up one year later (T2). Furthermore, reciprocal associations were found between disordered eating and psychological difficulties (but not body dissatisfaction) so that disordered eating at T1 predicted general psychological difficulties (including hyperactivity–inattention) at T2. The results support the notion of a vicious interplay between disordered eating and other subclinical psychological problems, which may represent a potential mechanism for the development of clinically significant eating disorders. It is suggested that it could be important to identify these kinds of bidirectional processes at an early stage, in order to prevent further developments of clinical forms of psychopathology. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Eating-related problems are common among adolescent girls. Studies from the Nordic countries report percentages of eating-related problems in around 15–30% of adolescent girls (Hautala et al., 2008; Kjelsås, Bjørnstrøm, & Götestam, 2004; Thorsteinsdottir & Ulfarsdottir, 2008; Viborg, Wångby-Lundh, Lundh, & Johnsson, 2012; Waaddegaard, Davidsen, & Kjøller, 2009). Similar rates of eating-related problems have also been observed among girls in Spain (Ferreiro, Seoane, & Senra, 2012), Italy (Toselli et al., 2005), Germany (Herpertz-Dahlmann et al., 2008), and in North America (Croll, Neumark-Sztainer, Story, & Ireland, 2002; Haines et al., 2011; Haley, Hedberg, & Leman, 2010). Although these different studies used different terminologies (disordered eating, unhealthy weight loss practices, etc.) and various cut-offs and criteria for their respective classifications of eating-related problems, the results strongly indicate that disordered eating is a major public health issue among adolescent girls. To increase our knowledge about these matters, there is a need for more research on the development of disordered eating in community samples, and its associations over time with other forms of problems. For example, what counts as risk factors for disordered eating during
⁎ Corresponding author. Tel.: +46 46 17 31 72. E-mail addresses: [email protected] (N. Viborg), [email protected] (M. Wångby-Lundh), [email protected] (L.-G. Lundh). 1 Tel.: +46 46 222 87 49. 2 Tel.: +46 46 222 36 47. 1471-0153/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.eatbeh.2013.11.002
early adolescence, and to what extent does disordered eating during this period represent a risk factor for the development of other problems? Although research on the development of disordered eating in this wider sense is limited, research on clinical eating disorders (EDs) with prospective longitudinal designs may also be used to derive hypotheses concerning the development of disordered eating in adolescents. To what extent can we find similar risk factors and developmental processes in the study of subclinical forms of disordered eating as has been found in research on clinical EDs? According to the continuity hypothesis of eating pathology, EDs represent the endpoint of a continuum along which subclinical forms of disordered eating differ only by degree (e.g., Stice, Killen, Hayward, & Taylor, 1998); if this hypothesis is true, we may expect to find similar risk factors for nonclinical forms of disordered eating as for clinical EDs. EDs typically develop during adolescence, especially among young females (Eddy, Herzog, & Zucker, 2011; Golden et al., 2003) and are associated with a number of other mental disorders and health problems, including mood and anxiety disorders, impulse control disorders and substance use disorder (Halmi, 2010; Herpertz-Dahlmann, 2009; Hudson, Hiripi, Pope, & Kessler, 2007). According to Jacobi, Jones, and Beintner (2011), the most potent and best replicated risk factors for Bulimia Nervosa (BN) and to a lesser degree also Anorexia Nervosa (AN) are gender, weight and shape concerns, various forms of negative affect, neuroticism and general psychiatric morbidity. In an 8-year prospective study, Stice, Marti, and Durant (2011) found that the most potent predictor for the onset of EDs in adolescent girls was body dissatisfaction, and that depressive symptoms interacted with body dissatisfaction in the prediction of ED onset. In a longitudinal study with a Spanish community-based
160
N. Viborg et al. / Eating Behaviors 15 (2014) 159–163
sample of young adolescent girls, Ferreiro et al. (2012) similarly found that body dissatisfaction and depressive symptoms, as well as BMI and perfectionism, predicted the development of disordered eating. As pointed out by Slane, Burt, and Klump (2010), much of the research on risk factors for disordered eating has focused on internalizing problems, whereas little research has been done on externalizing problems. There is, however, some evidence that ADHD symptoms represent risk factors for the development of EDs. For example, a prospective study by Biederman et al. (2007) found that adolescent girls with ADHD were 3.5 times more likely to develop an ED and 5.6 times more likely to develop BN throughout the follow-up period compared to a control group. A five-year longitudinal study by Mikami, Hinshaw, Patterson, and Lee (2008) similarly found that children and adolescents (mean age 14.2 years) with the combined subtype of ADHD showed more eating pathology (body dissatisfaction and BN symptoms) at follow-up compared to a control group. Further, Francis (2011) used longitudinal data available for 940 families and reported that externalizing problems at age 3, and all time points thereafter, were consistently positively associated with disordered eating characteristics at age 15. Another relevant question is if there are also prospective associations in the other direction. That is, does disordered eating (or clinical ED) represent a risk factor for the development of other psychological problems (or psychiatric disorders)? According to Herpertz-Dahlmann (2009), some studies suggest that depression precedes EDs, and others that EDs precede depression. Interestingly, reciprocal relations between depressive and bulimic symptoms were found in an 8-year longitudinal study of adolescents, which showed both that depressive symptoms predicted future increases in bulimic symptoms and vice versa (Presnell, Stice, Seidel, & Madeley, 2009). This raises the question to what extent such processes may lead to “vicious cycles” or “vicious spirals”, where disordered eating and depressive symptoms reciprocally reinforce each other, leading into successively aggravated problems. The same question may be raised with regard to externalizing problems. A recent prospective study of a Finnish community sample (Viinamäki, Marttunen, Fröjd, Ruuska, & Kaltiala-Heino, 2013) showed that 15-year old girls with subclinical bulimia were at risk for conduct disorder at the age of 17, whereas the opposite was not the case — that is, conduct disorder at age 15 was not predictive of subclinical bulimia at age 17. In view of other results which suggest that ADHD symptoms (e.g., Biederman et al., 2007; Mikami et al., 2008) and externalizing problems (Francis, 2011) may predict the future development of disordered eating or EDs, however, the possibility of a reciprocal association between disordered eating and externalizing problems deserves more study. As pointed out by Racine, Root, Klump, and Bulik (2011), the identification and reversal of early symptoms of EDs may be important to prevent the development of chronic conditions that are less amenable to intervention. If “vicious cycle” processes (e.g., between disordered eating and other psychological difficulties) can be identified already at a subclinical level, the early detection of symptoms might make it possible to intervene before such processes expand into clinical forms of psychopathology. The purpose of the present study was to prospectively investigate if certain previously described risk factors for EDs and non-clinical forms of disordered eating predict disordered eating in a community sample of Swedish adolescent girls, and also to test for the possibility of reciprocal associations over time between disordered eating and other psychological problems. Based on previous findings, it was expected that body dissatisfaction (Hypothesis 1) and general psychological difficulties (Hypothesis 2) would be predictive of disordered eating over a one-year period. Further, based on the results from Presnell et al. (2009), who described a reciprocal relation between depression and EDs, it was hypothesized (Hypothesis 3) that disordered eating would predict general psychological difficulties over the same one-year period. Support for both Hypotheses 2 and 3 would be evidence for a reciprocal, bidirectional relationship between general psychological difficulties and disordered eating. Finally, with regard to more specific forms of
psychological difficulties, we also expected that emotional symptoms (Hypothesis 4) and hyperactivity–inattention (Hypothesis 5) would predict disordered eating; further, we also explored if these associations were reciprocal. 2. Method 2.1. Participants The participants were a community sample of all female students in two grades of regular school in a Swedish municipality who took part in a two-wave longitudinal study with a one-year interval. This municipality had approximately 40,000 inhabitants and five schools with 504 female students in Grades 7–8 at T1. At T2 they attended Grades 8–9; the test–retest interval for the different schools ranged from 12 months and 7 days to 13 months and 11 days. The municipality was fairly representative of Sweden in terms of demographics, although slightly more rural, and with a slightly lower mean income level and lower educational level than the rest of Sweden (for more detailed information, see Lundh, Wångby-Lundh, & Bjärehed, 2008). There were complete data on the RiBED-8 for 484 girls at T1. At T2, there were complete data for 477 girls. Imputing 0 for each missing value in participants with at most 3 missing values (10 individuals at T1 and 19 individuals at T2), the final number of participants with RiBED-8 data was 494 girls at T1 and 496 girls at T2. In total, there were longitudinal data on RiBED-8 for 445 girls, representing 85% of all female students who were eligible for inclusion at T1. 2.2. Instruments 2.2.1. Risk Behaviour related to Eating disorders (RiBED-8; Waaddegaard, Thoning, & Petersson, 2003) Risk Behaviour related to Eating disorders (RiBED-8; Waaddegaard et al., 2003) is an eight-item instrument where participants are asked to rate statements about eating-related behaviors and attitudes as to how often each statement applies to them, on a scale from 1 to 4 (with the response alternatives “never”, “seldom”, “often”, and “very often”). The RiBED-8 was specifically designed to capture risk behavior for eating disorders and has been found to be successful in predicting clinical eating disorders in a Danish sample (Waaddegaard et al., 2003). The RiBED-8 comprises the following questions: (1)“I diet”, (2)“I have a bad conscience because I eat sweets”, (3)“I throw up to get rid of food that I have eaten”, (4)“I am satisfied with my eating habits”, (5)“I feel I have to control my eating either by maintaining a strict diet or in some other way”, (6)“I'm afraid of not being able to stop eating once I've started”, (7),“I feel that my desire to lose weight has completely taken over”, and (8)“I feel uncomfortable when I eat with others”. The original Danish version of the RiBED-8 showed good psychometric properties for girls but not for boys. Also, the Swedish version of RiBED-8 (validated by Viborg et al., 2012) showed good reliability and validity for girls, but not for boys. 2.2.2. The Strengths and Difficulties Questionnaire — self-report version (SDQ-s; Goodman, 2001) The Strengths and Difficulties Questionnaire — self-report version (SDQ-s; Goodman, 2001) is a widely used screening instrument for psychological problems among children and adolescents, which contains 25 statements. The participants are instructed to respond to each item on the basis of how things have been for them during the last six months. The items are divided into four difficulty scales (hyperactivity–inattention, emotional symptoms, conduct problems, peer problems) and one prosocial scale (with five items each). The items are scored 0 for “not true”, 1 for “somewhat true” and 2 for “certainly true”. The four first-mentioned scales are summed to generate a Total Difficulties score. The SDQ was translated into Swedish by
N. Viborg et al. / Eating Behaviors 15 (2014) 159–163
Smedje, Broman, Hetta., and von Knorring (1999), and the self-report version was empirically validated by Lundh et al. (2008). 2.2.3. Body-Esteem Scale for Adolescents and Adults — Appearance subscale (BEAA; Mendelson, Mendelson, & White, 2001) Body-Esteem Scale for Adolescents and Adults — Appearance subscale (BEAA; Mendelson et al., 2001) is a 10 item measure which assesses participants' general satisfaction with their own appearance. Participants were asked to rate their degree of agreement with different statements (e.g., “I look as good as I'd like”, “My appearance bothers me” [reversed]) in this version on a 4-point Likert scale. The instrument was translated into Swedish and validated by Erling and Hwang (2004). Whereas the original version uses a 5-point Likert scale, we used a 4-point Likert scale (so that the participants could use the same response format as with the RiBED-8). 2.2.4. Body mass index (ISO-BMI) Body mass index (ISO-BMI) was computed to control for overweight. The questionnaire included one question about the participants' length (with ten response alternatives, from “not taller than 150 cm” and“151–155 cm”, up to“186–190 cm” and “taller than 190 cm”), and one about their weight (with eleven response alternatives, from “not more 35 kg” and “36–40 kg”, up to“76–80 kg” and “more than 80 kg”). The midpoints of the weight and height intervals were used to compute the BMI. The ISO BMI describes internationally established cut-offs for child overweight for girls and boys at different ages (Cole, Bellizzi, Flegal, & Dietz, 2000). 2.3. Procedure The Regional Ethics Committee at Lund University approved of the study. A passive consent procedure was used; parents were instructed to contact the responsible researchers or school personnel if they disapproved of their girls' participation. Information about the form and purpose of the study was sent to the parents of all the students considered for participation. Nine participants (or their parents) chose to refrain from participation. Data collection was conducted in a classroom or large assembly hall during ordinary lecture time by research assistants who were either licensed psychologists or senior students on the psychology program. Respondents were informed that the purpose of the research project was to study the situation for young people today in terms of how they feel and how they perceive themselves, their relations, and their general living conditions. They were instructed to answer all questions as best as they could without thinking too much about each answer. Finally, they were reminded that participation was voluntary, that their answers were treated confidentially, that no school personnel would have access to their answers, and that they should not write their names anywhere on the questionnaire. To enable the matching of answers from Time 1 to Time 2 while preserving the anonymity of the participants, each participant was assigned a code number; each questionnaire was then matched by these code numbers. After completion
161
of the questionnaire, it was sealed in an envelope by the student. Before dismissal, students were informed that the administrator would be around a short while afterward to answer any questions that might have risen. This was done to ensure that students would have someone to turn to if they were upset by anything in the questionnaires. Students were also reminded of how to go about doing so if they wanted to contact the school counselor to talk about any kind of personal problems.
3. Results Table 1 shows that disordered eating, body satisfaction and psychological difficulties showed a high degree of stability from T1 to T2, with one-year stability coefficients ranging between r = .64 and r = .68. A strong negative correlation was noted between disordered eating at T1 and body satisfaction at T1, as well as between disordered eating at T2 and body satisfaction at T2. To test the hypotheses, a series of prospective hierarchical regression analyses were conducted. As seen in Table 2, Hypotheses 1 and 2 were supported, as the hierarchical regression analyses showed that body satisfaction and general psychological difficulties at T1 predicted disordered eating at T2, when disordered eating and BMI at T1 were controlled for. Hypothesis 3 was also supported; that is, disordered eating at T1 predicted general psychological difficulties at T2, when psychological difficulties and BMI at T1 were controlled for. In combination with the support for Hypothesis 2, this is evidence for a bidirectional relation between disordered eating and general psychological difficulties. To test whether disordered eating would be predicted by more specific forms of psychological difficulties as measured by the SDQ, or if prosocial behavior as measured by the SDQ could be a protective factor, a hierarchical regression was conducted with disordered eating and BMI at T1 entered in step 1, and the five SDQ subscales entered at step 2. Although the results supported Hypothesis 5 (i.e., hyperactivity–inattention was a unique predictor of disordered eating), there was no support for Hypothesis 4 (i.e., emotional symptoms did not predict disordered eating). Separate hierarchical regressions for emotional symptoms and hyperactivity–inattention confirmed these results. For exploratory purposes, a hierarchical regression was also conducted to test whether disordered eating would show a reciprocal association with hyperactivity–inattention. Using T2 hyperactivity–inattention as the dependent variable, and entering T1 hyperactivity–inattention as the independent variable at step 1 and T1 disordered eating at step 2, the results showed a significant increase in explained variance when disordered eating was included in the model (B = 0.06, SE B = 0.02, β = 0.12, R2Δ = 0.014, F = 10.74, p b .001). Similarly for exploratory purposes, a hierarchical regression was conducted to test whether disordered eating at T1 would predict body satisfaction at T2, while controlling for body satisfaction at T1. There was, however, no such effect (p N .24), thus showing no evidence of a reciprocal association over time between disordered eating and body satisfaction.
Table 1 Descriptive statistics and reliability (Cronbach's alpha) for RiBED-8, SDQ Total Difficulties, and Body Esteem — Appearance (BEAA) at Time 1 and Time 2, and Pearson correlations between these variables. Variable
M (SD)
Possible range
Alpha
1
1. T1 RiBED-8 2. T2 RiBED-8 3. T1 SDQ Total difficulties 4. T2 SDQ Total difficulties 5. T1 BEAA 6. T2 BEAA 7. T1 BMI 8. T1 SDQ prosocial behavior
14.4 (4.5) 14.6 (4.8) 10.3 (5.2) 10.6 (5.1) 27.4 (5.2) 26.6 (7.0) 19.5 (2.8) 8.1 (1.7)
8–32 8–32 0–40 0–40 10–40 10–40
.80 .83 .78 .80 .92 .92 – .67
– .65⁎⁎ .45⁎⁎ .39⁎⁎
⁎ Significant at the 0.05 level. ⁎⁎ Significant at the 0.01 level.
0–10
−.59⁎⁎ −.42⁎⁎ .35⁎⁎ −.14⁎⁎
2 – .37⁎⁎ .44⁎⁎ −.45⁎⁎ −.59⁎⁎ .26⁎⁎ −.13⁎
3
– .68⁎⁎
−.47⁎⁎ −.37⁎⁎ .19⁎⁎ −.24⁎⁎
4
5
6
7
8
– −.42⁎⁎ −.42⁎⁎ .16⁎⁎ −.25⁎⁎
– .67⁎⁎ −.25⁎⁎ .15⁎⁎
– −.17⁎⁎ .18⁎⁎
– −.05
–
162
N. Viborg et al. / Eating Behaviors 15 (2014) 159–163
Table 2 Prospective hierarchical regressions. Variables Predicting disordered eating at T2 from body satisfaction at T1 Step 1 T1 disordered eating Step 2 T1 disordered eating T1 BMI T1 body satisfaction
R2Δ
.42 .01
Predicting disordered eating at T2 from general psychological difficulties at T1 Step 1 T1 disordered eating .43 Step 2 .01 T1 disordered eating T1 BMI T1 general psychological difficulties Predicting general psychological difficulties at T2 from disordered eating at T1 Step 1 T1 general psychological difficulties .47 Step 2 .01 T1 general psychological difficulties T1 disordered eating T1 BMI Predicting disordered eating at T2 from strengths and difficulties at T1 Step 1 .42 T1 disordered eating T1 BMI Step 2 .02 T1 disordered eating T1 BMI T1 hyperactivity–inattention T1 emotional problems T1 peer problems T1 conduct problems T1 prosocial behavior
B
SE B
β
F-step
.71
.04
.65
.63 .05 −.07
.05 .07 .03
.58⁎⁎⁎ .03 −.11⁎
.72
.04
.66
.65 .06 .10
0.04 .07 .04
.60⁎⁎⁎ .04 .11⁎⁎
.68
0.4
.68
.63 .14 .01
.04 .05 .07
.63⁎⁎⁎ .12⁎⁎ .00
.69 .07
.04 .07
.63⁎⁎⁎ .04
.65 .05 .29 .10 .18 −.24 −.13
.05 .07 .10 .09 .12 .15 .11
.59⁎⁎⁎ .03 .13⁎⁎ .05 .06 −.07 −.04
Partial r
Collinearity (VIF)
.52 .04 −.12
1.60 1.13 1.50
.57 .05 .13
1.34 1.13 1.22
.62 .14 .01
1.22 1.35 1.13
.62 .05
1.12 1.12
.56 .04 .15 .05 .07 −.08 −.05
1.35 1.14 1.38 1.35 1.19 1.48 1.20
316.0⁎⁎⁎ 3.3⁎
326.7⁎⁎⁎ 4.2⁎
382.2⁎⁎⁎ 5.01⁎⁎
156.7⁎⁎⁎
3.6⁎
⁎ p b .05. ⁎⁎ p b .01. ⁎⁎⁎ p b .001.
4. Discussion There are several important findings of the present study. First, as expected and in line with previous research, it was found that body satisfaction and general psychological difficulties predicted disordered eating. These results show that nonclinical forms of disordered eating seem to have a similar relation to various forms of psychological problems as has been shown for clinical EDs, in line with the continuity hypothesis. Girls with a low degree of body satisfaction or with general psychological difficulties, or both, were found to be more likely to develop disordered eating over a one-year period. Second, a reciprocal relationship between disordered eating and general psychological difficulties was found as disordered eating predicted general psychological difficulties. This result resembles that of Presnell et al. (2009) who reported a reciprocal relation between bulimic and depressive symptoms, although the present results refer to disordered eating in a wider sense (rather than bulimic symptoms) and to psychological difficulties in a wider sense (rather than depression). We are not aware of previous studies that have shown that disordered eating predicts general psychological difficulties over time, making the results an important contribution to the field. Third, of the four subscales of the SDQ, it was only hyperactivity– inattention that significantly predicted disordered eating. Somewhat surprisingly, the SDQ scale emotional symptoms was not a significant predictor. The results with regard to hyperactivity, however, are in line with those of Slane et al. (2010), who found that hyperactivity was associated with disordered eating in a cross-sectional study. Some prospective studies have also reported similar results with regard to ADHD symptoms, although Mikami et al. (2008) found that impulsivity, rather than hyperactivity and inattention, best predicted adolescent eating pathology. In addition, the present results indicate that disordered
eating predicts hyperactivity–inattention over a one-year period, thus suggesting that there is a reciprocal association over time between disordered eating and hyperactivity–inattention. Previous studies have shown that adolescent girls with both ADHD and disordered eating have increased levels of mood, anxiety and disruptive behavior disorders (see Herpertz-Dahlmann et al., 2008, for a review). Related to this issue of comorbidity are the present findings that disordered eating and general psychological difficulties were moderately correlated at both test occasions, indicating a stable and important association between disordered eating and other psychological difficulties. Beyond these “static” associations, however, the present results add to this picture by suggesting the hypothesis that disordered eating and other psychological difficulties (e.g., hyperactivity– inattention) in young girls may form a dynamic system, where feedback processes lead to the emergence and stabilization of self-generating “vicious cycles” of disordered eating and other psychological difficulties. Although these effects were small, they were statistically significant, and their dynamic nature may make them even more important. The results thereby suggest a possible “risk mechanism” for the development of successively more severe problems. If it is the case that general psychological difficulties tend to increase the risk for disordered eating, and that disordered eating at the same time tends to increase the risk for other psychological difficulties already at a subclinical level, then it is possible that “vicious cycles” of this kind may serve a role in the development from subclinical to clinical problems. It would then be important to identify these kinds of bidirectional processes at an early stage, to enable some form of intervention before such processes expand into clinical forms of psychopathology. On the other hand, it is important to point out that, because the present results do not contain any data on clinical EDs, this remains speculative and calls for further study.
N. Viborg et al. / Eating Behaviors 15 (2014) 159–163
At the same time it must be stressed that these effects are relatively small (typically 1–2% of the variance). From a prevention perspective, the high stability of disordered eating from T1 to T2 (i.e., disordered eating at T1 predicts more than 40% of disordered eating at T2) means that the presence of disordered eating is what needs to be addressed primarily, with general psychopathology and low body esteem as small additional factors. Still, the demonstration of bidirectional associations between general psychopathology and disordered eating, of the kind that was demonstrated in the present study, adds potentially important information about risk factors of increased disordered eating beyond its apparent stability. The strengths of the present study were that it used a large representative sample of adolescent girls and that there were longitudinal data on disordered eating, psychological difficulties and body satisfaction for almost the entire sample. Furthermore, the high response rate increases the generalizability of the results. The main limitation of the present study was that it is entirely based on self-reported data and that it only used two points of measurement with a one-year interval. A further limitation is that the SDQ may not be an optimal instrument for testing the hypothesis of emotional problems as a risk factor for disordered eating, as the emotional symptoms subscale of the SDQ asks primarily about less severe forms of emotional problems (fear, nervousness, worry) and contains only one depression-related item. The absence of a reciprocal association over time between disordered eating and emotional problems may thus possibly be due to the use of an insufficiently sensitive measure. Future research is needed to replicate these findings with more comprehensive measures of depression as well as externalizing problems. Role of funding sources Funding for this study was provided by a grant from the Swedish Council for Working life and Social Research (FAS), Grant nr 2005-0597. FAS had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication. Contributors Lundh and Wångby-Lundh designed the larger project within which this study was conducted, and wrote the protocol for it. Viborg and Lundh conducted literature searches and provided summaries of previous research studies. Viborg formulated the research questions for the present study and conducted the statistical analyses, under supervision by Lundh and Wångby-Lundh. Viborg wrote the first full draft of the manuscript. Lundh and Wångby-Lundh worked on several edits of the paper. All authors contributed to and have approved the final manuscript. Conflict of interest All authors declare that they have no conflicts of interest.
References Biederman, J., Ball, W., Monuteaux, M. C., Surman, C. B., Johnson, J. L., & Zeitlin, S. (2007). Are girls with ADHD at risk for eating disorders? Results from a controlled, five-year prospective study. Journal of Developmental and Behavioral Pediatrics, 28, 302–307. Cole, T. J., Bellizzi, M. C., Flegal, K. M., & Dietz, W. H. (2000). Establishing a standard definition for child overweight and obesity worldwide: International survey. BMJ, 320(6), 1–6. 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, 33, 166–175. Eddy, K. T., Herzog, D. B., & Zucker, N. L. (2011). Diagnosis and classification of eating disorders in adolescence. In D. Le Grange, & J. Lock (Eds.), Eating disorders in children and adolescents, a clinical handbook (pp. 126–136). New York: The Guilford Press. Erling, A., & Hwang, P. (2004). Body-esteem in Swedish 10-year old children. Perceptual and Motor Skills, 99(2), 437–444. Ferreiro, F., Seoane, G., & Senra, C. (2012). Gender-related risk and protective factors for depressive symptoms and disordered eating in adolescence. A four-year longitudinal study. Journal of Youth and Adolescence, 41(5), 607–622. Francis, L. A. (2011). Body mass index and externalizing behaviour as early as 3 predict disordered eating characteristics at 15 years. Appetite, 57 (Abstracts).
163
Golden, N. H., Katzman, D. K., Kreipe, R. E., Steven, S. L., Sawyer, S. M. R., Nicolls, J., et al. (2003). Eating disorders in adolescents: Position paper of the society of adolescent medicine. Journal of Adolescent Health, 33(6), 496–503. Goodman, R. (2001). Psychometric properties of the Strengths and Difficulties Questionnaire. Journal of the American Academy of Child and Adolescent Psychiatry, 40(11), 1337–1345. Haines, J., Ziyadeh, N. J., Franko, D. L., McDonald, J., Mondo, J. M., & Austin, S. B. (2011). Screening high school students for eating disorders: Validity of brief behavioral and attitudinal measures. Journal of School Health, 81(9), 530–535. Haley, C. C., Hedberg, K., & Leman, R. F. (2010). Disordered eating and unhealthy weight loss practices: Which adolescents are at highest risk? Journal of Adolescent Health, 47, 102–105. Halmi, K. A. (2010). Psychological comorbidity of eating disorders. In W. S. Agras (Ed.), Oxford handbook of eating disorders (pp. 292–303). New York: Oxford University Press. Hautala, L., Junnila, J., Helenius, H., Väänänen, A. -M., Liuksila, P. -R., Räihä, H., et al. (2008). Towards understanding gender differences in disordered eating among adolescents. Journal of Clinical Nursing, 17(13), 1803–1813. Herpertz-Dahlmann, B. (2009). Adolescent eating disorders: Definitions, symptomatology, epidemiology, and comorbidity. Child and Adolescent Psychiatric Clinics of North America, 18(1), 31–47. Herpertz-Dahlmann, B., Wille, N., Hölling, H., Vloet, T. D., Ravens-Sieberer, U., & the Bella Study group (2008). Disordered eating behaviour and attitudes, associated psychopathology and health-related quality of life: Results of the BELLA study group. European Child & Adolescent Psychiatry, 17(1), 82–90. Hudson, J. I., Hiripi, E., Pope, H. G., Jr., & Kessler, R. C. (2007). The prevalence and correlates of eating disorders in the national comorbidity survey replication. Biological Psychiatry, 61(3), 348–358. Jacobi, C., Jones, M., & Beintner, I. (2011). Prevention of eating disorders in children and adolescents. In J. Locke (Ed.), Oxford handbook of child and adolescent eating disorders (pp. 163–181). New York: Oxford University Press. Kjelsås, E., Bjørnstrøm, C., & Götestam, K. G. (2004). Prevalence of eating disorders in female and male adolescents (14–15 years). Eating Behaviors, 5, 13–25. Lundh, L. -G., Wångby-Lundh, M., & Bjärehed, J. (2008). Self-reported emotional and behavioural problems in Swedish 14 to 15-year-old adolescents: A study with the self-report version of the Strengths and Difficulties Questionnaire. Scandinavian Journal of Psychology, 49(6), 523–532. Mendelson, B. K., Mendelson, M. J., & White, D. R. (2001). Body Esteem Scale for Adolescents and Adults. Journal of Personality Assessment, 76(1), 90–106. Mikami, A. Y., Hinshaw, S. P., Patterson, K. A., & Lee, J. C. (2008). Eating pathology among adolescent girls with attention-deficit/hyperactivity disorder. Journal of Abnormal Psychology, 117(1), 225–235. Presnell, K., Stice, E., Seidel, A., & Madeley, M. C. (2009). Depression and eating pathology: Prospective reciprocal relations in adolescents. Clinical Psychology and Psychotherapy, 16, 357–365. Racine, S. E., Root, T. L., Klump, K. L., & Bulik, C. M. (2011). Environmental and genetic risk factors for eating disorders: A developmental perspective. In D. Le Grange, & J. Lock (Eds.), Eating disorders in children and adolescents, a clinical handbook (pp. 25–33). New York: The Guilford Press. Slane, J.D., Burt, S. A., & Klump, K. L. (2010). The road less traveled: Associations between externalizing behaviors and eating pathology. International Journal of Eating Disorders, 43(2), 149–160. Smedje, H., Broman, J. -E., Hetta., J., & von Knorring, A. -L (1999). Psychometric properties of a Swedish version of the “Strengths and Difficulties Questionnaire”. European Child and Adolescent Psychiatry, 8(2), 67–70. Stice, E., Killen, J.D., Hayward, C., & Taylor, C. B. (1998). Support for the continuity hypothesis of bulimic pathology. Journal of Consulting and Clinical Psychology, 66(5), 784–790. Stice, E., Marti, C. N., & Durant, S. (2011). Risk factors for onset of eating disorders: Evidence of multiple risk pathways from an 8-year prospective study. Behaviour Research and Therapy, 49(10), 622–627. Thorsteinsdottir, G., & Ulfarsdottir, L. (2008). Eating disorders in college students in Iceland. European Journal of Psychiatry, 22(2), 107–115. Toselli, A. L., Villani, S., Ferro, A.M., Verri, A., Cucurullo, L., & Marinoni, A. (2005). Eating disorders and their correlates in high school adolescents of Northern Italy. Epidemiologia e Psichiatria Sociale, 14(2), 91–99. Viborg, N., Wångby-Lundh, M., Lundh, L. -G., & Johnsson, P. (2012). Eating-related problems among 13–15 year old Swedish adolescents. Frequency and stability over a one-year period. Scandinavian Journal of Psychology, 53(5), 401–406. Viinamäki, A., Marttunen, M., Fröjd, S., Ruuska, J., & Kaltiala-Heino, R. (2013). Subclinical bulimia predicts conduct disorder in middle adolescent girls. European Eating Disorders Review, 21, 38–44. Waaddegaard, M., Davidsen, M., & Kjøller, M. (2009). Obesity and prevalence of risk behaviour for eating disorders among young Danish women. Scandinavian Journal of Public Health, 37(7), 736–743. Waaddegaard, M., Thoning, H., & Petersson, B. (2003). Validation of a screening instrument for identifying risk behaviour related to eating disorders. European Eating Disorders Review, 11(6), 433–455.
Contents lists available at ScienceDirect
Eating Behaviors
Reciprocal prospective associations between disordered eating and other psychological problems in a community sample of Swedish adolescent girls Njördur Viborg ⁎, Margit Wångby-Lundh 1, Lars-Gunnar Lundh 2 Department of Psychology, Lund University, Box 213, SE-221 00 Lund, Sweden
a r t i c l e
i n f o
Article history: Received 24 July 2013 Received in revised form 26 October 2013 Accepted 20 November 2013 Available online 28 November 2013 Keywords: Eating disorders Disordered eating Adolescents Prospective Risk factors
a b s t r a c t Disordered eating and its associations with psychological difficulties and body satisfaction were prospectively studied in a community sample of 13–15 year old adolescent girls (N = 428). General psychological difficulties (including hyperactivity–inattention) and lower levels of body satisfaction at T1 were found to predict disordered eating at follow-up one year later (T2). Furthermore, reciprocal associations were found between disordered eating and psychological difficulties (but not body dissatisfaction) so that disordered eating at T1 predicted general psychological difficulties (including hyperactivity–inattention) at T2. The results support the notion of a vicious interplay between disordered eating and other subclinical psychological problems, which may represent a potential mechanism for the development of clinically significant eating disorders. It is suggested that it could be important to identify these kinds of bidirectional processes at an early stage, in order to prevent further developments of clinical forms of psychopathology. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Eating-related problems are common among adolescent girls. Studies from the Nordic countries report percentages of eating-related problems in around 15–30% of adolescent girls (Hautala et al., 2008; Kjelsås, Bjørnstrøm, & Götestam, 2004; Thorsteinsdottir & Ulfarsdottir, 2008; Viborg, Wångby-Lundh, Lundh, & Johnsson, 2012; Waaddegaard, Davidsen, & Kjøller, 2009). Similar rates of eating-related problems have also been observed among girls in Spain (Ferreiro, Seoane, & Senra, 2012), Italy (Toselli et al., 2005), Germany (Herpertz-Dahlmann et al., 2008), and in North America (Croll, Neumark-Sztainer, Story, & Ireland, 2002; Haines et al., 2011; Haley, Hedberg, & Leman, 2010). Although these different studies used different terminologies (disordered eating, unhealthy weight loss practices, etc.) and various cut-offs and criteria for their respective classifications of eating-related problems, the results strongly indicate that disordered eating is a major public health issue among adolescent girls. To increase our knowledge about these matters, there is a need for more research on the development of disordered eating in community samples, and its associations over time with other forms of problems. For example, what counts as risk factors for disordered eating during
⁎ Corresponding author. Tel.: +46 46 17 31 72. E-mail addresses: [email protected] (N. Viborg), [email protected] (M. Wångby-Lundh), [email protected] (L.-G. Lundh). 1 Tel.: +46 46 222 87 49. 2 Tel.: +46 46 222 36 47. 1471-0153/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.eatbeh.2013.11.002
early adolescence, and to what extent does disordered eating during this period represent a risk factor for the development of other problems? Although research on the development of disordered eating in this wider sense is limited, research on clinical eating disorders (EDs) with prospective longitudinal designs may also be used to derive hypotheses concerning the development of disordered eating in adolescents. To what extent can we find similar risk factors and developmental processes in the study of subclinical forms of disordered eating as has been found in research on clinical EDs? According to the continuity hypothesis of eating pathology, EDs represent the endpoint of a continuum along which subclinical forms of disordered eating differ only by degree (e.g., Stice, Killen, Hayward, & Taylor, 1998); if this hypothesis is true, we may expect to find similar risk factors for nonclinical forms of disordered eating as for clinical EDs. EDs typically develop during adolescence, especially among young females (Eddy, Herzog, & Zucker, 2011; Golden et al., 2003) and are associated with a number of other mental disorders and health problems, including mood and anxiety disorders, impulse control disorders and substance use disorder (Halmi, 2010; Herpertz-Dahlmann, 2009; Hudson, Hiripi, Pope, & Kessler, 2007). According to Jacobi, Jones, and Beintner (2011), the most potent and best replicated risk factors for Bulimia Nervosa (BN) and to a lesser degree also Anorexia Nervosa (AN) are gender, weight and shape concerns, various forms of negative affect, neuroticism and general psychiatric morbidity. In an 8-year prospective study, Stice, Marti, and Durant (2011) found that the most potent predictor for the onset of EDs in adolescent girls was body dissatisfaction, and that depressive symptoms interacted with body dissatisfaction in the prediction of ED onset. In a longitudinal study with a Spanish community-based
160
N. Viborg et al. / Eating Behaviors 15 (2014) 159–163
sample of young adolescent girls, Ferreiro et al. (2012) similarly found that body dissatisfaction and depressive symptoms, as well as BMI and perfectionism, predicted the development of disordered eating. As pointed out by Slane, Burt, and Klump (2010), much of the research on risk factors for disordered eating has focused on internalizing problems, whereas little research has been done on externalizing problems. There is, however, some evidence that ADHD symptoms represent risk factors for the development of EDs. For example, a prospective study by Biederman et al. (2007) found that adolescent girls with ADHD were 3.5 times more likely to develop an ED and 5.6 times more likely to develop BN throughout the follow-up period compared to a control group. A five-year longitudinal study by Mikami, Hinshaw, Patterson, and Lee (2008) similarly found that children and adolescents (mean age 14.2 years) with the combined subtype of ADHD showed more eating pathology (body dissatisfaction and BN symptoms) at follow-up compared to a control group. Further, Francis (2011) used longitudinal data available for 940 families and reported that externalizing problems at age 3, and all time points thereafter, were consistently positively associated with disordered eating characteristics at age 15. Another relevant question is if there are also prospective associations in the other direction. That is, does disordered eating (or clinical ED) represent a risk factor for the development of other psychological problems (or psychiatric disorders)? According to Herpertz-Dahlmann (2009), some studies suggest that depression precedes EDs, and others that EDs precede depression. Interestingly, reciprocal relations between depressive and bulimic symptoms were found in an 8-year longitudinal study of adolescents, which showed both that depressive symptoms predicted future increases in bulimic symptoms and vice versa (Presnell, Stice, Seidel, & Madeley, 2009). This raises the question to what extent such processes may lead to “vicious cycles” or “vicious spirals”, where disordered eating and depressive symptoms reciprocally reinforce each other, leading into successively aggravated problems. The same question may be raised with regard to externalizing problems. A recent prospective study of a Finnish community sample (Viinamäki, Marttunen, Fröjd, Ruuska, & Kaltiala-Heino, 2013) showed that 15-year old girls with subclinical bulimia were at risk for conduct disorder at the age of 17, whereas the opposite was not the case — that is, conduct disorder at age 15 was not predictive of subclinical bulimia at age 17. In view of other results which suggest that ADHD symptoms (e.g., Biederman et al., 2007; Mikami et al., 2008) and externalizing problems (Francis, 2011) may predict the future development of disordered eating or EDs, however, the possibility of a reciprocal association between disordered eating and externalizing problems deserves more study. As pointed out by Racine, Root, Klump, and Bulik (2011), the identification and reversal of early symptoms of EDs may be important to prevent the development of chronic conditions that are less amenable to intervention. If “vicious cycle” processes (e.g., between disordered eating and other psychological difficulties) can be identified already at a subclinical level, the early detection of symptoms might make it possible to intervene before such processes expand into clinical forms of psychopathology. The purpose of the present study was to prospectively investigate if certain previously described risk factors for EDs and non-clinical forms of disordered eating predict disordered eating in a community sample of Swedish adolescent girls, and also to test for the possibility of reciprocal associations over time between disordered eating and other psychological problems. Based on previous findings, it was expected that body dissatisfaction (Hypothesis 1) and general psychological difficulties (Hypothesis 2) would be predictive of disordered eating over a one-year period. Further, based on the results from Presnell et al. (2009), who described a reciprocal relation between depression and EDs, it was hypothesized (Hypothesis 3) that disordered eating would predict general psychological difficulties over the same one-year period. Support for both Hypotheses 2 and 3 would be evidence for a reciprocal, bidirectional relationship between general psychological difficulties and disordered eating. Finally, with regard to more specific forms of
psychological difficulties, we also expected that emotional symptoms (Hypothesis 4) and hyperactivity–inattention (Hypothesis 5) would predict disordered eating; further, we also explored if these associations were reciprocal. 2. Method 2.1. Participants The participants were a community sample of all female students in two grades of regular school in a Swedish municipality who took part in a two-wave longitudinal study with a one-year interval. This municipality had approximately 40,000 inhabitants and five schools with 504 female students in Grades 7–8 at T1. At T2 they attended Grades 8–9; the test–retest interval for the different schools ranged from 12 months and 7 days to 13 months and 11 days. The municipality was fairly representative of Sweden in terms of demographics, although slightly more rural, and with a slightly lower mean income level and lower educational level than the rest of Sweden (for more detailed information, see Lundh, Wångby-Lundh, & Bjärehed, 2008). There were complete data on the RiBED-8 for 484 girls at T1. At T2, there were complete data for 477 girls. Imputing 0 for each missing value in participants with at most 3 missing values (10 individuals at T1 and 19 individuals at T2), the final number of participants with RiBED-8 data was 494 girls at T1 and 496 girls at T2. In total, there were longitudinal data on RiBED-8 for 445 girls, representing 85% of all female students who were eligible for inclusion at T1. 2.2. Instruments 2.2.1. Risk Behaviour related to Eating disorders (RiBED-8; Waaddegaard, Thoning, & Petersson, 2003) Risk Behaviour related to Eating disorders (RiBED-8; Waaddegaard et al., 2003) is an eight-item instrument where participants are asked to rate statements about eating-related behaviors and attitudes as to how often each statement applies to them, on a scale from 1 to 4 (with the response alternatives “never”, “seldom”, “often”, and “very often”). The RiBED-8 was specifically designed to capture risk behavior for eating disorders and has been found to be successful in predicting clinical eating disorders in a Danish sample (Waaddegaard et al., 2003). The RiBED-8 comprises the following questions: (1)“I diet”, (2)“I have a bad conscience because I eat sweets”, (3)“I throw up to get rid of food that I have eaten”, (4)“I am satisfied with my eating habits”, (5)“I feel I have to control my eating either by maintaining a strict diet or in some other way”, (6)“I'm afraid of not being able to stop eating once I've started”, (7),“I feel that my desire to lose weight has completely taken over”, and (8)“I feel uncomfortable when I eat with others”. The original Danish version of the RiBED-8 showed good psychometric properties for girls but not for boys. Also, the Swedish version of RiBED-8 (validated by Viborg et al., 2012) showed good reliability and validity for girls, but not for boys. 2.2.2. The Strengths and Difficulties Questionnaire — self-report version (SDQ-s; Goodman, 2001) The Strengths and Difficulties Questionnaire — self-report version (SDQ-s; Goodman, 2001) is a widely used screening instrument for psychological problems among children and adolescents, which contains 25 statements. The participants are instructed to respond to each item on the basis of how things have been for them during the last six months. The items are divided into four difficulty scales (hyperactivity–inattention, emotional symptoms, conduct problems, peer problems) and one prosocial scale (with five items each). The items are scored 0 for “not true”, 1 for “somewhat true” and 2 for “certainly true”. The four first-mentioned scales are summed to generate a Total Difficulties score. The SDQ was translated into Swedish by
N. Viborg et al. / Eating Behaviors 15 (2014) 159–163
Smedje, Broman, Hetta., and von Knorring (1999), and the self-report version was empirically validated by Lundh et al. (2008). 2.2.3. Body-Esteem Scale for Adolescents and Adults — Appearance subscale (BEAA; Mendelson, Mendelson, & White, 2001) Body-Esteem Scale for Adolescents and Adults — Appearance subscale (BEAA; Mendelson et al., 2001) is a 10 item measure which assesses participants' general satisfaction with their own appearance. Participants were asked to rate their degree of agreement with different statements (e.g., “I look as good as I'd like”, “My appearance bothers me” [reversed]) in this version on a 4-point Likert scale. The instrument was translated into Swedish and validated by Erling and Hwang (2004). Whereas the original version uses a 5-point Likert scale, we used a 4-point Likert scale (so that the participants could use the same response format as with the RiBED-8). 2.2.4. Body mass index (ISO-BMI) Body mass index (ISO-BMI) was computed to control for overweight. The questionnaire included one question about the participants' length (with ten response alternatives, from “not taller than 150 cm” and“151–155 cm”, up to“186–190 cm” and “taller than 190 cm”), and one about their weight (with eleven response alternatives, from “not more 35 kg” and “36–40 kg”, up to“76–80 kg” and “more than 80 kg”). The midpoints of the weight and height intervals were used to compute the BMI. The ISO BMI describes internationally established cut-offs for child overweight for girls and boys at different ages (Cole, Bellizzi, Flegal, & Dietz, 2000). 2.3. Procedure The Regional Ethics Committee at Lund University approved of the study. A passive consent procedure was used; parents were instructed to contact the responsible researchers or school personnel if they disapproved of their girls' participation. Information about the form and purpose of the study was sent to the parents of all the students considered for participation. Nine participants (or their parents) chose to refrain from participation. Data collection was conducted in a classroom or large assembly hall during ordinary lecture time by research assistants who were either licensed psychologists or senior students on the psychology program. Respondents were informed that the purpose of the research project was to study the situation for young people today in terms of how they feel and how they perceive themselves, their relations, and their general living conditions. They were instructed to answer all questions as best as they could without thinking too much about each answer. Finally, they were reminded that participation was voluntary, that their answers were treated confidentially, that no school personnel would have access to their answers, and that they should not write their names anywhere on the questionnaire. To enable the matching of answers from Time 1 to Time 2 while preserving the anonymity of the participants, each participant was assigned a code number; each questionnaire was then matched by these code numbers. After completion
161
of the questionnaire, it was sealed in an envelope by the student. Before dismissal, students were informed that the administrator would be around a short while afterward to answer any questions that might have risen. This was done to ensure that students would have someone to turn to if they were upset by anything in the questionnaires. Students were also reminded of how to go about doing so if they wanted to contact the school counselor to talk about any kind of personal problems.
3. Results Table 1 shows that disordered eating, body satisfaction and psychological difficulties showed a high degree of stability from T1 to T2, with one-year stability coefficients ranging between r = .64 and r = .68. A strong negative correlation was noted between disordered eating at T1 and body satisfaction at T1, as well as between disordered eating at T2 and body satisfaction at T2. To test the hypotheses, a series of prospective hierarchical regression analyses were conducted. As seen in Table 2, Hypotheses 1 and 2 were supported, as the hierarchical regression analyses showed that body satisfaction and general psychological difficulties at T1 predicted disordered eating at T2, when disordered eating and BMI at T1 were controlled for. Hypothesis 3 was also supported; that is, disordered eating at T1 predicted general psychological difficulties at T2, when psychological difficulties and BMI at T1 were controlled for. In combination with the support for Hypothesis 2, this is evidence for a bidirectional relation between disordered eating and general psychological difficulties. To test whether disordered eating would be predicted by more specific forms of psychological difficulties as measured by the SDQ, or if prosocial behavior as measured by the SDQ could be a protective factor, a hierarchical regression was conducted with disordered eating and BMI at T1 entered in step 1, and the five SDQ subscales entered at step 2. Although the results supported Hypothesis 5 (i.e., hyperactivity–inattention was a unique predictor of disordered eating), there was no support for Hypothesis 4 (i.e., emotional symptoms did not predict disordered eating). Separate hierarchical regressions for emotional symptoms and hyperactivity–inattention confirmed these results. For exploratory purposes, a hierarchical regression was also conducted to test whether disordered eating would show a reciprocal association with hyperactivity–inattention. Using T2 hyperactivity–inattention as the dependent variable, and entering T1 hyperactivity–inattention as the independent variable at step 1 and T1 disordered eating at step 2, the results showed a significant increase in explained variance when disordered eating was included in the model (B = 0.06, SE B = 0.02, β = 0.12, R2Δ = 0.014, F = 10.74, p b .001). Similarly for exploratory purposes, a hierarchical regression was conducted to test whether disordered eating at T1 would predict body satisfaction at T2, while controlling for body satisfaction at T1. There was, however, no such effect (p N .24), thus showing no evidence of a reciprocal association over time between disordered eating and body satisfaction.
Table 1 Descriptive statistics and reliability (Cronbach's alpha) for RiBED-8, SDQ Total Difficulties, and Body Esteem — Appearance (BEAA) at Time 1 and Time 2, and Pearson correlations between these variables. Variable
M (SD)
Possible range
Alpha
1
1. T1 RiBED-8 2. T2 RiBED-8 3. T1 SDQ Total difficulties 4. T2 SDQ Total difficulties 5. T1 BEAA 6. T2 BEAA 7. T1 BMI 8. T1 SDQ prosocial behavior
14.4 (4.5) 14.6 (4.8) 10.3 (5.2) 10.6 (5.1) 27.4 (5.2) 26.6 (7.0) 19.5 (2.8) 8.1 (1.7)
8–32 8–32 0–40 0–40 10–40 10–40
.80 .83 .78 .80 .92 .92 – .67
– .65⁎⁎ .45⁎⁎ .39⁎⁎
⁎ Significant at the 0.05 level. ⁎⁎ Significant at the 0.01 level.
0–10
−.59⁎⁎ −.42⁎⁎ .35⁎⁎ −.14⁎⁎
2 – .37⁎⁎ .44⁎⁎ −.45⁎⁎ −.59⁎⁎ .26⁎⁎ −.13⁎
3
– .68⁎⁎
−.47⁎⁎ −.37⁎⁎ .19⁎⁎ −.24⁎⁎
4
5
6
7
8
– −.42⁎⁎ −.42⁎⁎ .16⁎⁎ −.25⁎⁎
– .67⁎⁎ −.25⁎⁎ .15⁎⁎
– −.17⁎⁎ .18⁎⁎
– −.05
–
162
N. Viborg et al. / Eating Behaviors 15 (2014) 159–163
Table 2 Prospective hierarchical regressions. Variables Predicting disordered eating at T2 from body satisfaction at T1 Step 1 T1 disordered eating Step 2 T1 disordered eating T1 BMI T1 body satisfaction
R2Δ
.42 .01
Predicting disordered eating at T2 from general psychological difficulties at T1 Step 1 T1 disordered eating .43 Step 2 .01 T1 disordered eating T1 BMI T1 general psychological difficulties Predicting general psychological difficulties at T2 from disordered eating at T1 Step 1 T1 general psychological difficulties .47 Step 2 .01 T1 general psychological difficulties T1 disordered eating T1 BMI Predicting disordered eating at T2 from strengths and difficulties at T1 Step 1 .42 T1 disordered eating T1 BMI Step 2 .02 T1 disordered eating T1 BMI T1 hyperactivity–inattention T1 emotional problems T1 peer problems T1 conduct problems T1 prosocial behavior
B
SE B
β
F-step
.71
.04
.65
.63 .05 −.07
.05 .07 .03
.58⁎⁎⁎ .03 −.11⁎
.72
.04
.66
.65 .06 .10
0.04 .07 .04
.60⁎⁎⁎ .04 .11⁎⁎
.68
0.4
.68
.63 .14 .01
.04 .05 .07
.63⁎⁎⁎ .12⁎⁎ .00
.69 .07
.04 .07
.63⁎⁎⁎ .04
.65 .05 .29 .10 .18 −.24 −.13
.05 .07 .10 .09 .12 .15 .11
.59⁎⁎⁎ .03 .13⁎⁎ .05 .06 −.07 −.04
Partial r
Collinearity (VIF)
.52 .04 −.12
1.60 1.13 1.50
.57 .05 .13
1.34 1.13 1.22
.62 .14 .01
1.22 1.35 1.13
.62 .05
1.12 1.12
.56 .04 .15 .05 .07 −.08 −.05
1.35 1.14 1.38 1.35 1.19 1.48 1.20
316.0⁎⁎⁎ 3.3⁎
326.7⁎⁎⁎ 4.2⁎
382.2⁎⁎⁎ 5.01⁎⁎
156.7⁎⁎⁎
3.6⁎
⁎ p b .05. ⁎⁎ p b .01. ⁎⁎⁎ p b .001.
4. Discussion There are several important findings of the present study. First, as expected and in line with previous research, it was found that body satisfaction and general psychological difficulties predicted disordered eating. These results show that nonclinical forms of disordered eating seem to have a similar relation to various forms of psychological problems as has been shown for clinical EDs, in line with the continuity hypothesis. Girls with a low degree of body satisfaction or with general psychological difficulties, or both, were found to be more likely to develop disordered eating over a one-year period. Second, a reciprocal relationship between disordered eating and general psychological difficulties was found as disordered eating predicted general psychological difficulties. This result resembles that of Presnell et al. (2009) who reported a reciprocal relation between bulimic and depressive symptoms, although the present results refer to disordered eating in a wider sense (rather than bulimic symptoms) and to psychological difficulties in a wider sense (rather than depression). We are not aware of previous studies that have shown that disordered eating predicts general psychological difficulties over time, making the results an important contribution to the field. Third, of the four subscales of the SDQ, it was only hyperactivity– inattention that significantly predicted disordered eating. Somewhat surprisingly, the SDQ scale emotional symptoms was not a significant predictor. The results with regard to hyperactivity, however, are in line with those of Slane et al. (2010), who found that hyperactivity was associated with disordered eating in a cross-sectional study. Some prospective studies have also reported similar results with regard to ADHD symptoms, although Mikami et al. (2008) found that impulsivity, rather than hyperactivity and inattention, best predicted adolescent eating pathology. In addition, the present results indicate that disordered
eating predicts hyperactivity–inattention over a one-year period, thus suggesting that there is a reciprocal association over time between disordered eating and hyperactivity–inattention. Previous studies have shown that adolescent girls with both ADHD and disordered eating have increased levels of mood, anxiety and disruptive behavior disorders (see Herpertz-Dahlmann et al., 2008, for a review). Related to this issue of comorbidity are the present findings that disordered eating and general psychological difficulties were moderately correlated at both test occasions, indicating a stable and important association between disordered eating and other psychological difficulties. Beyond these “static” associations, however, the present results add to this picture by suggesting the hypothesis that disordered eating and other psychological difficulties (e.g., hyperactivity– inattention) in young girls may form a dynamic system, where feedback processes lead to the emergence and stabilization of self-generating “vicious cycles” of disordered eating and other psychological difficulties. Although these effects were small, they were statistically significant, and their dynamic nature may make them even more important. The results thereby suggest a possible “risk mechanism” for the development of successively more severe problems. If it is the case that general psychological difficulties tend to increase the risk for disordered eating, and that disordered eating at the same time tends to increase the risk for other psychological difficulties already at a subclinical level, then it is possible that “vicious cycles” of this kind may serve a role in the development from subclinical to clinical problems. It would then be important to identify these kinds of bidirectional processes at an early stage, to enable some form of intervention before such processes expand into clinical forms of psychopathology. On the other hand, it is important to point out that, because the present results do not contain any data on clinical EDs, this remains speculative and calls for further study.
N. Viborg et al. / Eating Behaviors 15 (2014) 159–163
At the same time it must be stressed that these effects are relatively small (typically 1–2% of the variance). From a prevention perspective, the high stability of disordered eating from T1 to T2 (i.e., disordered eating at T1 predicts more than 40% of disordered eating at T2) means that the presence of disordered eating is what needs to be addressed primarily, with general psychopathology and low body esteem as small additional factors. Still, the demonstration of bidirectional associations between general psychopathology and disordered eating, of the kind that was demonstrated in the present study, adds potentially important information about risk factors of increased disordered eating beyond its apparent stability. The strengths of the present study were that it used a large representative sample of adolescent girls and that there were longitudinal data on disordered eating, psychological difficulties and body satisfaction for almost the entire sample. Furthermore, the high response rate increases the generalizability of the results. The main limitation of the present study was that it is entirely based on self-reported data and that it only used two points of measurement with a one-year interval. A further limitation is that the SDQ may not be an optimal instrument for testing the hypothesis of emotional problems as a risk factor for disordered eating, as the emotional symptoms subscale of the SDQ asks primarily about less severe forms of emotional problems (fear, nervousness, worry) and contains only one depression-related item. The absence of a reciprocal association over time between disordered eating and emotional problems may thus possibly be due to the use of an insufficiently sensitive measure. Future research is needed to replicate these findings with more comprehensive measures of depression as well as externalizing problems. Role of funding sources Funding for this study was provided by a grant from the Swedish Council for Working life and Social Research (FAS), Grant nr 2005-0597. FAS had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication. Contributors Lundh and Wångby-Lundh designed the larger project within which this study was conducted, and wrote the protocol for it. Viborg and Lundh conducted literature searches and provided summaries of previous research studies. Viborg formulated the research questions for the present study and conducted the statistical analyses, under supervision by Lundh and Wångby-Lundh. Viborg wrote the first full draft of the manuscript. Lundh and Wångby-Lundh worked on several edits of the paper. All authors contributed to and have approved the final manuscript. Conflict of interest All authors declare that they have no conflicts of interest.
References Biederman, J., Ball, W., Monuteaux, M. C., Surman, C. B., Johnson, J. L., & Zeitlin, S. (2007). Are girls with ADHD at risk for eating disorders? Results from a controlled, five-year prospective study. Journal of Developmental and Behavioral Pediatrics, 28, 302–307. Cole, T. J., Bellizzi, M. C., Flegal, K. M., & Dietz, W. H. (2000). Establishing a standard definition for child overweight and obesity worldwide: International survey. BMJ, 320(6), 1–6. 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, 33, 166–175. Eddy, K. T., Herzog, D. B., & Zucker, N. L. (2011). Diagnosis and classification of eating disorders in adolescence. In D. Le Grange, & J. Lock (Eds.), Eating disorders in children and adolescents, a clinical handbook (pp. 126–136). New York: The Guilford Press. Erling, A., & Hwang, P. (2004). Body-esteem in Swedish 10-year old children. Perceptual and Motor Skills, 99(2), 437–444. Ferreiro, F., Seoane, G., & Senra, C. (2012). Gender-related risk and protective factors for depressive symptoms and disordered eating in adolescence. A four-year longitudinal study. Journal of Youth and Adolescence, 41(5), 607–622. Francis, L. A. (2011). Body mass index and externalizing behaviour as early as 3 predict disordered eating characteristics at 15 years. Appetite, 57 (Abstracts).
163
Golden, N. H., Katzman, D. K., Kreipe, R. E., Steven, S. L., Sawyer, S. M. R., Nicolls, J., et al. (2003). Eating disorders in adolescents: Position paper of the society of adolescent medicine. Journal of Adolescent Health, 33(6), 496–503. Goodman, R. (2001). Psychometric properties of the Strengths and Difficulties Questionnaire. Journal of the American Academy of Child and Adolescent Psychiatry, 40(11), 1337–1345. Haines, J., Ziyadeh, N. J., Franko, D. L., McDonald, J., Mondo, J. M., & Austin, S. B. (2011). Screening high school students for eating disorders: Validity of brief behavioral and attitudinal measures. Journal of School Health, 81(9), 530–535. Haley, C. C., Hedberg, K., & Leman, R. F. (2010). Disordered eating and unhealthy weight loss practices: Which adolescents are at highest risk? Journal of Adolescent Health, 47, 102–105. Halmi, K. A. (2010). Psychological comorbidity of eating disorders. In W. S. Agras (Ed.), Oxford handbook of eating disorders (pp. 292–303). New York: Oxford University Press. Hautala, L., Junnila, J., Helenius, H., Väänänen, A. -M., Liuksila, P. -R., Räihä, H., et al. (2008). Towards understanding gender differences in disordered eating among adolescents. Journal of Clinical Nursing, 17(13), 1803–1813. Herpertz-Dahlmann, B. (2009). Adolescent eating disorders: Definitions, symptomatology, epidemiology, and comorbidity. Child and Adolescent Psychiatric Clinics of North America, 18(1), 31–47. Herpertz-Dahlmann, B., Wille, N., Hölling, H., Vloet, T. D., Ravens-Sieberer, U., & the Bella Study group (2008). Disordered eating behaviour and attitudes, associated psychopathology and health-related quality of life: Results of the BELLA study group. European Child & Adolescent Psychiatry, 17(1), 82–90. Hudson, J. I., Hiripi, E., Pope, H. G., Jr., & Kessler, R. C. (2007). The prevalence and correlates of eating disorders in the national comorbidity survey replication. Biological Psychiatry, 61(3), 348–358. Jacobi, C., Jones, M., & Beintner, I. (2011). Prevention of eating disorders in children and adolescents. In J. Locke (Ed.), Oxford handbook of child and adolescent eating disorders (pp. 163–181). New York: Oxford University Press. Kjelsås, E., Bjørnstrøm, C., & Götestam, K. G. (2004). Prevalence of eating disorders in female and male adolescents (14–15 years). Eating Behaviors, 5, 13–25. Lundh, L. -G., Wångby-Lundh, M., & Bjärehed, J. (2008). Self-reported emotional and behavioural problems in Swedish 14 to 15-year-old adolescents: A study with the self-report version of the Strengths and Difficulties Questionnaire. Scandinavian Journal of Psychology, 49(6), 523–532. Mendelson, B. K., Mendelson, M. J., & White, D. R. (2001). Body Esteem Scale for Adolescents and Adults. Journal of Personality Assessment, 76(1), 90–106. Mikami, A. Y., Hinshaw, S. P., Patterson, K. A., & Lee, J. C. (2008). Eating pathology among adolescent girls with attention-deficit/hyperactivity disorder. Journal of Abnormal Psychology, 117(1), 225–235. Presnell, K., Stice, E., Seidel, A., & Madeley, M. C. (2009). Depression and eating pathology: Prospective reciprocal relations in adolescents. Clinical Psychology and Psychotherapy, 16, 357–365. Racine, S. E., Root, T. L., Klump, K. L., & Bulik, C. M. (2011). Environmental and genetic risk factors for eating disorders: A developmental perspective. In D. Le Grange, & J. Lock (Eds.), Eating disorders in children and adolescents, a clinical handbook (pp. 25–33). New York: The Guilford Press. Slane, J.D., Burt, S. A., & Klump, K. L. (2010). The road less traveled: Associations between externalizing behaviors and eating pathology. International Journal of Eating Disorders, 43(2), 149–160. Smedje, H., Broman, J. -E., Hetta., J., & von Knorring, A. -L (1999). Psychometric properties of a Swedish version of the “Strengths and Difficulties Questionnaire”. European Child and Adolescent Psychiatry, 8(2), 67–70. Stice, E., Killen, J.D., Hayward, C., & Taylor, C. B. (1998). Support for the continuity hypothesis of bulimic pathology. Journal of Consulting and Clinical Psychology, 66(5), 784–790. Stice, E., Marti, C. N., & Durant, S. (2011). Risk factors for onset of eating disorders: Evidence of multiple risk pathways from an 8-year prospective study. Behaviour Research and Therapy, 49(10), 622–627. Thorsteinsdottir, G., & Ulfarsdottir, L. (2008). Eating disorders in college students in Iceland. European Journal of Psychiatry, 22(2), 107–115. Toselli, A. L., Villani, S., Ferro, A.M., Verri, A., Cucurullo, L., & Marinoni, A. (2005). Eating disorders and their correlates in high school adolescents of Northern Italy. Epidemiologia e Psichiatria Sociale, 14(2), 91–99. Viborg, N., Wångby-Lundh, M., Lundh, L. -G., & Johnsson, P. (2012). Eating-related problems among 13–15 year old Swedish adolescents. Frequency and stability over a one-year period. Scandinavian Journal of Psychology, 53(5), 401–406. Viinamäki, A., Marttunen, M., Fröjd, S., Ruuska, J., & Kaltiala-Heino, R. (2013). Subclinical bulimia predicts conduct disorder in middle adolescent girls. European Eating Disorders Review, 21, 38–44. Waaddegaard, M., Davidsen, M., & Kjøller, M. (2009). Obesity and prevalence of risk behaviour for eating disorders among young Danish women. Scandinavian Journal of Public Health, 37(7), 736–743. Waaddegaard, M., Thoning, H., & Petersson, B. (2003). Validation of a screening instrument for identifying risk behaviour related to eating disorders. European Eating Disorders Review, 11(6), 433–455.