EMPIRICAL ARTICLE

Cross-Cultural Examination of Beliefs About the Causes of Bulimia Nervosa Among Australian and Japanese Females Rachel Dryer, PhD1* Yuri Uesaka, PhD2 Emmanuel Manalo, PhD3 Graham Tyson, PhD1

ABSTRACT Objective: To identify similarities and differences in beliefs about the causes of Bulimia Nervosa (BN) held by Asian (Japanese) women and Western (Australian) women, and hence, to examine the applicability of belief models of eating disorders (ED) across different cultures. Method: Four hundred three Japanese and 256 Australian female university students (aged 17–35 years) completed a questionnaire that gauged beliefs about the causes of BN. Results: Among the Australian women, the four-component structure of perceived causes (dieting and eating practices, family dynamics, socio-cultural pressure, and psychological vulnerability) found in Dryer et al. (2012) was replicated. Among the Japanese women, however, a three-component structure (without the psychological vulnerability component) was obtained. The groups also differed in the causal component

Introduction Eating disorders (ED), such as anorexia nervosa (AN) and bulimia nervosa (BN), have traditionally been considered as conditions that primarily affect individuals from Western-industrialized countries. However, recent cross-cultural research findings have suggested that individuals from non-Western countries are equally susceptible to these conditions.1–3 In countries like Japan, although lower prevalence rates have been reported,4 the incidence of ED has significantly risen over the past Accepted 22 February 2014 *Correspondence to: Rachel Dryer; School of Psychology, Charles Sturt University, Bathurst, NSW 2795, AUSTRALIA. E-mail: [email protected] 1 School of Psychology, Charles Sturt University, Bathurst, Australia 2 Centre for Excellence in School Education, Graduate School of Education, University of Tokyo, Japan 3 Faculty of Science and Engineering, Waseda University, Tokyo, Japan Published online 6 March 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/eat.22269 C 2014 Wiley Periodicals, Inc. V

176

they most strongly endorsed, that being socio-cultural pressure for the Australian women, and dieting and eating practices for the Japanese women. Discussion: The Japanese participants were found to endorse three out of the four Western-based causal explanations for BN, but the relative importance they placed on those explanations differed from that of the Australian participants. Further research is needed, particularly to establish whether Japanese women simply fail to see psychological vulnerability as a viable cause of BN, or there are in fact cultural differences in the extent to which such vulnerability causes BN. C 2014 Wiley Periodicals, Inc. V Keywords: eating disorders; bulimia nervosa; causal beliefs; cultural similarities and differences; Australian and Japanese women (Int J Eat Disord 2015; 48:176–186)

three decades5 and is predicted to soon reach similar levels to those observed in the West.6 Several studies have also reported relatively high levels of body dissatisfaction, drive for thinness, dieting, and disordered eating, in the Japanese population.7–10 Studies comparing Japanese and American women have reported that although Japanese women’s average BMI is lower than that of American women, the Japanese women’s “ideal” BMI is also lower11 and their ideal female body shape is significantly thinner compared to those of the American women.12 Furthermore, it is common for young Japanese women to consider themselves fat despite having a normal BMI.13 Given the increasing prevalence of ED and the high incidence of body-image disturbances in Japan, it would be important to understand how ED are conceptualized in that country. Etiological explanations developed in the West have identified the crucial role of sociocultural influences such as the popular media in promoting the “thinnessideal.” Women’s internalization of this thin-ideal is considered to be a contributor to increased body International Journal of Eating Disorders 48:2 176–186 2015

CROSS-CULTURAL COMPARISONS OF BELIEFS ABOUT CAUSES OF BN

dissatisfaction, drive for thinness, body change behaviors, and disordered eating.14,15 Several researchers have attempted to explain the increased prevalence of ED in Japan in terms of internalization of Western beauty ideals and Western-based etiological theories.16,17 Although these approaches have some explanatory power, they fail to resolve some important differences that have been reported in the presentation and symptomatology of ED in Japan as compared to Western countries. For example, Japanese women diagnosed with ED have been reported to have lower levels of drive for thinness and greater maturity fears compared with their American counterparts.18 Furthermore, many of the intervention programs that have been implemented in Japan have been based on programs developed in Western countries and have been criticized as lacking in appropriate cultural awareness.5 Not surprisingly, these programs have also met with limited success.19,20 Japanese researchers such as Yasuhara et al.6 have proposed that the rise in ED in Japan may partly be due to the rapid westernization and industrialization that has occurred in that country. They have argued that the rapidly changing social system has resulted in a vulnerable family structure and insufficient social support systems. This conceptualisation of ED is echoed in NishizonoMaher’s description of ED as a “culture change syndrome.” She argued that young women in Japan are faced with trying to build a personal identity inside and outside the family context and are caught between the conflicting pressures of traditional rules of the family and an increasingly individualistic society focused on personal freedoms.21 Chisuwa and O’Dea5 have also suggested that the rise in ED in Japan may be partly due to modern Japanese society having a strong ideal for thinness, which has emerged from combining Japanese traditions (that encourage thinness) with the Western thin ideal of beauty. Despite the Japanese government’s efforts at promoting better dietary habits (e.g., “Health Japan 21” campaign), the prevalence of ED and body-image disturbance continue to rise in that country.22 One way to develop an understanding of how ED is conceptualized in Japan is to examine Japanese women’s attitudes and beliefs about the causes of ED. To date, however, no such examination seems to have been undertaken. Understanding such attitudes and beliefs, and whether they are similar to those held by women in the West, could lead to better delivery of preventative education programs and interventions that address sociocultural factors specific to Japan. International Journal of Eating Disorders 48:2 176–186 2015

Previous research on the attitudes/beliefs about ED held by professional and lay groups has identified important differences that may have implications for determining the content of information campaigns and treatment programs. For example, Dryer et al.23 examined beliefs about the causes of BN among professionals in the medical, psychological and allied health fields, the general public, and female university students in Australia. Their findings indicated four underlying dimensions to such beliefs, corresponding to (1) socio-cultural pressure, (2) dieting and eating practices, (3) family dynamics, and (4) psychological vulnerability. Although there was congruence in beliefs among the professional groups, important differences were observed between the professional and lay groups. Compared to the professional groups, female university students endorsed socio-cultural pressure more strongly; and both lay groups placed greater importance on dieting and eating practices. In contrast, the professional groups provided stronger endorsement for family dynamics dimension compared to the two lay groups. The authors argued that the congruence in causal beliefs among the professional groups would contribute to the ease of interdisciplinary collaboration required in the widely recommended multimodal treatment approaches to BN. However, the differences in beliefs between professionals and lay public need to be considered when developing educational campaigns to improve public knowledge of this disorder, early help-seeking behaviors, and treatment compliance. Overview of the Current Study

As previously noted, there is currently a dearth of information available in the research literature about Japanese women’s beliefs and attitudes concerning ED. To address this gap, the present study examined the beliefs held by Japanese women about the causes of BN and compared those beliefs to those held by women in a Western country (i.e., Australia). One important objective was to find out whether the causal explanations for BN that are commonly found in the Western-based literature would be seen by Japanese women as viable causes for this disorder. This project focused on BN, instead of AN, for two reasons. First, BN and other bulimic associated disorders (e.g., binge-eating disorder and purging disorder) are not only more prevalent among young women,24 but also many women with bulimic pathology do not seek treatment.25 Second, the current researchers were interested in using the questionnaire developed by Dryer et al.23 given that this measure already contained causal explanations for BN commonly found 177

DRYER ET AL.

in the ED literature. To ensure that the participants (in particular the Japanese women) had an opportunity to identify other causal explanations not contained in the questionnaire, it was modified to include open-ended questions. Only young adult women were invited to participate in the current project because bulimic pathology is more common among women than men, and among women it is more prevalent in younger age groups.

Method Participants The participants were composed of 256 Australian and 403 Japanese female university students between the ages of 17 and 35 years. Both Australian and Japanese participants were recruited during the course of one academic teaching year. The Australian participants (mean age 5 26.00 years, SD 5 5.26) were recruited from undergraduate students completing their studies either through a regular on-campus study program or through distance education at Charles Sturt University.a The Japanese participants (mean age 5 20.49 years, SD 5 3.23) were recruited from the campuses of three universities in Tokyo. The students were enrolled in undergraduate courses across a number of different disciplines. The Australian participants were found to have higher BMI scores (M 5 24.90, SD 5 4.82) compared to the Japanese participants (M 5 20.30, SD 5 2.38). The Australian participants also indicated greater familiarity with ED: 77% reported personally knowing someone who had suffered from ED, 7.5% reported having been diagnosed with ED, and 14.5% believed they had ED but had not been formally diagnosed. In contrast, only 35% of the Japanese participants reported knowing someone who had suffered from an ED or of having first-hand experience of ED through personal experience (1.6% diagnosed, 2.3% undiagnosed). Questionnaire To examine beliefs about the causes of BN, participants were first provided with a brief description of this condition (see Appendix) to ensure that all participants understood the eating disorder being examined. The description was based on the DSM-IV TR (APA, 2005) definition of BN and was restricted to the essential features of the disorder so as not to bias participants to a particular causal explanation. Participants were then presented with 26 possible risk factors and asked to rate how important each factor was in the development of

this disorder on a 5-point scale (0 5 unimportant, 4 5 extremely important). The questionnaire which was developed by Dryer et al.23 measures beliefs about the causes of BN and has been shown to have a four-factor structure, with each factor having good internal reliability (Cronbach’s a range of 0.73 to 0.86). Participants were also asked to list and rate (using the same 5-point scale) any other causes that they believed were important in BN which were not covered by the previous 26 items. This was performed to ensure that the full range of beliefs about the causes of this disorder was explored especially among the Japanese participants. The final section of this questionnaire requested participants to report their age, height, weight, and first-hand knowledge of eating disorders through either personal experience or knowing a sufferer. Procedure The questionnaire was translated into Japanese by a professional translator (who was independent of the study) then back-translated into English by the second author. This procedure allowed the authors to confirm the accuracy and appropriateness of the original translation. Permission to undertake this study was obtained from the Charles Sturt University Human Research Ethics Committee, and the appropriate teaching staff in and Heads of Departments of the courses/programs from which the participants were recruited. Participants were approached during lectures, tutorials, or residential schools, and were provided with a brief description of the study. They were also informed that participation was voluntary and anonymous. Students interested in participating were given a study information sheet, the survey questionnaire, and a sealable envelope that they could use to return their completed questionnaire at the end of the lecture or tutorial. Those participants who wanted to complete their questionnaire at another time were given a reply-paid envelope. Australian students studying through distance education were alerted about the study through a message posted on their online study sites. These students were provided with the option of completing a paper-and-pencil version or an online version of the questionnaire. Those who opted for the online questionnaire were provided with the study information sheet and the questionnaire through SurveyMonkeyTM. The two versions of the study information sheet and questionnaire were identical except for the mode of delivery. Return of the completed questionnaire was taken as provision of informed consent.

Results a

Approximately 70% of the Australian participants completed an online questionnaire. No significant differences were observed between participants who completed the two versions of the questionnaire.

178

Separate principal component analyses (PCA) using a promax rotation were conducted on the International Journal of Eating Disorders 48:2 176–186 2015

CROSS-CULTURAL COMPARISONS OF BELIEFS ABOUT CAUSES OF BN TABLE 1. The mean rating (and standard deviation) for each component and each item, the rotated component loadings for the Australian participants (the ranking for each item is also shown)

Dieting and eating practices (Cronbach’s a 5 0.83) Being overweight Poor eating patterns Rigid attitude to dieting Physical changes caused by dieting Failure to learn control over eating Inappropriate patterns of dieting Eating in private Increase in body weight Family dynamics (Cronbach’s a 5 0.81) Conflict in the family A rebellion against parents An inability to cope with change Problems in an individual’s relationship with their father A desire to please others Sexual abuse in childhood Socio-cultural pressure (Cronbach’s a 5 0.76) A fear of getting fat A feeling of being desirable only when thin An obsession with self-image A distortion in how individuals see their body A belief that to be thin will bring happiness Society’s emphasis upon physical thinness Societal pressure to be slim A belief that to lose weight is always good Psychological vulnerability (Cronbach’s a 5 0.82) Low self-esteem Poor self-confidence

Mean

SD

2.30 2.04 2.40 2.56 2.29 2.24 2.43 2.04 2.24 1.83 2.19 1.32 2.06 1.40 2.09 2.22 3.03 3.17 3.11 3.32 3.44 2.99 2.92 2.81 2.39 3.34 3.31 3.33

0.81 1.27 1.18 1.05 1.04 1.23 1.09 1.22 1.12 0.77 1.01 1.06 1.08 1.01 1.16 1.06 0.55 0.96 0.92 0.77 0.77 0.91 0.93 0.92 1.01 0.70 0.81 0.72

Australian and Japanese participants’ ratings of the 26 causal explanations. For both groups, the item means were initially inspected for low mean scores (i.e., means of 1.0 or lower). None of the items met this criteria indicating that all the items were seen as realistic causal explanations by both the Australian and Japanese participants. Consequently, all items were included for further analyses. Velicer’s Minimum Average Partial (MAP) Tests26 were used to determine the appropriate number of components to extract in the PCA. Items were regarded as contributing to the interpretation of a component only if their loading was 0.40 on the primary component and 0.25 on any other component. Causal Belief of the Australian Sample

A four-component solution, which accounted for 49.75% of the variance, was extracted from the ratings of the 26 items obtained from the Australian participants. One item concerned with a desire to control all aspects of life was deleted as it did not meet the factor loading criterion (of 0.40). Another item referring to preoccupation with body weight/image had loadings on two of the components, and hence, was not considered in defining the components or in the calculation of scores. Table 1 shows the loadings of the four rotated component solution. This table also contains the Cronbach’s a for each component, the mean rating (and International Journal of Eating Disorders 48:2 176–186 2015

Rank

1

21 12 10 14 15 11 21 15

0.52 0.72 0.58 0.77 0.81 0.77 0.75 0.71

18 24 20 23 19 17 5 6 3 1 7 8 9 13 4 2

2

3

4

0.74 0.73 0.69 0.80 0.53 0.63 0.64 0.73 0.65 0.50 0.66 0.46 0.55 0.50 0.86 0.84

SD), and the rank for each of the 24 items and the four components. All four components of the causal belief questionnaire obtained by Dryer et al.23 were replicated in the current sample of Australian women. These four components were dieting and eating practices (23.4% of the variance), family dynamics (12.29% of the variance), socio-cultural pressure (7.78% of the variance), and psychological vulnerability (6.04% of the variance). All but 1 of the 24 items loaded on the same components as those obtained in the Dryer et al. study. In the current sample, the item referring to a desire to please others was found to load on the family dynamics component, instead of the psychological vulnerability component. As can be seen in Table 1, the internal consistencies of the four components are all of acceptable levels. Scores for each of the four components were calculated by adding the responses for the items that loaded onto a component, and then dividing by the number of items. A one-way repeated analysis of variance (ANOVA) showed that there were significant differences in the overall levels of endorsement for the four components (F (2.72, 650.91) 5 295.70, p < 0.0001, gp2 5 0.55).b As can be seen in Table 1, the most endorsed causal b As the assumption of sphericity was violated, the Huynh-Feldt correction was used to calculate the F-value.

179

DRYER ET AL. TABLE 2. The mean rating (and standard deviation) for each component and each item, the rotated component loadings for the Japanese participants (the ranking for each item is also shown)

Socio-cultural pressure (Cronbach’s a 5 0.82) A feeling of being desirable only when thin An obsession with self-image A distortion in how individuals see their body A belief that to be thin will bring happiness Society’s emphasis upon physical thinness A belief that to lose weight is always good Dieting and eating practices (Cronbach’s a 5 0.79) Poor eating patterns Physical changes caused by dieting Failure to learn control over eating Inappropriate patterns of dieting Eating in private Increase in body weight Family dynamics (Cronbach’s a 5 0.72) Conflict in the family A rebellion against parents An inability to cope with change Problems in an individual’s relationship with their father A desire to please others

Mean

SD

2.57 2.52 2.82 2.72 2.65 2.36 2.31 2.73 2.76 2.58 3.05 2.92 2.29 2.78 1.84 2.27 1.54 2.04 1.50 1.92

0.85 1.22 1.01 1.10 1.11 1.26 1.30 0.79 1.12 1.14 1.07 1.23 1.13 1.04 0.78 1.19 1.02 1.08 1.13 1.23

components among the Australian participants were psychological vulnerability and socio-cultural pressure, followed by dieting and eating practices. The least endorsed component was family dynamics. Pair-wise comparisons (with Bonferroni adjustment to maintain the error rate at 0.05) indicated that there were significant differences in the levels of endorsement for all four components. To examine whether the participants’ levels of experience of ED influenced their endorsement of these four components, participants were placed into one of two groups on the basis of their personal experiences of ED (i.e., being diagnosed with ED and/or knowing someone with a diagnosis of ED). However, the analyses showed no significant differences between these two groupings of participants. Causal Belief of the Japanese Sample

On the basis of the results of MAP test, the PCA was run extracting three components, which accounted for 47.65% of the variance. Inspection of the item loadings resulted in eight items being removed due to similar loadings on two of the components. One item referring to being overweight did not meet the criteria for a factor loading of 0.40 and was therefore deleted. Table 2 shows the loadings of the three rotated component solution and the Cronbach’s a for each component and the mean rating (and SD) for each of the 17 items. As can be seen in Table 2, three components (i.e., socio-cultural pressure, eating and dieting practices, and family dynamics) that were identified in the current sample of Australian participants 180

Rank

1

9 3 6 7 10 11

0.73 0.56 0.69 0.73 0.77 0.73

5 8 1 2 12 4 13 16 14 17 15

2

3

0.73 0.56 0.74 0.68 0.64 0.63 0.60 0.74 0.50 0.84 0.60

were also obtained in the ratings obtained from the Japanese participants. For example, the first component, which accounted for 29.64% of the variance, was composed of six items that were identical to six of the eight items that loaded on the socio-cultural pressure component identified in the Australian sample. Likewise, the items comprising the second component (11.21% of the variance) were consistent with six of the eight items that comprised the eating and dieting practices component obtained in the Australian sample. This similarity was also observed in the third component (6.80% of the variance) in that it was composed of items identical to those in the family dynamics component. However, note that an equivalent component to psychological vulnerability was not obtained in the Japanese sample. The two items that comprised this component in the Australian sample (i.e., “low self-esteem” and “poor self-confidence”) were deleted in the Japanese sample due to high cross-loadings. To examine whether there were significant differences in the endorsement of these three components, a one-way repeated ANOVA was conducted on the mean scores for the three components. This analysis showed significant differences in the overall levels of endorsement for the three components (F (1.91, 737.74) 5 195.48, p < 0.0001, gp2 5 0.34).b As can be seen in Table 2, the most endorsed causal components among the Japanese participants were dieting and eating practices followed by socio-cultural pressure. Pair-wise comparisons (with Bonferroni adjustment to maintain the error rate at 0.05) indicated a significant difference in the endorsement of these two components. International Journal of Eating Disorders 48:2 176–186 2015

CROSS-CULTURAL COMPARISONS OF BELIEFS ABOUT CAUSES OF BN TABLE 3. Adjusted and unadjusted means for the groups on each of the components after controlling for age and familiarity with ED of the participants Group Australian Causal Belief Component Socio-cultural pressure Dieting and eating practices Family dynamics

Japanese

Unadjusted Means

Adjusted Means

Unadjusted Means

Adjusted Means

M

M

M

M

3.03 2.27 1.82

SD 0.55 0.87 0.77

SE a

3.00 2.33a 1.75

0.06 0.06 0.06

2.57 2.75 1.86

SD 0.84 0.78 0.77

SE b

2.59 2.71b 1.90

0.04 0.05 0.04

Significant group differences (p < 0.0001) are indicated by different lowercase letters across the columns.

Endorsement for these two components also differed significantly from the family dynamics component, which was the least endorsed by the Japanese participants. Participants with personal experiences of ED were not found to differ from participants with no such experiences in terms of their endorsement for the three causal components. Comparisons Between the Australian and Japanese Samples

As there was a large degree of consistency across the two groups in the item composition of three of the four causal components (i.e., socio-cultural pressure, eating and dieting practices, and family dynamics), comparisons could be made between the Australian and Japanese participants in their endorsement for these components. The scores used to make these comparisons were based on the item composition of the components identified in the Japanese sample, instead of the Australian sample. This decision was made on the basis of fewer items comprising these three components identified in the Japanese sample. Scores were calculated by adding the responses for the items that loaded onto a component, and then dividing by the number of items. Due to differences between the two groups in their first-hand experience of ED, this information was coded and used as a covariate in the one-way ANCOVAs conducted to examine group differences in endorsement of the causal components. Because Australian participants (M 5 26, SD 5 5.26) were significantly older than the Japanese participants (M 5 20.49, SD 5 3.23), (t (378.08) 5 15.05, p < 0.0001, d 5 1.26); age was also used as a covariate in these analyses. The adjusted mean scores (and standard errors) for the three components, after controlling for the covariates are shown in Table 3. A Bonferroni corrected significance level of 0.02 was used to control the experiment-wise error rate. After adjusting for the covariates, the analyses indicated that there was a significant difference between the two groups in their endorseInternational Journal of Eating Disorders 48:2 176–186 2015

ment of the socio-cultural pressure (F (1, 635) 5 28.22, p < 0.0001, gp2 5 0.04) and the dieting and eating practices (F (1, 628) 5 21.12, p < 0.0001, gp2 5 0.03) components. The Australian participants provided stronger endorsement for the sociocultural pressure component as a cause of BN compared to the Japanese participants. In contrast, the Japanese participants rated the dieting and eating practices component higher than the Australian participants. No group difference was obtained on the family dynamics component. Qualitative Responses From the Japanese and Australian Participants

Only 12.41% of the Japanese participants completed the open-ended section of the questionnaire that requested participants to list and rate other causes that they believed were important in BN. Many of these respondents identified causes that were similar to or covered already by the items in the questionnaire. For example, 3% of the participants identified socio-cultural pressures to be thin or to look cute that can often be found in Japanese popular media. A small proportion (1.2%) identified dieting and eating practices (e.g., snacking too much, family eating practices) and causes related to family conflict (1%) such as relationship problems with the mother. Causes identified by participants that were not included in the questionnaire included stress (3%) and peer-pressure and bullying (3.7%). For example, scrutiny and comments from friends, competition to be thin with friends, maintenance of peer acceptance, and being bullied when someone is fat, were identified as possible causes of BN. A further 1.5% of the participants also identified pressure from boyfriends or partners to attain a certain beauty ideal (e.g., insensitive comments from boyfriends/ partners about the woman’s figure, trying to meet the boyfriend/partner’s ideal figure for a woman). Similar to the Japanese participants, only a small proportion (17%) of the Australian participants responded to the open-ended questions. As with 181

DRYER ET AL.

the Japanese participants, most of the Australian participants who responded to these questions provided more specific examples of items already covered in the questionnaire. For example, 5.1% of participants identified socio-cultural pressures to meet the thin-ideal (e.g., media portrayal of what is beautiful is almost exclusively of thin women, advertisements linking a thin body to success and happiness). Smaller proportions of the Australian participants also identified causes related to dieting and eating practices (2%), conflict in the family (1.6%), and psychological vulnerabilities such as low self-esteem and self-confidence (1.6%). Although stress was identified by the Australian participants as a cause of BN (2.7%), a larger proportion of participants (6.3%) also identified mental health issues such as anxiety and depression as a cause of the disorder. Interestingly a small proportion of respondents (3.5%) also attributed the disorder to personality flaws in (e.g., weak character) or characteristics of (e.g., perfectionism, addictive personality) the sufferer. This is in contrast to the Japanese participants who did not identify mental health issues or personality flaws/traits as possible causes. Similar to the Japanese participants, peerpressure and bullying (6.6%) were also identified by the Australian participants as causes of BN. However, unlike the Japanese participants, none of the Australian respondents identified causes related to pressure from boyfriends/partners to attain a certain standard of beauty. Instead some of the Australian respondents (2.3%) attributed BN to difficulties experienced by the sufferer in forming relationships with peers and partners (e.g., lack of intimacy, loneliness, desire for acceptance).

Discussion The four-component structure of the Dryer et al.23 questionnaire was replicated in the current sample of Australian women. These four components were dieting and eating practices, family dynamics, socio-cultural pressure, and psychological vulnerability. Australian women were found to provide the strongest endorsement for the psychological vulnerability component indicating that these women believe that BN can be caused by low selfesteem and low self-confidence. This is consistent with previous studies that have found that young women in Western communities regard these psychological vulnerabilities as the most likely causes of BN.23,27,28 The current sample of Australian women was also found to strongly endorse the 182

socio-cultural pressure component, which contained causal explanations relating to confirmed risk factors such as internalization of the thin ideal and body/weight concerns.29 This finding is consistent with the results of previous research23 and is not surprising given that the discourse on BN in Western countries is often dominated by sociocultural explanations.30 The third most endorsed group of causal explanations was that contained within the dieting and eating practices component indicating that Australian women are aware of the poor eating and dieting practices/attitudes that are often associated with BN. The least endorsed causal explanations among this group of women were those contained within the family dynamics component, which related to dysfunctional family systems. This is consistent with the results obtained by Dryer et al.23 who found that this component received the lowest level of endorsement among their sample of female university students, and those drawn from the general public. Compared to socio-cultural variables, much less is known about the contribution of familial variables to the development of BN.30 Consequently, causal explanations involving familial variables are less likely to receive coverage in the popular media. The causal belief structure obtained for Japanese women was composed of three components that corresponded to the dieting and eating practices, socio-cultural pressure, and family dynamics components obtained in the Australian sample. The absence of a psychological vulnerability component in the belief structure obtained for the Japanese women suggests that, among this group, there is either uncertainty or lack of agreement about the causal role of low self-esteem and poor selfconfidence in the development of BN. This finding is in direct contrast to the causal beliefs of Australian women who provided strongest endorsement to this causal component. The group difference found in regards to the psychological vulnerability component may be due to differences between the two cultures in the value placed on the concept of self-esteem. In Western societies like Australia, where greater emphasis is placed on individual performance rather than on the maintenance of group cohesiveness, individuals are encouraged to assert themselves and to develop a robust self-esteem.31 High self-esteem is also widely regarded as a protective factor against a range of mental health illnesses, including bulimiaassociated disorders, in that individuals with low self-esteem are regarded to be more vulnerable to developing such disorders.32 In contrast, in more collectivist societies like Japan, the ability to adjust, International Journal of Eating Disorders 48:2 176–186 2015

CROSS-CULTURAL COMPARISONS OF BELIEFS ABOUT CAUSES OF BN

restrain the self, and maintain group harmony is viewed as desirable and encouraged. Within this society, individual abilities, feelings, and views may need to be controlled or downplayed so as maintain group cohesiveness.31 Therefore, the focus is less on the individual developing a robust selfesteem but more on defining a sense of self in terms of interpersonal relationships with significant others. It follows that low self-esteem may likewise not be associated with other negative attributes such as susceptibility to illnesses like BN. A strong negative stigma associated with mental illness in Japan may have also contributed to the psychological vulnerability component being absent in the belief structure of Japanese women. Studies conducted in Japan indicate that perceived stigma associated with mental illness (i.e., the belief that other individuals hold stigmatizing views or attitudes) and concerns about being identified as having a psychiatric disorder are key deterrents to seeking help from professional mental health services.33,34 The Japanese general public have been reported to have the ability to recognize mental health problems however, unlike their Western counterparts; psychiatric labels tend to only be used to describe severe cases of mental disorders.35 Many psychiatric labels are regarded to have negative connotations in Japan, such that even psychiatrists are often reluctant to use these labels in diagnosing their patients.36 Therefore, in a society where there is an aversive perception of mental health problems, an external (e.g., sociocultural) or behavioral (e.g., dieting/eating habits) cause for BN may be more socially and personally acceptable. Of the three causal components, Japanese women provided the strongest endorsement for the dieting and eating practices component. The Japanese women’s level of endorsement for this component was found to be higher than the endorsement observed among the Australian women. There are a number of possible explanations for this finding. Firstly, recent government initiatives into improving the health of its citizens (e.g., “Health Japan 21”) have focused primarily on the provision of basic food/nutrition education and information about healthy lifestyle choices. The goal of Health Japan 21 is not only to reduce the incidence of ED and being underweight but also to reduce the incidence of obesity in the general population.22 However, the dual focus on obesity and ED, combined with the promotion of general healthy eating habits, is likely to encourage conceptualisation of BN as a disorder caused by inappropriate dieting habits. Secondly, in contemInternational Journal of Eating Disorders 48:2 176–186 2015

porary Japanese households, mothers are often given the responsibility of educating and promoting healthy eating practices within the family. It is not uncommon for adolescents to engage in weight monitoring behaviors and share dieting and weight concerns with their mothers.9 Problems experienced by adolescents with weight can often be regarded as the mother’s failure to properly educate her children in appropriate healthy dieting and eating habits. The Japanese women were found not to endorse the socio-cultural pressure component as strongly as the Australian women. One possible explanation for this difference is that socio-cultural explanations of BN may not be as prevalent in Japan as they are in Australia. Despite the active promotion of slimming information and products in the Japanese media, public discourse on BN has only emerged relatively recently. Unlike in many Western countries such as Australia, where there are widespread media campaigns that include a discussion of the various causes and risks associated with ED, there are currently no equivalent campaign or program in Japan. Combined with the recent “Health Japan 21” campaign, information in the public arena about BN is likely to have been dominated by dietary explanations and much less by other causal explanations. Consequently, although many Japanese women may be aware of society’s emphasis on physical thinness, they may not necessarily regard these pressures to be as important as dietary habits in causing BN. Similar to the Australian women, the Japanese women provided the lowest level of endorsement for the family dynamics component. This finding is slightly surprising given that recent sociocultural explanations of ED in Japan have focused on the difficulty many women face in negotiating traditional family and societal expectations with modern societal pressures.21 One might expect such difficulties to manifest as conflict in the family, a rebellion against parents, or an inability to cope with change. However, the questionnaire items concerning these issues were regarded by the Japanese women as relatively unimportant in causing BN. It is possible that the wording of the questionnaire items may have contributed to the low endorsement of these items. The terms “conflict” and “rebellion” may have been interpreted by the respondents as referring to open conflict and overt expressions of hostilities between family members. In a collectivist society like Japan where group cohesion is desirable, public display of emotions such as anger, sadness, and fear can be at odds with the maintenance of group cohesiveness and 183

DRYER ET AL.

are often discouraged.31 Consequently, overt expressions of anger and conflict within the family may be relatively rare. Family discord may more likely manifest in Japanese society with minimal expressions of open conflict. Both the Australian and Japanese women identified a number of additional causal explanations. Both groups identified peer-pressure and bullying as causes of BN which suggest that women in both cultures are under pressure from their peers to be physically thin. Many of these respondents identified the need to diet for peer acceptance and the need to be “just like their friends”. However, the Japanese women also identified the pressure to attain Japanese male’s ideal body shape for women. This is consistent with results obtained by Smith and Joiner12 who found that, when young Japanese women were asked to identify the ideal female body shape for both men and women, the women believed that their male counterparts preferred a female body that was even thinner than their own female body ideal. Both groups of women identified stress as a cause of BN, with some of the Australian respondents also identifying mental health issues such as anxiety and depression. Many of these “causes” are also comorbid or secondary conditions that can be experienced by ED sufferers; and are conditions that are more likely to be emphasized in “illness” stories commonly found in popular media. This suggests that there may be a level of confusion among some of the respondents in differentiating the symptoms/characteristics of BN from the causes/risks associated with this condition. The most notable finding of the current study is that although Japanese women endorsed a number of Western-based causal explanations for BN, they regarded this condition as being caused by inappropriate dieting and eating habits. This result suggests that BN may not be regarded by Japanese women as a mental health condition requiring specialist treatment but rather as a condition arising from lifestyle choice or dieting that has gone wrong. This conceptualization of BN may partly explain why many Japanese women who suffer from this condition fail to present for treatment.3 The current research is primarily an exploratory and descriptive study. Further research is needed to fully understand how Japanese women conceptualise ED and the culture-specific factors that need to be incorporated into treatment programs. For example, the lack of endorsement by the Japanese participants for the notion of low self-esteem and poor self-confidence being causes of BN needs to be better understood. Without examining the 184

possible connection between low self-esteem/confidence and BN among Japanese women, it would be difficult to confidently determine whether the findings of this study are indicative of a perceptual/belief problem among Japanese women (i.e., not realizing the causal link) or a cultural difference in illness causation (i.e., the causal link between low self-esteem/self-confidence and BN may be absent in Japan and possibly other non-Western societies). The latter possibility is important to consider as previous studies have demonstrated cultural differences not only in value placed on psychological constructs/conditions, but also in the connection of those constructs/conditions to behavior and performance. In Japan, for example, there is far greater value placed on people’s development of the ability to self-reflect (and to learn from that self-reflection) than on their development of high self-esteem.37 There is also evidence indicating that, unlike students from Western cultures, Asian students’ levels of self-efficacy appear not to affect their levels of academic achievement and use of learning strategies like critical thinking.38,39 Several limitations of the study should be noted. First, it could be argued that the methodology used in the current study (i.e., use of a preexisting questionnaire) did not allow exploration of all possible beliefs about the causes of BN, especially among the Japanese participants. However, respondents were given the opportunity to express additional beliefs through the open-ended section of the questionnaire. The majority of the Japanese and Australian women chose not to provide additional causal explanations. Furthermore, none of the questionnaire items received low levels of endorsement which suggests that all the items were seen as realistic causal explanations by both the Australian and the Japanese participants. Second, the sample consisted of university female students which may limit generalizability, although it should be noted that the findings obtained in the Australian women are consistent with previous research on members of the general public and professionals.23 Third, the current study did not examine the respondents’ risk for developing BN. It could be argued that personal experiences of BN may influence beliefs/attitudes toward this disorder. Causal beliefs of BN held by participants with personal experiences of ED were not found to differ from participants with no such experiences. However, the majority of participants comprising the former group were those who had known someone with an ED rather than having these conditions themselves. Measuring the respondents eating behaviors and attitudes may International Journal of Eating Disorders 48:2 176–186 2015

CROSS-CULTURAL COMPARISONS OF BELIEFS ABOUT CAUSES OF BN

have been a better way of examining the issue of whether personal experience influences causal beliefs. Finally, although the emphasis has been on the differences between the two groups of women, the effect sizes show that the magnitudes of these differences are not large, and that there are many similarities between the Australian and Japanese women with regards to their causal beliefs. It is important that future research continues to investigate the similarities and differences in causal beliefs between different cultures. Such research is necessary for the development of more robust theoretical explanations and culturally sensitive treatments for ED. References 1. Gordon RA. Eating disorders. In: Malden, MA, editor. Anatomy of a Social Epidemic, 2nd ed. Blackwell Publications, 2000. 2. Lee S, Lee AM. Disordered eating in three communities of China: A comparative study of female high school students in Hong Kong, Shenzhen, and rural Hunan. Int J Eat Disorder 2000;27:317–327. 3. Pike KM, Borovoy A. The rise of eating disorders in Japan: Issues of culture and limitations of the model of ‘Westernization’. Cult Med Psychiat 2004;28: 493–531. 4. Nakamura K, Yamamoto M, Yamazaki O, Kawashima Y, Muto K, Someya T, et al. Prevalence of anorexia nervosa and bulimia nervosa in a geographically defined area in Japan. Int J Eat Disorder 2000;28:173–180. 5. Chisuwa N, O’Dea JA. Body image and eating disorders amongst Japanese adolescents: A review of the literature. Appetite 2010;54:5–15. 6. Yasuhara D, Homan N, Nagai N, Naruo T, Komaki G, Nakao K, et al. A significant nationwide increase in the prevalence of eating disorders in Japan: 1998-year survey. Int Congr Ser 2002;1241:297–301. 7. Kowner R. Japanese body image: Structure and esteem scores in a crosscultural perspective. Int J Psychol 2002;37:149–159. 8. Kowner R. When ideals are too “far off”: Physical self-ideal discrepancy and body dissatisfaction in Japan. Genet Psychol Monogr 2004;130:333–361. 9. Mukai T, Crago M, Shisslak CM. Eating attitudes and weight preoccupation among female high-school students in Japan. J Child Psychol Psyc 1994;35: 677–688. 10. Nishizawa Y, Kida K, Nishizawa K, Hasiba S, Saito K, Mita R. Perception of self-physique and eating behaviour of high school students in Japan. Psychiat Clin Neuros 2003;57:189–196. 11. Mukai T, Kambara A, Sasaki Y. Body dissatisfaction, need for social approval, and eating disturbances among Japanese and American college women. Sex Roles 1998;39:751–763. 12. Smith AR, Joiner TE. Examining body image discrepancies and perceived weight status in adult Japanese women. Eat Behav 2008;9:513–515. 13. Makino M, Hashizume M, Yasushi M, Tsuboi K, Dennerstein L. Factors associated with abnormal eating attitudes among female college students in Japan. Arch Womens Ment Health 2006;9:203–208. 14. Thompson JK, Heinberg LJ, Altabe M, Tantleff-Dunn S. Exacting Beauty. Washington DC: American Psychological Association, 1999. 15. Stice E, Shaw HE. Role of body dissatisfaction in the onset and maintenance of eating pathology: A synthesis of research findings. J Psychosom Res 2002; 53:985–993. 16. Stark-Wroblewski K, Yanico BJ, Lupe S. Acculturation, internalization of Western appearance norms, and eating pathology among Japanese and Chinese international student women. Psychol Women Q 2005;29:38–46. 17. Yamamiya Y, Shroff H, Thompson JK. The tripartite influence model of body image and eating disturbance: A replication with a Japanese sample. Int J Eat Disorder 2008;41:88–91.

International Journal of Eating Disorders 48:2 176–186 2015

18. Pike KM, Mizushima H. The clinical presentation of Japanese women with anorexia nervosa and bulimia nervosa: A study of the eating disorders inventory-2. Int J Eat Disorder 2005;37:26–31. 19. Nagai M, Aoki K, Masuda K, Iwafuji H. The primary prevention program for eating disorders among high school girls. Jpn J School Health 2007;47:436– 451 (in Japanese). 20. Takahata K, Katsuhara C. A trial of prevention programs on eating-disorders for young female students. Clin Educ 2005;12:13–30 (in Japanese). 21. Nishizono-Maher A. Eating disorders in Japan: Finding the right context. Psychiat Clin Neurosci 1998;52:S320–S323. 22. .Ministry of Health Labour and Welfare. Section 3: Measures against lifestyle-related diseases through “Health Japan 21” and promotion of “Shokuiku (food and nutrition education)”. Available at: http:// www.mhlw.go.jp/english/wp/wp-hw2/part2/p2c1s3.pdf. 23. Dryer R, Tyson GA, Kiernan MJ. Bulimia nervosa: Professional and lay people’s beliefs about the causes. Australian Psychol 2013;48:338–344. 24. Hoek HW, van Hoeken D. Review of the prevalence and incidence of eating disorders. Int J Eat Disorder 2003;34:383–396. 25. Stice E, Bulik CM. Eating disorders. In: Beauchaine TP, Hinshaw SP, editors. Child and Adolescent Psychopathology. Hoboken, New Jersey: Wiley, 2008, pp. 643–669. 26. Zwick WR, Velicer WF. Comparison of five rules for determining the number of components to retain. Psychol Bull 1986;99:423–442. 27. Mond JM, Hay PJ, Rodgers B, Owen C. Beliefs of women concerning causes and risk factors for bulimia nervosa. Aust NZ J Psychiat 2004;38:463–469. 28. Mond JM, Hay PJ, Rodgers B, Owen C. Mental health literacy and eating disorders: What do women with bulimic eating disorders think and know about bulimia nervosa and its treatment? J Ment Health 2008;17: 565–575. 29. Jacobi C, Fittig E. Psychosocial risk factors for eating disorders. In: Agras WS, editor. The Oxford Handbook of Eating Disorders. New York: Oxford University Press, 2010, pp. 123–136. 30. Striegel-Moore RH, Bulik CM. Risk factors for eating disorders. Am Psychol 2007;62:181–198. 31. Markus HR, Kitayama S. Culture and the self: Implications for cognition, emotion and motivation. Psychol Rev 1991;2:224–253. 32. Vohs KD, Bardone AM, Joiner TE Jr, Abramson LY. Perfectionism, perceived weight status, and self-esteem interact to predict bulimic symptoms: A model of bulimic symptom development. J Abnorm Psychol 1999;108:695– 700. 33. Miyaji S. Shinriteki enjyo yousei wo yokusei suru youinn ni tuiteno kentou: Nayamino sinkokudo, jikosutxiguma tono kanrenn kara. [A study of restraining factors on the professional-psychological help seeking: Related to the degree of distress and self-stigma]. Japan Women’s Univ J 2010;16: 153–172. 34. Takano A, Yoshitakem K, Ikeda T, Sato S, Sekitani K. Gakusei soudann kikann heno enjyoyousei no purosesu ni kansuru tansakuteki kenkyu [Exploratory study about the behavioral processes involved in seeking help from university mental health professionals]. CAHE J High Educ Tohoku Univ 2007;2:157–164. 35. Jorm AF, Nakane Y, Christensen H, Yoshioka K, Griffiths KM, Wata Y. Public beliefs about treatment and outcome of mental disorders: A comparison of Australia and Japan. BMC Med 2005;3:12. 36. McDonald-Scott P, Machizawa S, Satoh H. Diagnostic disclosure: A tale of two cultures. Psychol Med 1992;22:147–157. 37. Akiba D. Jikotyousei gakusyuu kenkyu ni okeru bunkateki kousatsu [Cultural considerations in self-regulated learning research]. In: Jiko chousei gakusyu kenkyu kai [Society for Self-Regulated Learning] (Ed.), Jiko chousei gakusyu: Riron to jissen no aratana tenkai he [Self-regulated learning: Toward new developments in theory and practice]. Kyoto: Kitaoji Shobo, 2012, pp. 115–134. 38. Eaton MJ, Dembo MH. Differences in the motivational beliefs of Asian American and non-Asian students. J Educ Psychol 1997;89:433–440. 39. Manalo E, Kusumi T, Koyasu M, Michita Y, Tanaka Y. To what extent do culture-related factors influence university students’ critical thinking use? Think Skills Creativity 2013;10:121–132.

185

DRYER ET AL.

Appendix Bulimia Nervosa is an eating disorder that is characterized by repeated episodes of binge eating which is followed by extreme weight control strategies. A

186

person with bulimia nervosa will regularly eat large amounts of food at once and then they may try to get rid of the food by vomiting, using laxatives, diuretics, or enemas. They may also engage in excessive exercising or fasting.

International Journal of Eating Disorders 48:2 176–186 2015

Copyright of International Journal of Eating Disorders is the property of John Wiley & Sons, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.