The relationship between spiritual, religious and personal beliefs and disordered eating psychopathology

ABSTRACT

Links between religiosity, spirituality and disordered eating have been posited theoretically and empirically, though most studies have accessed predominantly Christian or Jewish samples from private educational institutions, using surveys which contain a heavy Judeo-Christian bias. The aim of the current study is to explore the relationship between disordered eating psychopathology (DEP) and spiritual, religious and personal beliefs (SRPBs) in a diverse sample of students with a wide range of cultural, religious and spiritual affiliations. Using a cross-sectional design, female students (n = 687) across two universities in Sydney, Australia completed the Eating Disorder Inventory-3 and the SRPB portion of the World Health Organisation Quality of Life – Spiritual, Religious and Personal Beliefs bref (WHOQOL-SRPB bref) questionnaire. While both existential and religious beliefs were significantly correlated with lower levels of DEP, multivariate analysis found that existential beliefs alone predicted DEP (p < 0.001). These results reveal that spiritual and personal beliefs may have a greater role in predicting lower levels of DEP than religious beliefs in a secular university setting. Furthermore, the findings may have important clinical implications including the exploration of existential and religious mechanisms within current treatment models.

KEYWORDS:

• Disordered eating
• anorexia
• bulimia
• body dissatisfaction
• spirituality
• religiosity
• existential
• quality of life

1. Introduction

As with many industrialised countries, the number of Australians with an eating disorder has been increasing. Current estimates suggest that 9% of the total population have been affected by an eating disorder at some point in their life (Hay, Mond, Buttner, & Darby, 2008; Weltzin et al., 2005). Data from American universities reveal that 17% of female students may suffer from an eating disorder (Prouty, Protinsky, & Canady, 2002), with 61% having subclinical disordered eating problems (Mintz & Bets, 1988). While possible causal factors are numerous, a combination of social factors that include negative pressure from peers and the media (Polivy & Herman, 2002), and anxiety derived from academic and social transition (Barker & Galambos, 2007) may contribute to these elevated numbers in these populations. Within rich multicultural societies, women may be further at risk of developing an eating disorder due to the clash between cultural and western ideals of body image (Bhugra & Jones, 2001; Di Nicola, 1990; Furnham & Alibhai, 1983).

Disordered eating psychopathology (DEP) describes psychological constructs relevant to the development and maintenance of eating disorders (Garner, 2004). Three symptom clusters commonly used to measure DEP include drive for thinness (DT), bulimia (B) and body dissatisfaction (BD) (Garner, 2004; Gleaves, Pearson, Ambwani, & Morey, 2014). Those who experience aspects of DEP are at a greater risk of developing or having a clinical eating disorder (Fredrickson & Roberts, 1997; Kotler, Cohen, Davies, Pine, & Walsh, 2001; Treasure, Claudino, & Zucker, 2010). With serious physical and mental health consequences, and elevated mortality rates (Arcelus, Mitchell, Wales, & Nielsen, 2011), eating disorders remain difficult to treat and novel avenues need to be explored, including a focus on risk factors and ideally strategies in prevention.

From early Christianity to recent times, many historians have speculated as to the possible links between eating disorder symptoms and religiosity and spirituality (Bell, 1985; Bemporad, 1996; Bynum, 1978). These links gained prominence in the sixteenth and seventeenth centuries, following an increased focus on fasting and prayer as the means of achieving spiritual communion, and the Puritan/Calvinist association of menstruation with the doctrine of original sin. This provided young women with further religious justification for fasting to the point of amenorrhoea (Huline-Dickens, 2000; Wood, 1981). These same forces are linked with influential social learning and psychobiological models of anorexia nervosa (Bruch, 1978; Crisp, 1995). For example, in such theories mechanisms of asceticism and dualism are employed in anorexia nervosa in attempts to regain control over self (Surgenor, Horn, Plumridge, & Hudson, 2002), with bingeing arising from the breakdown of such control. These theories would suggest that at least some mechanisms or concepts within religiosity may overdevelop with eating disorder phenomena, even if the extent of religious/spiritual affiliation in such clinical groups may be comparable to the general population (Joughin, Crisp, Halek, & Humphrey, 1992).

Over the past few years increasing empirical attention has emerged exploring these links, though again mainly in respect to anorexia nervosa symptoms. In summarising this literature, Rider, Terrell, Sisemore, and Hecht (2014) described and provided additional empirical support to two themes: religion as a cultivator and preserver of eating disorder pathology, and religion as a treatment adjunct and benefactor of recovery. Thus, any relationship between religiosity/spirituality and eating pathology is likely to be highly nuanced, with possibly both positive and negative influences.

A recent systematic review by Akrawi, Bartrop, Potter, and Touyz (2015) concluded that a superficially extrinsic faith with loosely held beliefs coupled with a doubtful, anxious and questioning relationship with God may be associated with greater levels of DEP. Conversely, strong and internalised religious beliefs along with having a secure and satisfying relationship with God were associated with lower levels of DEP. However, the systematic review highlighted major limitations in that most of these studies failed to explore the nature of these beliefs and the extent that they influence a person’s life. Of the studies that did explore the strength and impact of certain religious and spiritual beliefs, most made use of scales containing heavy Judeo-Christian bias, hence limiting the response of participants without a Judeo-Christian based faith. The instruments used included the Quest (Batson & Schoenrade, 1991), RCOPE (Pargament, Koenig, & Perez, 2000), Religious Orientation (Allport & Ross, 1967) and Spiritual Well-Being (Daaleman & Frey, 2004) scales. Furthermore, most studies accessed predominantly Christian or Jewish samples from private educational institutions in America and Israel, with sparse use of participants attending secular universities. In summary, there is a need for further research accessing more religiously diverse samples using measures that transcend all spiritual, religious and personal beliefs.

Societal trends have revealed an increasing aversion against religious practice and a greater embrace of spirituality amongst university aged students (Astin, Astin, & Lindholm, 2011). For the purpose of this study, religiosity will be defined as ‘an organised system of beliefs, practices, and symbols designed to facilitate closeness to the transcendent’ (Koenig, King, & Carson, 2012, p. 45). Spirituality will be defined as ‘striving for inspiration, reverence, awe, meaning, and purpose even in those who do not believe in God’ (Murray & Zentner, 1989), also encompassing strong personal beliefs. The measurement of specific beliefs poses a challenge in a religiously and spiritually diverse sample of students. However, exploring the strength of beliefs through the impact they may have on an individual’s quality of life provides an alternative means of assessing spiritual, religious and personal beliefs (SRPBs). The World Health Organisation, based on Ellison’s (1983) spiritual conceptual dimensions, has developed an instrument which explores the strength of religious and existential (spiritual and personal) beliefs through the impact they have on quality of life (Skevington, Gunson, & O’Connell, 2013).

Using a methodology intended to minimise past design problems, the purpose of this study is to explore the relationship between SRPBs and DEP through general and multidimensional facets in a diverse sample of students with a wide range of cultural, religious and spiritual affiliations. Religiosity and spirituality have been associated with lower levels of DEP amongst predominantly religious university samples. Thus we hypothesise that strong religious and existential beliefs which greatly impact the quality of life of participants will be associated with lower levels of DEP. We also aim to explore which belief system, if any, best predicts DEP levels in a secular university setting.

2. Method

2.1. Participant recruitment and setting

The survey was constructed using the online survey software Qualtrics (Qualtrics, Provo, UT, USA), with participants given a web-link for access. Consent was obtained through the submission of the survey, and all responses were anonymous. In addition to answering structured measures (see below), participants were asked to provide information on their height, weight, age and religious affiliation.

A total of 936 female university students from Western Sydney University were recruited through a combination of student emails, university social media and community engagement. Participants were given a chance of winning one of four shopping gift vouchers. Of those targeted by the recruitment, 200 failed to complete the survey in its entirety, with a further 121 students failing to attempt the survey. An additional 72 first-year psychology students at the University of Sydney were recruited, obtaining credit points for research participation, leaving a final sample of n = 687. Ethics approval was obtained from both institution’s Human Research Ethics Committees.

2.2. Measures

2.2.1. Disordered eating psychopathology

DEP was measured using the DT, B and BD subscales of the Eating Disorder Inventory 3 (Garner, 2004). DT describes an extreme desire to be thin, excessive concern with dieting, a preoccupation with weight, and fear of weight gain. B describes the extent of thoughts about engaging in episodes of binge eating. BD describes the general dissatisfaction with the size and shape of particular areas of the body. The three subscales have established reliability and validity in non-clinical samples of females (Clausen, Rosenvinge, Friborg, & Rokkedal, 2011). In the present study, all Cronbach alphas were high: 0.88 (DT), 0.85 (B) and 0.89 (BD).

2.2.2. General levels of religiosity and spirituality

The ‘general’ SRPB section of the World Health Organisation Quality of Life – Spiritual, Religious and Personal Beliefs (WHOQOL-SRPB) instrument (2006) was used to measure general levels of religiosity and spirituality. Religiosity was measured using the question ‘To what extent do you consider yourself to be a religious person?’ with a 5-point Likert-type scale ranging from 5 (an extreme amount) to 1 (not at all). Using a single item to measure religiosity is consistent with similar studies (Boisvert & Harrell, 2012, 2013; Homan & Cavanaugh, 2013). Spirituality was measured using the question ‘To what extent do you have spiritual beliefs?’ with an identical scale.

2.2.3. Spiritual, religious and personal beliefs

The WHOQOL-SRPB bref (Skevington et al., 2013) measures quality of life across five alternate domains. The SRPB domain of the WHOQOL-SRPB bref (Skevington et al., 2013) was used to assess the strength of spiritual, religious, and personal beliefs through the impact they may have on quality of life. The initial instrument was developed and reviewed by 92 focus groups across 15 countries, and transcends cultural, religious and spiritual affiliation, making it suitable for use in a secular university setting (WHOQOL SPRB Group, 2006). With the original instrument having nine facets, factor analysis by the authors of the instrument revealed two subscales: religious beliefs and existential beliefs (Skevington et al., 2013). Religious beliefs include facets relating to spiritual connection, spiritual strength, faith and wholeness. Existential beliefs relate to facets which include hope and optimism, meaning and purpose, and inner peace. For each facet, answers are scored on a 5-point Likert scale ranging from 1 (not at all) to 5 (an extreme amount). The instrument has shown good validity and reliability (Skevington et al., 2013; WHOQOL SPRB Group, 2006).

2.3. Statistical analyses

All statistical analyses were conducted using SPSS (v22; IBM, Armonk, NY, USA). Initial descriptive analyses of the data were conducted including frequency, variability and central tendency (where relevant) for all demographic variables and scale/subscale measures. Thereafter scores on the standard measures were further examined for normal distribution. Analysing BMI scores revealed six low BMI outliers which were removed from further analyses. Because the EDI-3 subscales were all heavily skewed to the positive, and the WHOQOL-SRPB bref subscale scores showed both positive and negative skew, non-parametric analyses were used. Spearman correlation coefficients assessed the association between the disordered eating subscales (DT, B, BD) (dependent variables) and general levels of religiosity and spirituality (independent variables). Spearman correlation coefficients were also used to assess the association between the disordered eating subscales and spiritual, religious and personal beliefs subscales. As a final step, hierarchical regression analyses were conducted to determine significant unique predictors of each disordered eating variable.

3. Results

3.1. Participants

The mean age of participants was 21.19 years (SD = 3.18), with a range of 17–35. The mean body mass index (BMI) of participants was 23.67 (SD = 5.02), with a range of 16.14–45.72. Most (69.6%) had some degree of religious belief while 85.3% had some degree of spiritual belief. Regarding religious affiliation, 44.9% had no religious affiliation, 33.8% were Christian, 7.5% were Muslim, 5.4% were Hindu, 3.7% were Buddhist, 0.7% were Jewish, 0.9% identified as other, and 3.08% did not identify a specific religious affiliation.

Descriptive statistics for the EDI-3 and WHOQOL-SRPB bref subscales are shown in Table 1.

3.2. Association with general levels of religiosity and spirituality

There was a significant correlation between religiosity and BD (r = –0.08, p < 0.05), indicating that lower levels of religiosity were associated with higher levels of BD. However no other correlations reached statistical significance (see Table 2).

3.3. Associations with spiritual, religious and personal beliefs

DT was negatively correlated with both Religious Beliefs (r = –0.09, p < 0.05) and Existential Beliefs (r = –0.20, p < 0.001) (see Table 2). B was also negatively correlated with both Religious Beliefs (r = –0.11, p < 0.01) and Existential Beliefs (r = –0.30, p < 0.001). Furthermore, BD was negatively correlated with both Religious Beliefs (r = –0.20, p < 0.001) and Existential Beliefs (r = –0.32, p < 0.001). It was evident that both SRPB facets were associated with lower levels of DEP.

Table 1. Descriptive statistics for standardised measures

Measure	Range	Mean (SD)
EDI subscales
Drive for thinness	0–28	11.46 (7.80)
Bulimia	0–32	6.41 (6.23)
Body dissatisfaction	0–40	17.46 (9.34)
General levels of religiosity and spirituality
Religiosity	1–5	2.39 (1.18)
Spirituality	1–5	2.96 (1.28)
WHOQOL-SRPB bref subscales
Religious beliefs	4–20	11.67 (4.38)
Existential beliefs	3–15	10.43 (2.55)

Table 2. Spearman correlations between disordered eating psychopathology, demographic characteristics, general levels of religiosity and spirituality and SRPB subscales

	EDI subscales
	Drive for thinness	p	Bulimia	p	Body dissatisfaction	p
Demographics
Age	0.01	0.84	−0.03	0.46	−0.05	0.24
BMI	0.34	0.00	0.31	0.00	0.51	0.00
Religiosity and spirituality variables
General levels
Religiosity	−0.01	0.90	0.01	0.96	−0.08	0.03
Spirituality	−0.02	0.58	0.01	0.83	−0.07	0.06
WHOQOL-SRPB bref subscales
Religious beliefs	−0.09	0.02	−0.11	0.01	−0.20	0.00
Existential beliefs	−0.20	0.00	−0.30	0.00	−0.32	0.00

3.4. Hierarchical regression analyses

Hierarchical regression analyses assessed the contribution of significant univariate variables in predicting DEP. Step one consisted of age and BMI, with step two containing the two SRPB subscales. Age and BMI predicted 8% (F [2, 569] = 25.18, p < 0.001) of variance, with both Religious Beliefs and Existential Beliefs explaining an additional 4% (F [4, 567] = 19.78, p < 0.001) of variance in DT (see Table 3). Age and BMI predicted 10% (F [2, 569] = 32.44, p < 0.001) of variance, with both Religious Beliefs and Existential Beliefs explaining an additional 8% (F [4, 567] = 32.85, p < 0.001) of variance in B. Finally, age and BMI predicted 24% (F [2, 569] = 91.06, p < 0.001) of variance, with both Religious Beliefs and Existential Beliefs explaining an additional 10% (F [4, 567] = 75.60, p < 0.001) of variance in BD.

Table 3. Hierarchical multiple regression analyses predicting disordered eating psychopathology from age, BMI and SRPB subscales

	Drive for thinness
	R^2	β
Predictor
Step 1	0.08***
Age and BMI
Step 2	0.12***
Religious beliefs		0.06
Existential beliefs		–0.23***
Bulimia
Step 1	0.10***
Age and BMI
Step 2	0.18***
Religious beliefs		0.08
Existential beliefs		−0.33***
Body dissatisfaction
Step 1	0.24***
Age and BMI
Step 2	0.34***
Religious beliefs		0.03
Existential beliefs		–0.34***

*p < 0.05, **p < 0.01, ***p < 0.001

The sample was then divided into two groups consisting of participants with and without a reported religious affiliation. Identical hierarchical regression analyses were conducted revealing similar results, with only Existential Beliefs remaining a significant predictor of lower levels of DT, B and BD amongst both groups.

4. Discussion

In addressing the limitations of the current literature, the aim of this study was to explore religiosity, spirituality and the strength of these beliefs through the impact on an individual’s quality of life in relation to DEP.

Although the study revealed a small (Cohen, 1988) association between religiosity and BD, general levels of religiosity and spiritualty were not significantly related to DEP. The result is at odds with current literature reporting a positive relationship between religious observance and lower levels of DEP (Gluck & Geliebter, 2002; Latzer, Tzischinsky, & Gefen, 2007; Mahoney et al., 2005). Our finding may be explained by the recruitment of a religiously and spiritually diverse sample of secular university students compared to recruitment from private American universities reporting high percentages of Christian (53–99%) and/or Jewish participants (Akrawi et al., 2015). While Judeo-Christian practices, including prayer and the reading of religious text, have been shown to improve body image concerns (Boyatzis, Kline, & Backof, 2007; Kim, 2006), only 34.5% of participants in the current study identified with a traditional Judeo-Christian religious affiliation. Furthermore, religiosity and spirituality are multidimensional concepts which have alternative meanings to people of varying religious and spiritual affiliations. Capturing these concepts exclusively through a single item scale limits this complexity.

In exploring the strength of SRPBs through the impact they may have on an individual’s quality of life, results suggest that existential beliefs, and to a lesser degree religious beliefs, are associated with lower levels of DT, B and BD. However, only existential beliefs remained a significant predictor of lower levels of DEP throughout hierarchical regression analyses, even amongst participants who associated with a religious affiliation. This contrasts with evidence reported in a recent systematic review (Akrawi et al., 2015) which summarised studies showing evidence for associations between religious beliefs and lower levels of DEP in traditional religious samples. According to the authors of the SRPB scale (Skevington et al., 2013), religious beliefs reflect the extent to which a person’s faith and a connection to a transcendental being helps an individual get through hard times and live a better life. Existential beliefs, on the other hand, represent the extent to which the participant’s spiritual or personal beliefs, possibly independent from a transcendental being, provide hope, meaning and inner peace. Evidently, spiritual and personal beliefs may have a greater role in predicting lower levels of DEP than religious beliefs in a secular university setting.

While disordered eating is often driven and compounded by anxiety regarding food and body image concerns, with psychological stresses and poor coping skills associated with the development of DEP (Ball & Lee, 2000; Troop & Treasure, 1997), certain theoretical accounts of eating disorders incorporate mind and body battles and interoceptive awareness issues (Garner, Olmstead, & Polivy, 1983; Garrett, 1996; Lester, 1997). However, profound spiritual and personal beliefs have the potential to supply sacred meaning to daily living, and thus provide a mechanism to cope with and alleviate stress (Bryant-Davis & Wong, 2013; Fleck & Skevington, 2007; Hill & Pargament, 2003; Miller & Thoresen, 2003; Pargament, 1997; Pargament et al., 2000). Indeed, yoga, mindfulness and meditation have all been associated with lower levels of disordered eating attitudes and behaviours (Lavender, Jardin, & Anderson, 2009; Scime & Cook-Cottone, 2007).

4.1. Conclusions

Existential beliefs and, to a lesser degree, religious beliefs were both associated with lower levels of DT, B and BD. However strong existential beliefs alone best predicted lower levels of DEP, even amongst participants with a religious affiliation. These results reveal that spiritual and personal beliefs may have a greater role in predicting lower levels of DEP than religious beliefs in a secular university setting.

4.2. Limitations and strengths

There are several limitations that should be noted. Firstly, the study design was cross-sectional in nature, rendering the inability to determine directionality and causation. Secondly, a non-randomised convenience method of sampling was used which could have led to sampling bias. Thirdly, the study made use of a non-clinical sample, with results not generalisable to clinical and sub-clinical populations. Furthermore, our results displayed low R² values, meaning that other complex variables are also contributing to the variance in DEP. Nevertheless, a large sample of students was obtained, contributing to a well-powered study. Valid and reliable measures, which transcend religious and spiritual beliefs, were also utilised to suit the diversity of the sample population. A final strength of the study was that it was the first study to explore WHOQOL-SRPBs and DEP in a secular university setting.

4.3. Implications and future research

The findings may have important clinical implications including the exploration of existential and religious mechanisms within treatment models. The current ‘gold standard’ treatments for eating disorders do not directly refer to religious and existential concepts as derived from SRPBs, though aspects of many therapies do use related concepts. For example, acceptance and mindfulness based therapies/techniques incorporate accepting the moment, being non-judgemental, and fostering perspectives of self/spiritual transcendence (Hofman & Asmundson, 2008). Similarly, Cognitive Behaviour Therapy (CBT) targets mind-body processes directly relevant in promoting emotion-regulation, and ‘adopt(ing) a more relaxed and satisfying system of values’ (Hofman & Asmundson, 2008, p. 10). This too seems aligned with concepts of achieving inner peace, meaning and hope. Indeed, CBT frameworks have been developed that more deliberately emphasise religious constructs (Pearce et al., 2015). The results from this study may help in understanding what components of standard psychological treatments may be helpful adapting into the treatment of eating disorders, taking into account differing cultural, religious and spiritual affiliations. Finally, such findings may allow health professions to develop more informed prevention programs for disordered eating.

Future research should be directed at the protective role of existential and religious beliefs through mediation and moderation analyses, and how the facets of these beliefs are best fostered, if indeed these aspects prove to be helpful in addressing symptoms. This would provide stronger evidence for their use in community settings. Further exploration of the nuanced mechanisms by which SRPBs may influence DEP, including how people with disordered eating may use SRPBs to cultivate and justify symptoms, is also needed. Additionally, longitudinal and experimentally designed studies should be conducted to explore directionality and causality.

Acknowledgements

Roger Bartrop passed away on the 3rd February 2017.  We would like to dedicate this paper to him.  Roger had both  the vision and foresight to bring this group of investigators together from diverse backgrounds to study religiosity in eating disorders. We are eternally grateful to him. We would like to acknowledge Michael Lee and Rohit Ghosh for assisting with survey administration.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The project was funded through a summer research scholarship grant provided by the School of Medicine at Western Sydney University.

Notes on contributors

Daniel Akrawi

Daniel Akrawi, MBBS (Hons) (1st class), is a medical officer at Liverpool Hospital, and holds a conjoint associate lecturer position at Western Sydney University. His research to date has focused on the roles of religiosity and spirituality in the development of disordered eating psychopathology.

Roger Bartrop

Roger Bartrop, was the Foundation Professor of Mental Health at Western Sydney University Blacktown clinical school, and the Professor of Psychiatry at the University of Sydney Medical School-Northern. He had developed diverse areas of research interest over his lifetime, which included eating disorders, smoking cessation, music therapy in acute psychosis, resilience amongst spinal injury victims and immunological and psychosocial aspects relating to grief.

Lois Surgenor

Lois Surgenor is an Associate Professor in Psychological Medicine at the University of Otago, Christchurch, New Zealand. Her research interests are in the area of eating disorders, psychological aspects of long-term and life-shortening medical conditions, and health workforce well-being and regulation. As a clinical psychologist she has over 25 years experiencing assessing and treating people with eating disorders and has twice been involved in developing Australasian practice guidelines in this clinical area. She is an Associate Editor of the Journal of Eating Disorders and on the Editorial Board of the Eating Disorders journal.

Shantiban Shanmugam

Dr. Shantiban Shanmugam is a medical officer at Concord Hospital. His research interests include eating disorders, end of life care in the intensive care unit, end of life planning in aged care and quality and safety improvements in the clinical setting.

Ursula Potter

Dr. Ursula Potter is a literary historian and Honorary Research Associate with the Department of English at the University of Sydney. She is researching green sickness, the disease of virgins, in young women in sixteenth- and seventeenth-century English drama and the relationship between Protestant reform and the rise of religious anorexia at this time. She is currently preparing a monograph for publication entitled Shakespeare and the Taming of the Womb: plotting women’s health in early modern drama, and has a chapter on the association of menarche with sin, Coming of Age with the Curse, forthcoming with Macmillan Interdisciplinary Handbooks.

Stephen Touyz

Professor Stephen Touyz obtained his PhD degree from the University of Cape Town in 1976.  He is the Chair of the Executive Committee of the Centre for Eating and Dieting Disorders at the Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders at the University of Sydney. He has written 7 books, over 330 papers/book chapters and 400 conference abstracts.  In 2012, he was given the prestigious Leadership in Research Award by the Academy of Eating Disorders (USA) in recognition of his pioneering research in the field of eating disorders and In 2014, he was awarded the first Lifetime Achievement Award by the Australian and New Zealand Academy of Eating Disorders. He is the founding co-editor in chief of the Journal of Eating Disorders and a member of the Editorial Advisory Boards of the International Journal of Eating Disorders, European Eating Disorders Review and Advances in Eating Disorders: Theory, Research and Practice