Introduction

Eating disorders (ED) are a complex group of psychological disorders that cause detrimental individual hardship and high costs to the American health care system [1–3]. Psychological research aims to understand the causes and comorbidities of ED; one possible important pathway to the development of an ED is the experience of sexual trauma. This review will be an in-depth assessment of the link between sexual trauma and ED. Sexual trauma will be considered a distal factor in the development of ED, something that may not directly cause the onset of the ED, but may precede the onset by a significant amount of time [4]. Thus, experiencing sexual trauma may add to a person’s risk to develop an ED and falls within the diathesis-stress model. The diathesis-stress model theoretically represents the development of psychopathology as a manifestation of individual vulnerabilities (including neurobiology and genetics) combined with difficult life events (that may be chronic or acute) [5]. Such a theoretical model may help elucidate the mechanisms underlying the development of ED.

Based on current literature, there appear to be two main etiological pathways between sexual assault and ED. The first etiological pathway consists of the survivor’s negative perception of her body, including body dissatisfaction, shame, and fear of future sexual trauma [6–9]. The second pathway consists of the survivor of sexual trauma’s need to cope with psychological difficulties, including failure of the average expected environment, need for control, coping with psychological diagnoses, and regulation of emotions [10–14]. This review will assess research related to these two pathways: body perceptions and psychological difficulties.

Previous reviews of the literature: the need for an updated review

Previous literature reviews have aimed to assess the potential relationship between sexual trauma and ED and have shown mixed results [15–18]. There has not been a clear consensus on whether a significant relationship is present, and if so, whether this relationship is causational. A 1993 review did not find a higher rate of sexual assault in ED populations compared to the general population [17]. A 2002 review paper established a small relationship between ED and childhood sexual abuse, but noted the inclusion of problematically heterogeneous studies [18]. In a meta-analytic review of risk factors for ED, longitudinal studies did not show a relationship between sexual trauma and ED [16]. However, recent modern meta-analyses of the relationship between sexual trauma and psychiatric disorders may be limited by not including the ED diagnosis from the DSM-IV: ED not otherwise specified (EDNOS) [15]. Therefore, the current literature review includes studies that assess for EDNOS, binge ED, and sub-clinical ED symptomatology, in addition to anorexia nervosa (AN) and bulimia nervosa (BN). Due to the small number of studies that address the relationship between ED and sexual trauma, the authors have chosen to be inclusive when differentiating which studies should be included in this review. As such, this literature review includes studies that do not differentiate between the different ED diagnoses, studies that assess eating disorder symptomatology, and studies that include more than one ED diagnosis.

Method

This article identified and reviewed literature from the past 10 years (2004–2014) that pertains to the potential relationship between sexual trauma and ED using PubMed, GoogleScholar, and PsychINFO as search engines. We utilized the search terms “sexual assault”, “sexual abuse”, “sexual trauma”, and “rape” in conjunction with relevant ED terminology, such as “bulimia”, “anorexia” and “binge eating”. As this is not a systematic review, we evaluated individual pieces of literature based on their potential contribution to understanding the etiology of ED as it relates to sexual trauma. This article excluded studies that are exclusively qualitative, nonhuman, unpublished, non-English, literature reviews, and letters to the editor. Moreover, this review only included studies that report statistical significance of factor(s). Assessment of study quality was adapted from the ten criteria established by Zhang and Wang [19]. The ten Zhang and Wang criteria for study acceptability for inclusion in a review paper were as follows: “provided operational definition of dependent variable (DV)”, “tested own DV data’s validity”, “tested own DV data’s reliability”, “tested own independent variable (IV) data’s validity or reliability”, “random/probability sampling”, “employed longitudinal design”, “interview conducted…to ensure cultural sensitivity and accuracy”, “used multiple/logistic regression as the highest level of statistical analysis”, “reported effect sizes”, “presented theoretical linkages to connect IV and DV” [19]. No studies were included with a quality score less than three. Although this criterion level appears relatively low, we chose to include a large range of studies due to the nature of the research question and to assess strengths and weaknesses found in the research literature. Although rates of ED have been shown to be much higher in women compared to men [2], study samples of both genders will be included in this literature review. The literature review yielded 32 relevant empirical studies (see Table 1). An additional 36 articles were found given the above search terms. However, upon further analysis of the relevancy of the research, these articles were not included in the literature review (list available upon request).

Table 1 Reviewed manuscripts covering the potential relationship between sexual trauma and eating disorders
Full size table

Results

Co-prevalence of sexual trauma and ED: 2004–2014

Results showed an increased prevalence of sexual trauma in ED women and that sexual trauma was associated with several different types of ED symptoms [20, 21]. Several studies have examined the temporal relationship between ED and trauma within inpatient settings. One inpatient unit reported that 38 of 95 consecutively admitted patients endorsed childhood sexual trauma that preceded ED onset. Additionally, twelve patients endorsed sexual trauma that occurred post ED onset, and six patients endorsed sexual trauma that occurred in adulthood [22]. This rate of 56 of 95 patients (59 %) is much higher than the rates of sexual trauma in the community: roughly 30 % [23]. A 64 % rate of sexual trauma has also been reported for mid-life onset ED inpatients [24].

Studies conducted with non-treatment seeking individuals also show relevant findings of a relationship between sexual trauma and ED. Women with ED versus healthy controls showed a higher rate of sexual trauma; results also indicated that sexual trauma preceded ED onset for most women who endorsed both sexual trauma and ED [25]. An 11-year longitudinal study in Australia found that young women were 4.9 times more likely to obtain a diagnosis of BN during adolescence after multiple incidences of childhood sexual trauma, and 2.5 times more likely after one episode of child sexual trauma [26]. In a large population-based study, ED women showed a prevalence of childhood sexual trauma more than twice that of non-ED women, a statistically significant difference [27]. In a college-aged sample that queried both childhood and adult experiences of sexual trauma, results showed that childhood sexual trauma was not associated with ED symptomatology, while sexual trauma within the last 3 months was significantly related to ED symptoms [28]. These differing results may point to a dynamic relationship between the timing of sexual trauma and ED symptoms.

The relationship between sexual trauma and specific ED diagnostic criteria unique to a particular ED has also been studied. Researchers assessed the variability of risk factors between the different ED diagnoses (full or partial diagnostic criteria), and found sexual trauma was a risk factor for all ED diagnoses, and did not add an increased level of risk for any specific ED diagnosis [29]. A more recent, large-scale population study showed that women with BN had higher rates of sexual trauma compared to women who met criteria for binge ED or AN [30]. Although large epidemiological studies rarely evaluate full diagnostic criteria, large scale data has shown increased likelihood to maintain one or more ED symptoms if one has experienced a sexual trauma in the past [31]. These results are replicated by a recent study of sister-dyads, which found that sisters diagnosed with ED were significantly more likely to have experienced sexual trauma [31]. However, many other studies have concentrated on individual ED diagnoses, possibly to obtain more homogeneous samples.

Examining the relationship between non-sexual trauma, sexual trauma, and ED, researchers compared women who had experienced one sexual trauma to women who had experienced one non-sexual trauma. Results showed women with sexual trauma are more likely to develop ED compared to women with non-sexual trauma [8]. Research literature has also shown a two-fold increase in the odds of developing ED if a woman has experienced physically abusive trauma as a child compared to women who reported no trauma during childhood [32]. Women in this study who reported both physical and sexual abuse experienced a four-fold increase in the odds of developing ED [32].

Many studies have focused on AN, perhaps because it is the ED with the highest rates of mortality [3]. In a retrospective study, AN participants endorsed higher rates of sexual trauma compared to both normal controls and a comparison group of non-ED psychiatric participants [34]. A significant subset of the literature has explored the subtypes of AN and has published a number of studies showing an increased rate of childhood sexual trauma in cases of AN binge-purge type (AN-BP) as opposed to AN restricting type (AN-R) and control participants. Rates of sexual trauma were higher in women with AN-BP than in women with AN-R [22, 35]. A German study compared rates of sexual trauma among three groups: women with AN-BP, women with AN-R, and healthy control women. AN-BP women had higher rates of childhood sexual trauma, while there were no significant difference between the AN-R and healthy control groups [35]. In a related study, women with comorbid AN and post traumatic stress disorder (PTSD) were more likely to have a diagnosis of AN-R than AN-BP. Although the sample was not exclusively women who had experienced sexual trauma, child sexual trauma was the most common precipitating factor for PTSD in the sample [14]. Different findings across ED subtypes may indicate that a variety of etiological factors may contribute to specific ED symptoms and diagnoses after sexual trauma.

In sum, the literature suggests that sexual trauma and ED often co-occur and some studies support the temporal relationship that sexual trauma precedes ED. The presence of this relationship is relevant as the relationship between sexual trauma and ED may aid both trauma therapists and eating disorder specialists to more accurately assess and treat their patients who may be suffering from both psychological stressors. Further studies are discussed according to the two theorized etiological pathways: the body perception pathway and the psychological difficulties pathway.

Etiological pathway 1: body perceptions

An individual may experience widespread difficulties in her relationship with her body following a sexual trauma. Recent research literature evaluates a potential etiological pathway to the development of ED after sexual trauma that occurs through mechanisms of body dissatisfaction, shame, sexual dysfunction, and fear of future sexual trauma. Eubanks et al. [36] have established that adult women with a history of childhood physical and/or sexual trauma have greater variability in their body-size perception compared to women with no history of trauma. Perhaps this difficulty in accurate self-evaluation may also work through the mechanisms of body dissatisfaction, shame, and sexual dysfunction following sexual trauma.

Body dissatisfaction

ED are directly linked to body dissatisfaction and low self-esteem, often in conjunction with underlying perfectionism [37]. AN could result from body dissatisfaction based on two motivations: the desire to change one’s body to be thinner, and the desire to deny one’s body nourishment as punishment. An Italian research study found individuals with a history of childhood sexual trauma had higher body dissatisfaction and ED symptomatology than those who experienced no sexual trauma, and those who experienced sexual trauma after puberty [7]. Further evaluation of this data found that body dissatisfaction maintains a mediating role between childhood sexual trauma and ED symptomatology [7]. These results are contrary to previous findings showing that a history of physical trauma, but not sexual trauma, increased body dissatisfaction. However, this study found a very high rate of sexual trauma within the sample of ED women (29 %) [38].

Shame

Related to body dissatisfaction, shame has shown a distinct relationship to AN symptomatology [39]. The relationship between sexual trauma and shame is well documented [40–42], as is the relationship between shame and ED [43, 44]. Separate research showed feelings of shame to be more severe in ED populations than in populations experiencing anxiety and depression [45]. Women who feel shame regarding their own sexual trauma may restrict their eating as a form of punishment. It is possible for shame to develop into feelings of hatred for one’s own body [44]. This could also result in a desire to limit food if the individual develops shameful thoughts such as, “I do not deserve to live”, or “the sexual trauma was my fault and I should be punished”. These negative thought patterns could lead to ED if left untreated.

Sexual dysfunction

Feelings of shame and body dissatisfaction are highly comorbid with sexual dysfunction in women, and sexual dysfunction may play a role in the relationship between sexual trauma and ED. One study found that women who had survived sexual trauma were significantly more likely to experience aversion to sexual activity (85 % of sexual trauma sample) compared to women who had experienced a non-sexual trauma [8]. A treatment study found that improved ED symptomatology was associated with improved sexual functioning; however, this relationship was not significant for individuals with a history of sexual trauma [9]. Thus, a history of sexual trauma may interfere with ED treatment outcomes, including outcomes related to sexual functioning. However, more research is necessary to tease apart potential temporal relationships and possible causal factors.

Fear of future sexual trauma

Another pathway from sexual trauma to an ED is the desire to minimize the chances of future sexual trauma. ED patients identify the desire to “become unattractive” to the opposite sex as motivation behind their eating pathology [6]. To a woman who has survived sexual trauma, becoming unattractive may be a defensive strategy. A further strategy involves the binging cycle, which includes gaining weight in order to “become bigger to be intimidating” so as to protect one’s self from future trauma [6]. These motivations may result in different ED diagnoses (e.g., binge eating to gain weight to defend one’s self physically or restricting food intake to lose physical signs of sexual maturity) [46]. Boynton-Jarrett et al. [21] recently found that African-American women had an increased likelihood of becoming obese as adult if they experienced trauma as adolescents. This pathway could contribute to the development of an ED independently or in conjunction with other etiological pathways, such as the psychological difficulties pathway discussed below.

Etiological pathway 2: management of psychological difficulties

The second etiological pathway from a sexual trauma to an ED conceptualizes the ED as a coping mechanism or reaction to psychological stress. Previous literature reviews have conceptualized ED as directly related to negative affect and cognitive biases [16, 47]. This pathway may develop in a number of ways: an individual may develop an ED after experiencing a failure of the average expectable environment, out of a need for control of the environment, as a reaction to psychological difficulties, or for regulation of emotions and impulsivity.

Failure of the average expectable environment

Failure of the average expectable environment may be a unique etiological pathway from a sexual trauma to ED. The relationship that the perpetrator of sexual trauma has with the survivor of sexual trauma may influence this pathway. The identity of the perpetrator of sexual trauma relates to willingness to disclose sexual trauma, and childhood sexual assault perpetrated by a relative is associated with higher levels of shame compared to when a perpetrator is a non-relative [48, 49]. Indeed, female adolescents with a history of sexual trauma are less likely to consider their family cohesive and more likely to consider their family in conflict compared to adolescents without a history of sexual trauma [50]. These findings are particularly relevant because AN women with a history of childhood sexual trauma report that the majority of the perpetrators were the AN patient’s father [13]. Family difficulties may also arise following childhood sexual abuse, as fear of abandonment and mistrust may mediate the relationship between childhood sexual trauma and ED symptoms [51].

Desire to control environment

The failure of the average expected environment may also relate to an unhealthy desire to control one’s environment, which may be another factor contributing to the etiological pathway from sexual trauma to ED. Sexual trauma contributes to a loss of trust in the environment [52] and a desire to control the environment to prevent a reoccurrence of trauma. Similarly, ED are linked to a desire to control one’s body [37]. Thus, the desire to control one’s body may be an outgrowth of a desire to control one’s environment. Female ED college students have been shown to experience losing control more than female controls, and that this effect is higher in ED women with a history of sexual assault [53]. Additionally, compulsivity mediated the relationship between childhood sexual trauma and ED symptoms in a sample of coed college students [10]. Compulsivity may manifest in the desire to control one’s environment as a means to alleviate distress [54].

Response to psychological difficulties

A maladaptive stress response to psychological difficulties may also play a role in the etiological pathway from sexual trauma to ED. Over 40 % of men and women with binge eating currently suffer from a comorbid psychiatric diagnosis, and over 70 % suffer an additional psychiatric diagnosis during their lifetime. Current psychiatric comorbidity is associated with higher ED symptomatology [55]. ED and anxiety are highly comorbid (65 %), and the majority of anxiety disorders onset before the ED [56]. In adolescents who have previously attempted suicide, childhood sexual assault is positively related to ED [15, 57]. A separate study shows psychological distress and alexithymia mediated the relationship between childhood sexual trauma and ED in adulthood [12]. The highest rates of lifetime psychiatric comorbidities in adolescents with ED are in adolescents with BN (88 %) and binge ED (83.5 %). Adolescents with the lowest rates of lifetime psychiatric comorbidities were those with AN (55 %) [58]. Thus the psychological difficulties etiological pathway may be more related to BN and the body perceptions etiological pathway may be more related to AN. This could have implications when considering theories behind the etiology of an ED following sexual trauma.

In a model of sexual trauma and ED symptoms, results show depression and PTSD to be significant mediators, thus showing either comorbidity can play a mediating role [59]. In particular, the fact that rates of ED increase following both sexual and physical trauma may indicate that the common factor between the two—PTSD—may be the mediating factor between trauma and ED. A recent review of the literature shows significant overlap between PTSD and ED, with a significant portion of PTSD symptoms presenting as a result of sexual trauma [60]. A study of the comorbidity of PTSD and AN showed 13.7 % of AN patients met criteria for PTSD, and that the majority of these PTSD cases were due to sexual trauma. Of those AN patients with comorbid PTSD, 64 % endorsed that the onset of PTSD preceded the onset of AN [14]. The strongest evidence for PTSD as a mediator of sexual abuse and ED may come from statistical regression models that show a decrease in the relationship between sexual trauma and ED symptomatology when PTSD is added to the model [61].

Emotion and behavior regulation

The final aspect of the speculated etiological pathway concerns two underlying mechanisms—emotion regulation and regulation of impulsivity—that may relate to both sexual trauma and ED and may help to elucidate the link between sexual trauma and ED. In ED, the attempt to regulate emotions and impulsivity is one of the prominent symptoms of the disorder. AN patients prefer to feel “numb” as they engage in disordered eating, rather than feel emotions [62]. BN patients tend to be emotionally dysregulated and impulsive [62, 63]. BN patients engage in all types of disordered eating—binging, purging, restricting—in an effort to regulate emotion [64]. Of note, BN patients appear to be overly dysregulated, and AN patients appear to be emotionally over-regulated [65]. Indeed, the concept that impulsivity may be a precursor to ED has been a long-debated theory [16, 66, 67]. BN in particular relates to impulsivity [68], but meta-analysis has also shown that individuals with any ED diagnosis have higher impulsivity compared to controls [69].

Survivors of sexual trauma have also been shown to have difficulty with emotional regulation and impulsivity [70, 71]. Impulsivity mediates the relationship between childhood sexual trauma and disordered eating symptomatology [10]. Although ED patients may develop ED due to abnormal emotional regulation and impulsivity, sexual trauma may be a unique catalyst in this process. Sexual trauma may be too stressful for the individual to tolerate without developing a psychological diagnosis. Researchers must assess the entire etiological model to understand the development of ED, including sexual trauma, emotional regulation, and impulsivity. In the recent literature, researchers have identified evidence of a link between childhood sexual trauma and a later emergence of impulsive behaviors in ED women, including alcohol abuse, substance abuse, and cutting [72]. A mediation model showed impulsivity to be a significant mediator of the relationship between childhood sexual trauma and disordered eating [10]. Child sexual trauma alone is also significantly related emotional dysregulation diagnoses [73].

Overall, there appear to be two different etiological pathways that potentially lead from sexual trauma to ED. Although there is variability in reported prevalence rates, many more individuals experience sexual trauma [74], compared to the number of individuals who develop ED [2, 3]. Therefore, there is not a direct pathway from sexual trauma to ED. It is possible that an innate or learned vulnerability must be present in order for an individual to develop ED after experiencing sexual trauma. Sexual trauma precedes ED at the same rate as other psychiatric diagnoses [34] and thus, may be only one of the many precursors to ED. Likewise, sexual trauma may be a catalyst for psychiatric difficulty, and ED may be only one manifestation of symptoms. It is also possible that an underlying biological or neurological vulnerability initiate an individual on the etiological pathway from a sexual trauma to an ED diagnosis (stress-diathesis model).

Summary and future directions

The literature included in this review shows a unique relationship between sexual trauma and ED across various domains. Although less than half of the studies reviewed reported specific findings regarding the chronology of sexual trauma and ED onset, each study that reported findings regarding directionality asserted that sexual trauma preceded the onset of ED. However, many of these studies used retrospective reports of chronology of sexual trauma, which have been shown to have questionable reliability [75]. Of the studies included in this literature review, only one study could establish evidence of causality [8] and showed support for sexual trauma preceding and contributing to the ED. However, the lack of replicated evidence for causality and directionality of the relationship between sexual trauma and ED is a problematic gap in the research literature.

Findings from the current literature review were able to build upon the findings of previous literature reviews concerning the relationship between sexual trauma and ED. Two of the most recent reviews of the literature found a significant relationship between sexual trauma and ED [15, 18]. Smolak and Murnen’s [18] review concentrated on childhood sexual abuse, which may account for the significant relationship; it is possible that sexual trauma causes distinctly severe psychological trauma if it occurs in childhood. The Chen et al. literature review is the most recent publication, utilized effect sizes, and established a relationship between sexual trauma and ED. However, the current literature review contradicts findings of two previously discussed literature reviews. While two aforementioned literature reviews [16, 17] did not find a significant relationship between sexual trauma and ED, there were several potential reasons to explain their null findings. One possible explanation for the lack of findings in the 1993 literature review was variability in the diagnostic criteria and samples used in the reviewed studies [17]. A potential explanation for the 2002 findings is the stringent inclusion criteria for studies limited the sexual trauma literature to one study of binge eaters [16]. While the findings of the current literature review contradict these earlier publications, the research evidence published in the last decade suggests a compelling link between sexual trauma and ED that should not be overlooked.

While this literature review establishes a relationship between ED and sexual trauma, the mechanisms and pathways underlying this relationship are still not fully understood. Understanding the etiological pathways between sexual trauma and ED is invaluable to ED research and treatment. The research literature has pointed to a need for accurate diagnosis of comorbidities and assessment of past trauma in order to optimize ED treatment [76]. Moreover, success in treatment programs change depending on ED or sexual trauma history. A longitudinal study of 77 women’s outcomes following inpatient ED treatment found that childhood sexual trauma predicted more severe ED symptomatology 5 years post-treatment [77]. In a study of treatment for childhood sexual trauma, AN was shown to moderate treatment effects such that women who were participating in trauma treatment had worse outcomes if they also had a history of an ED [78]. Due to the high cost of ED to the community and the individual [1–3], every effort should be made to increase scientific understanding of the etiology of ED.

Researchers are limited in their ability to explore the relationship between ED and sexual trauma due to a number of factors. ED diagnoses have a low base-rate in the community [2], and individuals with ED are unlikely to seek treatment for this psychological problem [79, 80]. Many persons with ED who are unmotivated to seek treatment for their diagnosis may be equally unmotivated to enroll in research studies. Thus, acquiring large voluntary research samples is difficult. Many research studies have utilized inpatient units to recruit participants, but this is only a subset of the larger population of ED [3]. This type of sampling may be omitting much of the ED population including less acute cases. Thus, the research surrounding the relationship between ED and sexual trauma will be improved via the same avenue that all ED research may be improved: better recruitment and retention of samples for research studies. The strategies for this goal may be multifaceted and include lowering the stigma of mental health treatment and ED treatment, better screening of at-risk populations, more education, additional resources for underserved populations, and more acceptable research studies. Together, these strategies could improve sample sizes, statistical power and conclusions drawn from empirical research studies.

The paramount goal for future research should be to establish the directionality and potential causality of the relationship between ED and sexual trauma. Research only demonstrates the temporal development of ED following sexual assault a handful of times, and many of these studies are retrospective (see Table 1). Researchers remain limited due to the nature of this question because there would be no ethical way to design a randomized study to test the directionality of this relationship. However, individual research studies may prioritize assessments that will provide a timeline of the onset of ED symptoms relative to the occurrence(s) of sexual trauma for each ED participant. Ideally, large-scale population based studies could evaluate both sexual trauma and ED using longitudinal designs to assess the directionality and causality of the relationship. Future research would also benefit from assessing the relationship between sexual trauma and ED diagnosis at different developmental time periods.

Much work remains to be done in the intersection of the fields of ED and sexual trauma research. This manuscript aimed to draw conclusions from the most relevant pieces of research literature published in the last 10 years to synthesize potential etiological pathways from sexual trauma to ED. However, this is just one piece of the greater puzzle. Future psychological research must build upon current theories and empirical evidence.