Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 May 1.
Published in final edited form as: Eat Disord. 2023 Dec 13;32(3):266–282. doi: 10.1080/10640266.2023.2293597

The role of sexual assault history and PTSD in responses to food intake among women with bulimic-spectrum eating disorders

Naomi G Hill 1,1, Sophie R Abber 1, Pamela K Keel 1,*
PMCID: PMC11116069  NIHMSID: NIHMS1951158  PMID: 38093449

Abstract

Sexual abuse or assault (SA) history is associated with eating disorder severity and psychiatric comorbidity, including post-traumatic stress disorder (PTSD). Beyond persistent alterations in mood and cognitions characterizing PTSD, PTSD due to SA may contribute to greater increases in negative affect and body image concerns following food intake in bulimic syndromes (BN-S). To test this, participants (n=172) with BN-S who reported PTSD due to SA, PTSD due to other forms of trauma, or neither completed clinical interviews and momentary reports of negative affect and shape/weight preoccupation before and after food intake. Participants with PTSD, regardless of trauma source, reported higher purging frequency whereas PTSD due to SA was associated with more frequent loss of control eating. For one task, changes in negative affect following food intake differed across the three groups. Negative affect decreased significantly in participants with PTSD without SA whereas nonsignificant increases were observed in those with PTSD with SA. Results of the present study suggest that source of trauma in PTSD may impact likelihood that eating regulates affect and provide insight into ways current treatments may be adapted to better target purging in BN-S comorbid with PTSD.

Keywords: Sexual assault, post-traumatic stress disorder, bulimic syndrome, negative affect, body image, purging

Sexual abuse and assault (SA) history often arise during treatment for an eating disorder, raising questions about how SA history may impact clinical presentation (Backholm et al., 2013). SA history is associated with greater eating disorder severity (Kjaersdam Telléus et al., 2021), particularly in the form of bulimic symptoms (Ackard & Neumark-Sztainer, 2002; Dansky et al., 1997; Fonseca et al., 2002; Groff Stephens & Wilke, 2016; Hazzard et al., 2022; Lie et al., 2021; Molendijk et al., 2017; Perkins & Luster, 1999). Individuals with SA history are more likely to be diagnosed with other psychiatric disorders such as depression, anxiety, and substance use (Dworkin, 2020; Malet-Karas et al., 2022), and SA history, by its nature, is a form of trauma that can lead to post-traumatic stress disorder (PTSD; American Psychiatric Association, 2013). SA history is also associated with greater concerns about shape and weight (Groff Stephens & Wilke, 2016; Hazzard et al., 2022; Jaconis et al., 2020; Preti et al., 2006). In addition to persistent differences in affect and shape/weight concerns, a history of SA may contribute to heightened emotional and cognitive responses to food intake such as greater increases in negative affect or cognitions around shape and weight (Karr et al., 2013; Madowitz et al., 2015). These responses could negatively impact adherence to a regular pattern of eating (Murphy et al., 2010), which comes first in cognitive-behavioral therapy for bulimic syndromes (Fairburn, 2008). Therefore, it is important to understand how SA history impacts responses to food intake in women with bulimic syndromes (BN-S) so that clinicians can tailor treatments to address their unique concerns.

PTSD co-occurs at high rates among individuals with eating disorders (Brewerton, 2007; Dansky et al., 1997; Ferrell et al., 2022). Diagnostic criteria for PTSD include persistent alterations of mood and cognitions following the traumatic event (American Psychiatric Association, 2013). Among women with bulimia nervosa (BN), the presence of PTSD was associated with higher negative affect before and after binge episodes but not with changes in negative affect (Karr et al., 2013). However, this study did not focus on PTSD due to SA specifically, which has been associated with greater increases in negative emotions following the traumatic event than non-sexual trauma (Amstadter & Vernon, 2008). Women with eating disorders are more likely to experience sexual trauma than other forms of trauma, such as combat exposure, accidents, or life-threatening illness (Breland et al., 2018; Gomez et al., 2021; Lie et al., 2021). PTSD resulting from SA may contribute to greater increases in negative affect following food intake compared to PTSD resulting from other forms of trauma.

Shape and weight concerns are elevated among individuals with PTSD and SA history compared to those without (Groff Stephens & Wilke, 2016; Hazzard et al., 2022; Jaconis et al., 2020; Preti et al., 2006). Indeed, SA is more likely to contribute to shape and weight concerns than non-sexual trauma (Sack et al., 2010). Although prior research has examined elevated shape and weight concerns following food intake among women with BN-S (Keel et al., 2018; McKenzie et al., 1993) and women with elevated body image concerns (Vocks et al., 2007), no research, to our knowledge, has examined changes in shape and weight-related cognitions in response to food intake among individuals with PTSD or SA or their combination. Cognitive responses to food intake in individuals with BN-S may distinguish those with PTSD resulting from SA compared to those with PTSD resulting from other forms of trauma.

Despite well-established links between SA, PTSD, and purging (Ackard & Neumark-Sztainer, 2002; Dansky et al., 1997; Fonseca et al., 2002; Groff Stephens & Wilke, 2016; Lie et al., 2021; Perkins & Luster, 1999), more work is required to understand factors that may predispose women to purge. Current theories and treatments tend to emphasize binge eating as the precursor and purging as a means to minimize caloric intake (Fairburn et al., 2003); however, many women with BN and all women with purging disorder purge after consuming normal or small amounts of food (Keel et al., 2005). In addition to compensating for the potential effects of food intake on weight, purging functions to alleviate elevated negative affect and body image concerns following food consumption (Keel et al., 2018). If a history of PTSD and SA contributes to even greater increases in negative emotional and cognitive responses to food consumption, this may increase severity of purging in BN-S. To our knowledge, no research has examined emotional or cognitive responses to consuming smaller amounts of food among women with eating disorders and a history of SA. Understanding the degree of these emotional and cognitive reactions, and whether stronger reactions occur even for small amounts of food, is important for case conceptualization efforts when patients present with an eating disorder and history of PTSD and SA. Linking PTSD and SA history to heightened negative emotional and cognitive responses to food intake may contribute to alternative understandings of the function of purging and enhance interventions by engaging purging as a primary versus compensatory behavior.

Taken together, women with a BN-S and PTSD are likely to have elevated trait levels of negative affect and body image concerns both before and after food intake given persistent elevations in mood and body image disturbance linked to PTSD (Amstadter & Vernon, 2008; Groff Stephens & Wilke, 2016; Hazzard et al., 2022; Jaconis et al., 2020; Preti et al., 2006; Sack et al., 2010). However, little is known about how PTSD resulting from SA versus other forms of trauma impacts momentary affective and cognitive responses to food intake in individuals with BN-S. Examining momentary responses to food intake will provide critical information to better understand factors that contribute to purging. To address this need, the present study utilized secondary analyses of existing data from a study in which women with BN-S completed comprehensive clinical assessments of eating disorder severity and comorbid diagnoses and participated in two behavioral tasks that involved food intake. Immediately before and after food intake, participants provided momentary reports of negative affect and preoccupation with shape and weight. In order to better understand differences between PTSD due to SA history and PTSD due to another type of traumatic event (e.g., combat, car accident), we created three groups: 1) PTSD and SA 2) PTSD without SA, and 3) No PTSD and No SA. The aims of the present study were to: 1) Confirm prior findings linking PTSD and SA history with eating disorder severity and comorbidity and 2) Examine whether PTSD due to SA contributed to different momentary responses to food intake in two behavioral tasks compared to PTSD due to other forms of trauma or those who had neither PTSD nor SA history. Based on prior research, we hypothesized that individuals with PTSD and SA history would have more severe eating disorder symptoms, and higher prevalence of mood, anxiety, and substance use disorders compared to the other groups. We also posited that individuals with PTSD and SA history would experience greater increases compared to other groups in negative affect following food consumption in behavioral tasks and explored whether these patterns would also be true for changes in shape and weight concerns.

Method

Participants

Participants came from a longitudinal study investigating biobehavioral predictors of illness trajectory among women with BN-S (R01MH111263). Participants were included in current analyses if they fell into one of three groups: 1) PTSD and SA (PTSD + SA; n = 46), 2) PTSD without SA (PTSD − SA; n = 19), and 3) No PTSD and No SA (No PTSD/No SA; n = 107). Data were excluded from participants who had a history of SA but no history of PTSD (n = 22), as we were unable to differentiate between those who experienced SA and did not go on to develop PTSD from those who experienced another type of traumatic events (e.g., combat, life-threatening accident) and did not go on to develop PTSD. Participants also had complete data for SA history and PTSD and for the food reward satiation task (N = 172, 76% total sample; see Procedure). A majority of the current sample (N=130, 76%) completed a food reward valuation task (see Procedure) prior to the onset of the COVID-19 pandemic. Thus, sample sizes vary for analyses for these two tasks. Data were excluded from participants who did not provide information on SA history (n = 6) or PTSD (n = 4) during interviews and who did not provide complete data for the behavioral tasks (n = 22; i.e., mechanical problems, incomplete self-report data).

Among the 172 women with BN-S included in analyses, 46 (26.7%) reported a history of SA and 65 (37.8%) reported a lifetime history of PTSD. There was no significant association between group membership and missing behavioral task data (χ2(2)=2.76, p=.43). The presence of a BN-S was operationalized as DSM-5 full-threshold ED or OSFED with the following minimum symptom levels for all BN-S diagnoses: ≥12 episodes of bulimic behaviors (subjective or objective binge eating, purging, or non-purging compensatory behaviors) in the past 3 months, including at least one objective binge episode. Current diagnoses of DSM-5 AN, BN, or BED were based on EDE interview assessments and objectively measured height and weight to calculate BMI. OSFED participants showed evidence of clinical significance based on a) scoring ≥16 on the CIA, b) endorsing distress or impairment on the SCID, and/or c) endorsing marked distress regarding binge eating on the EDE. The sample comprised 7 women with the binge-purge subtype of anorexia nervosa (AN-BP), 92 women with BN, 2 women with binge eating disorder (BED), and 71 women with an other specified feeding or eating disorder (OSFED). The OSFED group comprised 1 woman with atypical AN, 64 with sub-threshold BN, 4 with sub-threshold BED, and 2 with unspecified feeding or eating disorder. Participants were ages 18-35 years (Mage = 20.26, SD = 2.61), had a BMI between 16 and 35kg/m2 (MBMI = 25.12, SD = 4.56), were free of psychotropic medications or on a stable dose of selective serotonin reuptake inhibitors (SSRIs; n = 16) for ≥8 weeks, and were free of alcohol, illicit drugs, and other medications for 72 hours before completing behavioral tasks. Exclusion criteria were: a medical condition or treatment that could influence appetite, weight, or ability to participate; current pregnancy or nursing or planning to become pregnant within the next year; and planning to move >2 hours from the study site within the next year. With the exception of a stable dose (i.e., ≥8 weeks) of SSRIs or hormonal contraceptives, participants were instructed to refrain from taking any medication, including prescription or over the counter medications, for 72 hours prior to visits that involved food intake2. Thus, participants who were prescribed medications on a pro re nata (PRN) basis and did not require medication at the time of participation were included (e.g., seasonal allergy medications).

All participants identified as cisgender female. Most of the sample completed partial college (87.2%), 9.2% completed partial graduate school, 1.7% had a two-year degree, 1.2% had a four-year degree, and 0.6% completed graduate or professional school. Participants’ racial identities were 76.2% White, 11.0% African American or Black, 0.6% American Indian/Alaska Native, 1.2% Pacific Islander or Native Hawaiian, 3.5% Asian, and 7.6% endorsed more than one race; 27.3% of participants were Hispanic or LatinX.

Measures

Eating Disorder Symptoms and Severity.

The Eating Disorder Examination (EDE; Fairburn et al., 2014) established eating disorder diagnosis, severity, and symptom frequency. Interviews were audio recorded with participant consent, and 15% were randomly selected for assessment of interrater reliability. Internal consistency of the subscales and global score as continuous measures of severity among those with BN-S was very good (α = .99). Interrater reliability for frequency of loss of control eating episodes and compensatory behaviors was excellent (Pearson’s correlation coefficients between .91-.96). Interrater reliability for binge episode size was good (Intraclass correlation coefficient = .82).

History of Sexual Assault and Comorbid Diagnoses.

History of SA and PTSD and the presence of current and lifetime mental disorders was established using the Structured Clinical Interview for DSM-5 Research Version (SCID-5; First, 2015). Interrater reliability among interviewers on lifetime and current diagnoses showed good to very good agreement (K = .83-.98). We also collected data on first age of SA (Mage=12.57 years, SD=5.52), whether the perpetrator of the SA was a relative (60.8%), and whether the SA was repeated or a single event (60.8% repeated).

Visual Analog Scale (VAS) Scores.

Immediately before and after food consumption during the behavioral food tasks, described below, participants completed a series of VAS questions assessing negative affect and preoccupation with shape and weight, as well as other momentary states. Internal consistency of these composites was good to excellent (α = .86-.96).

Participants completed a progressive ratio (PR) task (Bodell & Keel, 2015) to earn 10 M&Ms after each trial. Participants could complete up to 10 trials of the task, earning up to 100 M&Ms. The number of key presses required to complete a trial began at 50 and increased by 200 in each subsequent trial. Participants were instructed to work for as many M&Ms as they wanted and to consume all 10 M&Ms before proceeding to the next trial. Breakpoint (i.e., number of key presses in the last completed trial) reflects food reward valuation. Breakpoint on this task demonstrates good convergent validity with breakpoint on a non-food progressive ratio task (i.e., r = .51) and good test-retest reliability over a one to two week period (ICC = .85; Keel et al., 2022). 130 participants (76% of the current sample) completed this task in the present study due to protocol changes to adapt to COVID-19.

Participants completed an ad lib test meal of frozen yogurt in the afternoon (Keel et al., 2018). Prior to the ad lib meal, participants consumed a fixed liquid meal for breakfast and refrained from eating or drinking anything other than water and refrained from inappropriate compensatory behaviors. For the ad lib meal, participants were served 1.5 quarts of vanilla frozen yogurt in a bowl with instructions to eat until full. Yogurt was weighed twice before and after the meal using an electronic balance. Total intake of frozen yogurt consumed reflects food reward satiation. The ad lib test meal is a reliable and valid method to measure satiation (Sysko et al., 2018). All participants included in analyses completed this task.

For the purposes of the current study, both tasks were used as a means to measure changes in affect and body image after food consumption. Although groups did not differ on calories consumed in each task, calories consumed in the ad lib test meal are significantly higher than calories consumed in the PR task (t(129)=9.655, p<.001. d=.84).

Procedure

Participants completed five study visits over the course of three days. All procedures were approved by Florida State University’s Institutional Review Board (#00000353). Informed consent was obtained from all participants. At the first visit, participants completed the EDE and SCID-5. During the third visit, in the afternoon of the second day, participants completed the food reward valuation (PR) task. During the fifth visit, in the afternoon of the third day, participants completed the food reward satiation (ad lib test meal) task (Keel et al., 2019).

Statistical Analysis

All analyses were conducted using SPSS 28.0. To compare eating disorder severity across groups, we ran a series of one-way ANOVA models for variables that were normally distributed or Kruskal-Wallis tests for variables that did not fit a normal distribution after attempting transformation (i.e., OBE episodes, SBE episodes, purging episodes, days of fasting, and days of driven exercise). We used Chi-square tests of independence to compare prevalence of comorbid disorders between the three groups of participants (i.e., PTSD + SA, PTSD − SA, No PTSD/No SA).

Prior to examining whether groups differed in their responses to food intake, we examined whether groups differed on breakpoint for M&Ms in the PR task (F(2, 127)=.50, p=.61, η2 =.01) or consumption of frozen yogurt in the ad lib meal (F(2, 169)=.13, p=.88, ηp2<.01). The lack of significant or clinically meaningful differences in food consumption among groups indicates that any differences in affective or cognitive responses to food intake are not confounded by amount of food consumed.

Next, we ran a series of repeated-measures ANOVA models with the three groups (i.e., PTSD + SA, PTSD − SA, No PTSD/No SA) as the between-subjects variable and timepoint (i.e., pre- and post- task) as the within-subjects variable to examine whether the PTSD + SA experienced greater changes in negative affect or shape/weight preoccupation following food intake in each behavioral task (interaction effect of group X time) relative to the PTSD − SA and No PTSD/No SA groups. Negative affect and shape/weight preoccupation were square root transformed to achieve a normal distribution. We used Tukey’s HSD post-hoc tests to control for Type I error when probing significant main effects of group. For any significant group X time interactions, we determined whether there was a pre- to post-task difference within each group to understand changes in negative affect or shape/weight preoccupation.

A post-hoc power analysis with an alpha of .05 and power of .80 indicated that we were powered to detect a medium to large effect size (d=.70) between the smallest (i.e., PTSD − SA, n = 19) and largest group (i.e., No PTSD/No SA, n = 107) in the ad lib test meal. We were powered to detect a large effect size (d=.75) between the smallest (i.e., PTSD − SA, n = 16) and largest group (i.e., No PTSD/No SA, n = 81) in the PR task.

Results

Clinical Features

Consistent with prior literature, participants with PTSD + SA experienced significantly greater eating disorder pathology than the No PTSD/No SA group (Table 1). The PTSD + SA group endorsed a significantly greater total number of loss of control eating episodes than the PTSD − SA group and the No PTSD/No SA group, and larger binge episodes, more frequent purging, and driven exercise than the No PTSD/No SA group (all p’s < .05). Finally, the PTSD − SA group largely reported intermediate levels of eating pathology and eating disorder symptom frequencies that did not differ from either the PTSD + SA or the No PTSD/No SA groups with the exception of purging, which was significantly greater than in the No PTSD/No SA group (all p’s < .05).

Table 1.

Eating disorder symptoms among the three study groups.

Sexual Assault History, PTSD History, and Eating Disorder Symptoms
PTSD + SA (n=46) M(SE) PTSD - SA (n=19) M(SE) No PTSD/No SA (n=107) M(SE) F (2, 169) ηp2
BMI 25.83 (.74) 26.21 (1.07) 24.62 (.42) 1.75 .02
EDE Global Score 3.56 (.13)a 3.17 (.24) 3.00 (.08)b 6.53** .07
EDE Restraint 3.73 (.17)a 3.49 (.37) 3.10 (.12)b 4.37* .05
EDE Eating Concern 2.26 (.20)a 1.69 (.30) 1.75 (.10)b 3.22* .04
EDE Weight Concern 4.03 (.16)a 3.55 (.23) 3.51 (.12)b 3.36* .04
EDE Shape Concern 4.21 (.12)a 3.93 (.23) 3.62 (.10)b 5.96** .07
OBE kcals 3626.29 (335.83)a 3347.05 (364.33) 2773.61 (116.13)b 5.07** .06
Total LOC Episodes 62.52 (8.53)a 31.32 (4.65)b 37.69 (3.12)b 5.25** .06
Mean Rank H (2) η2
OBE Episodes 92.60 67.13 87.32 3.60 .01
SBE Episodes 100.64 88.71 80.03 5.79 .02
Purging Episodes 107.49a 104.50a 74.28b 19.45*** .10
Fasting Days 105.38a 81.39 79.29b 9.13* .04
Driven Exercise Days 95.59 81.37 83.50 2.13 <.01
Open in a new tab

Note: We reported the untransformed mean and standard error for OBE kcals and Total LOC Episodes for descriptive purposes.

Kruskal-Wallis H tests were run for total OBE Episodes, SBE Episodes, Purging Episodes, Days of Fasting, and Days of Driven Exercise. Follow-up Mann-Whitney U tests were run to examine group differences if the omnibus test was significant.

Within each row, means with different superscripts are significantly different.

*

p<.05

**

p<.01

***

p<.001

The PTSD + SA group demonstrated a higher prevalence of lifetime and current mood disorders, substance use disorders, and anxiety disorders than the No PTSD/No SA group (Table 2). Similar to patterns for eating pathology, comorbidity in the PTSD − SA group was intermediate and did not differ significantly from either the PTSD + SA or No PTSD/No SA group prevalence estimates.

Table 2.

Prevalence of SCID-5 diagnoses among the three study groups.

Sexual Assault History, PTSD History, and Comorbidity
DSM-5 Diagnosis PTSD + SA (n=46) n (%) PTSD − SA (n=19) n (%) No PTSD/ No SA (n=107) n (%) X2(2)
Lifetime Mood Disorder 42 (91.3%)a 17 (89.5%) 75 (70.1%)b 10.07**
Current Mood Disorder 27 (58.7%)a 7 (36.8%) 28 (26.2%)b 14.79***
Lifetime Substance Use Disorder 36 (78.3%)a 12 (63.2%) 45 (42.06%)b 19.10***
Current Substance Use Disorder 24 (52.2%)a 7 (36.8%) 31 (29.0%)b 7.70*
Lifetime Anxiety Disorder 39 (84.8%)a 15 (78.9%) 60 (56.1%)b 14.79***
Current Anxiety Disorder 39 (84.8%)a 13 (68.4%) 52 (48.6%)b 18.79***
Open in a new tab

Note: Lifetime & Current mood disorders include Bipolar I, Bipolar II, Substance/Medically Induced Bipolar, Major Depressive Disorder, Persistent Depressive Disorder, Other Specified Depressive Disorder, and Substance/Medically-Induced Depressive Disorder. Lifetime & Current Substance Use Disorders include Alcohol Use Disorder, Sedative Hypnotic Use Disorder, Cocaine Use Disorder, Stimulant/Cocaine Use Disorder, Opioid Use Disorder, and Other Substance Use Disorder. Lifetime & Current anxiety disorders include Panic Disorder, Agoraphobia, Social Anxiety Disorder, specific phobias, Generalized Anxiety Disorder, and Other Specified Anxiety Disorder.

Within each row, means with different superscripts are significantly different.

*

p<.05

**

p<.01

***

p<.001

Food Intake Behavioral Tasks

Across all four models, there was a significant main effect of group that did not reflect a response to food intake (Table 3). For the PR task and ad lib test meal; the PTSD + SA group experienced higher negative affect throughout the task compared to individuals in the No PTSD/No SA group (Table 4). For the PR task, the PTSD + SA group had higher shape and weight preoccupation throughout the task compared to the PTSD − SA and No PTSD/No SA groups. Finally, the PTSD + SA group had higher shape and weight preoccupation throughout the ad lib test meal compared to the No PTSD/No SA group (Table 4). No other significant differences between groups were found.

Table 3.

Group differences in subjective responses to food intake in two behavioral tasks.

Subjective Responses to Food Intake
VAS Task SA History Pre – M(SE) Post – M(SE) Group Time Group x Time
F (df) ηp2 F (df) ηp2 F (df) ηp2
Negative Affect PR PTSD + SA 40.58 (4.14) 45.10 (4.24) 7.45 (2, 127) *** .11 4.74 (2, 127) * .04 3.45 (2, 127) * .05
PTSD − SA 36.15 (5.94) 24.17 (6.09)
No PTSD/No SA 25.14 (2.64) 22.36 (2.71)
Ad lib PTSD + SA 39.06 (3.74) 55.85 (4.33) 6.31 (2, 169) ** .07 20.38 (2, 169) *** .11 1.01 (2, 169) .01
PTSD − SA 27.95 (5.82) 35.32 (6.74)
No PTSD/ No SA 27.28 (2.45) 35.91 (2.84)
Shape/Weight Preoccupation PR PTSD + SA 63.23 (4.51) 72.03 (5.01) 5.52 (2, 127) ** .08 3.47 (2, 127) .03 .36 (2, 127) .01
PTSD − SA 45.31 (6.47) 50.38 (7.20)
No PTSD/ No SA 44.76 (2.88) 50.96 (3.20)
Ad lib PTSD + SA 62.38 (4.15) 80.64 (3.82) 3.85 (2, 169) * .04 51.74 (2, 169) *** .23 .05 (2, 169) <.01
PTSD − SA 49.45 (6.45) 67.37 (5.95)
No PTSD/ No SA 49.55 (2.72) 67.33 (2.51)
Open in a new tab

Note: All Visual Analogue Scale (VAS) score composites were square root transformed to achieve a normal distribution. We reported the untransformed mean and standard error and transformed test statistic, degrees of freedom, and effect size.

PR=Progressive Ratio Food Reward Valuation task, Ad lib=Food Reward Satiation task.

*

p<.05,

**

p<.01,

***

p<.001

Table 4.

Multiple comparisons for main effect of group on mood and body image.

Main Effect of Group on Mood and Body Image
VAS Task (I) (J) Mean Difference (I-J) (SE) d
Negative Affect PR PTSD + SA PTSD − SA 12.68 (6.73) .50
No PTSD/No SA 19.09 (4.56) * .71
PTSD − SA No PTSD/No SA 6.41 (6.04) .27
Ad lib PTSD + SA PTSD − SA 15.82 (6.76) .32
No PTSD/No SA 15.85 (4.37) * .35
PTSD − SA No PTSD/No SA .03 (6.17) <.01
Shape/Weight Preoccupation PR PTSD + SA PTSD − SA 19.79 (7.77) * .73
No PTSD/No SA 19.77 (5.27) * .69
PTSD − SA No PTSD/No SA −.01 (6.97) .04
Ad lib PTSD + SA PTSD − SA 13.10 (6.73) .50
No PTSD/No SA 13.07 (4.35) * .49
PTSD − SA No PTSD/No SA −.03 (6.15) <.01
Open in a new tab

Note: We reported the untransformed mean and standard error and transformed test statistic.

PR = Progressive Ratio Food Reward Valuation Task, Ad lib = Food Reward Satiation Task.

*

p<.05 using Tukey’s HSD post-hoc correction.

Negative Affect Response to Food Intake.

There was a significant group X time interaction for negative affect during the PR task (Table 3). Neither the PTSD + SA group nor the No PTSD/No SA group experienced significant changes in negative affect from pre- to post-task (p’s > .05). The PTSD − SA group experienced a significant decrease in negative affect from pre- to post-task (Mean (SE) difference = 11.98 (4.76), p=.008, d=.54; Figure 1), not observed in the other groups. Post-hoc comparisons supported that the change in affect in the PTSD − SA group differed significantly from change in the PTSD + SA group (F(1, 49) = 5.49, p=.02, ηp2 = .11) but not the No PTSD/No SA group (F(1, 95)=3.74, p=.06, ηp2 = .04), who also did not differ from the PTSD + SA group (F(1, 112) = 2.06, p=.15, ηp2 = .02). Pre-task, participants in the PTSD + SA group reported greater negative affect than the No PTSD/No SA group (Mean (SE) difference = 15.44 (4.91), p=.01, d=.63), but not the PTSD − SA group (Mean (SE) difference = 4.43 (7.24), p=1.00, d=.19). Similarly, post-task, participants in the PTSD + SA group reported greater negative affect than the No PTSD/No SA group (Mean (SE) difference = 22.74 (5.03), p<.001, d=.86), but not the PTSD − SA group (Mean (SE) difference = 20.93 (7.42), p=.08, d=.78). Finally, there was a main effect of time on negative affect for the PR task and the ad lib test meal, supporting an overall increase in this sample of women with BN-S.

Figure 1.

Figure 1.

Open in a new tab

Group x Time interaction for negative affect in the PR task.

Shape and Weight Preoccupation Response to Food Intake.

There was a main effect of time for the ad lib test meal; shape and weight preoccupation increased significantly following food intake regardless of group membership.

Discussion

The present study examined the role of SA history and PTSD on affective and cognitive responses to food intake during two behavioral tasks. To our knowledge, this was the first study to focus on PTSD + SA history and examine these factors on a momentary level among individuals with current BN-S. Results replicated prior findings that participants with PTSD + SA history had the greatest eating disorder severity and psychiatric comorbidity of the three groups (Dworkin, 2020; Kjaersdam Telléus et al., 2021; Malet-Karas et al., 2022). Purging was more frequent in both PTSD groups, with more frequent of loss of control eating in the PTSD + SA group. In addition, there was a significant group X time interaction in the PR task. Participants in the PTSD − SA group experienced a significant decrease in negative affect from pre- to post-task, whereas negative affect did not significantly change from pre- to post-task for the PTSD + SA and No PTSD/No SA groups. Negative affect was elevated throughout the ad lib test meal for participants in the PTSD + SA group relative to participants in the No PTSD/No SA group and negative affect increased throughout the task in all groups. Shape and weight preoccupation was also elevated for the PTSD + SA group relative to the No PTSD/No SA group for both the PR task and the ad lib test meal and increased throughout the task in all groups. Taken together, results suggest key differences in affective and cognitive responses to food intake between those with PTSD + SA and those with PTSD − SA.

Somewhat surprisingly, the PTSD + SA and PTSD − SA groups did not differ in frequency of purging behaviors, although both groups endorsed purging more frequently than the No PTSD/No SA group; this finding is partially consistent with past research which indicates a robust association between SA and purging (Ackard & Neumark-Sztainer, 2002; Dansky et al., 1997; Fonseca et al., 2002; Groff Stephens & Wilke, 2016; Hazzard et al., 2022; Lie et al., 2021; Molendijk et al., 2017; Perkins & Luster, 1999). Also consistent with prior research, the PTSD + SA group had a higher prevalence of current and lifetime mood, substance use, and anxiety disorders than the No PTSD/No SA group (Dworkin, 2020; Malet-Karas et al., 2022). During case conceptualization, therapists work with patients to understand how comorbid conditions are likely to exacerbate each other. Comorbidity may contribute to elevated negative affect prior to food intake among individuals with PTSD + SA history (Brewerton, 2007).

Participants in the PTSD − SA group reported a significant decrease in negative affect following food intake in the PR task in which they were given the opportunity to consume a normative amount of food. This finding contrasts with prior work on emotional responses to food intake among women with BN and PTSD (Karr et al., 2013). This suggests that some forms of food intake may function to regulate negative affect in those with PTSD with no SA history. This may impact how therapists interpret the function of eating as well as account for differences in experiences of loss of control eating between those with PTSD due to SA versus other forms of trauma. This finding should be interpreted with caution, given the small number of participants who comprised this group for the PR task analyses (n = 16). However, this finding provides evidence that those with PTSD + SA and those with PTSD − SA have different clinical presentations in terms of responses to food intake and may warrant different treatment targets.

Findings regarding main effects of group in the present study replicated and extended previous work on momentary responses to food intake among individuals with BN and PTSD (Karr et al., 2013). Karr and colleagues (2013) found that negative affect was elevated before and after binge eating episodes; we found that PTSD + SA history was associated with elevated negative affect before the opportunity to consume both a normative and objectively large amount of food. Greater negative affect may lead to a greater desire to decrease negative affect, and purging may function to alleviate negative affect (Haedt-Matt & Keel, 2015). Although the PTSD + SA and PTSD − SA groups did not differ from each other on purging frequency, both groups reported greater frequencies of purging than the No PTSD/No SA group. It is possible that purging may be associated with PTSD in general, however, given the small sample size of the PTSD − SA group, future research should seek to replicate this finding. The absence of a main effect of time for negative affect in the PR task may reflect the smaller amount of food available and consumed during this task compared to the ad lib test meal. The availability and consumption of larger amounts of food may be particularly distressing for individuals with BN-S, irrespective of SA history.

The present study had several strengths, including the use of well-validated clinical interviews with good interrater reliability to establish PTSD and SA history, eating disorder pathology, and comorbidity. Reliable and valid laboratory-based behavioral measures of food intake (Bodell & Keel, 2015; Keel et al., 2018) allowed us to examine affective and cognitive responses to normative amounts of food and an ad lib test meal. However, our findings should be interpreted within the context of limitations. To maximize sample size for analyses, the present study combined all forms of SA into a single category without differentiating based on age, recurrence of events, or type of SA. Individuals who experience sexual trauma in childhood and adulthood are more likely to exhibit eating disorder symptoms than those who only experience sexual trauma in childhood (Wonderlich et al., 2001). Future research should examine whether reactions to food intake are influenced by the number of sexual traumas and the age at which they were experienced.

Overall, findings suggest that women with a current BN-S and PTSA + SA had higher levels of negative affect and shape/weight preoccupation prior to tasks that involved food consumption. Women with PTSD + SA may benefit from increased focus on body image earlier in treatment, as this may represent a more trait-like factor and could account for noncompensatory purging (Liebman et al., 2020), that is, purging that occurs in the absence of food consumption. These women also endorsed a greater prevalence of mood, anxiety, and substance use disorders. Cognitive Behavior Therapy (CBT) for eating disorders recommends treating comorbid depression before treating the eating disorder (Fairburn, 2008). Results of the present study suggest that trauma-focused treatment for individuals with sub- and full-threshold PTSD may be particularly beneficial for individuals with BN-S and PTSD (Brewerton, 2007). Clinicians can work with patients to identify maintaining mechanisms for both the eating disorder and the PTSD (e.g., purging as a means to avoid traumatic memories and alleviate hyperarousal; Brewerton, 2007). Principles of CBT can be adapted across eating disorders, sub- and full-threshold PTSD, mood, and anxiety disorders to treat the whole person, rather than a single disorder. Adapting current evidence-based treatments may help women with current BN-S and PTSD + SA history regain integrity of their bodies, enhance compliance with prescription of a regular pattern of eating, and alleviate symptoms.

Clinical Implications.

  • PTSD + SA is associated with more loss of control eating.

  • PTSD regardless of SA history is associated with more purging.

  • PTSD − SA is associated with decreased negative affect (NA) following food intake.

  • PTSD + SA is associated with NA and body image concern before and after food intake.

  • Trauma-focused treatment may help target purging for patients with PTSD.

Funding Details:

This research was supported by the National Institute of Mental Health (R01MH111263).

Footnotes

Disclosure Statement: The authors report there are no competing interests to declare.

2

Participants taking hormonal contraceptives did not differ from participants not taking hormonal contraceptives on breakpoint in the PR task (t(124.84)=−.62, p=.54, d=.11) or grams of frozen yogurt consumed in the Ad lib test meal (t(166.96)=−1.03, p=.30, d=.15). Participants taking SSRIs did not differ from participants not taking SSRIs on breakpoint in the PR task (t(122)=.13, p=.90, d=.04) or grams of frozen yogurt consumed in the Ad lib test meal (t(164)=.12, p=.91, d=.03).

Data availability:

Data are available through the NIMH Data Archive (C2451).

References

  1. Ackard DM, & Neumark-Sztainer D (2002). Multiple Sexual Victimizations Among Adolescent Boys and Girls: Prevalence and Associations with Eating Behaviors and Psychological Health. Journal of Child Sexual Abuse, 12(1), 17–37. 10.1300/J070v12n01_02 [DOI] [PubMed] [Google Scholar]
  2. American Psychiatric Association (Ed.). (2013). Diagnostic and statistical manual of mental disorders: DSM-5 (5th ed). American Psychiatric Association. [Google Scholar]
  3. Amstadter AB, & Vernon LL (2008). Emotional Reactions During and After Trauma: A Comparison of Trauma Types. Journal of Aggression, Maltreatment & Trauma, 16(4), 391–408. 10.1080/10926770801926492 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Backholm K, Isomaa R, & Birgegård A (2013). The prevalence and impact of trauma history in eating disorder patients. European Journal of Psychotraumatology, 4(1), 22482. 10.3402/ejpt.v4i0.22482 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bodell LP, & Keel PK (2015). Weight suppression in bulimia nervosa: Associations with biology and behavior. Journal of Abnormal Psychology, 124(4), 994–1002. 10.1037/abn0000077 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Breland JY, Donalson R, Li Y, Hebenstreit CL, Goldstein LA, & Maguen S (2018). Military sexual trauma is associated with eating disorders, while combat exposure is not. Psychological Trauma: Theory, Research, Practice, and Policy, 10(3), 276–281. 10.1037/tra0000276 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brewerton TD (2007). Eating Disorders, Trauma, and Comorbidity: Focus on PTSD. Eating Disorders, 15(4), 285–304. 10.1080/10640260701454311 [DOI] [PubMed] [Google Scholar]
  8. Dansky BS, Brewerton TD, Kilpatrick DG, & O’Neil PM (1997). The National Women’s Study: Relationship of victimization and posttraumatic stress disorder to bulimia nervosa. International Journal of Eating Disorders, 21(3), 213–228. [DOI] [PubMed] [Google Scholar]
  9. Dworkin ER (2020). Risk for Mental Disorders Associated With Sexual Assault: A Meta-Analysis. Trauma, Violence, & Abuse, 21(5), 1011–1028. 10.1177/1524838018813198 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Fairburn C, Cooper Z, & O’Connor M (2014). Eating Disorder Examination [dataset]. American Psychological Association. 10.1037/t03975-000 [DOI] [Google Scholar]
  11. Fairburn CG (2008). Cognitive behavior therapy and eating disorders. Cognitive Behavior Therapy and Eating Disorders. [Google Scholar]
  12. Fairburn CG, Cooper Z, & Shafran R (2003). Cognitive behaviour therapy for eating disorders: A “transdiagnostic” theory and treatment. Behaviour Research and Therapy, 41(5), 509–528. 10.1016/S0005-7967(02)00088-8 [DOI] [PubMed] [Google Scholar]
  13. Ferrell EL, Russin SE, & Flint DD (2022). Prevalence Estimates of Comorbid Eating Disorders and Posttraumatic Stress Disorder: A Quantitative Synthesis. Journal of Aggression, Maltreatment & Trauma, 31(2), 264–282. 10.1080/10926771.2020.1832168 [DOI] [Google Scholar]
  14. First MB (2015). Structured Clinical Interview for the DSM (SCID). In The Encyclopedia of Clinical Psychology (pp. 1–6). John Wiley & Sons, Ltd. 10.1002/9781118625392.wbecp351 [DOI] [Google Scholar]
  15. Fonseca H, Ireland M, & Resnick MD (2002). Familial correlates of extreme weight control behaviors among adolescents. International Journal of Eating Disorders, 32(4), 441–448. 10.1002/eat.10078 [DOI] [PubMed] [Google Scholar]
  16. Gomez F, Kilpela LS, Middlemass KM, & Becker CB (2021). Sexual trauma uniquely associated with eating disorders: A replication study. Psychological Trauma: Theory, Research, Practice, and Policy, 13(2), 202–205. 10.1037/tra0000586 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Groff Stephens S, & Wilke DJ (2016). Sexual violence, weight perception, and eating disorder indicators in college females. Journal of American College Health, 64(1), 38–47. 10.1080/07448481.2015.1074237 [DOI] [PubMed] [Google Scholar]
  18. Haedt-Matt AA, & Keel PK (2015). Affect Regulation and Purging: An Ecological Momentary Assessment Study in Purging Disorder. Journal of Abnormal Psychology, 124(2), 399–411. 10.1037/a0038815 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hazzard VM, Ziobrowski HN, Borg SL, Schaefer LM, Mangold AC, Herting NA, Lipson SK, Crosby RD, & Wonderlich SA (2022). Past-Year Abuse and Eating Disorder Symptoms Among U.S. College Students. Journal of Interpersonal Violence, 37(15–16), NP13226–NP13244. 10.1177/08862605211005156 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Jaconis M, Boyd SJ, & Gray MJ (2020). History of Sexual Violence and Associated Negative Consequences: The Mediating Role of Body Image Dissatisfaction. Journal of Loss and Trauma, 25(2), 107–123. 10.1080/15325024.2019.1660500 [DOI] [Google Scholar]
  21. Karr TM, Crosby RD, Cao L, Engel SG, Mitchell JE, Simonich H, & Wonderlich SA (2013). Posttraumatic stress disorder as a moderator of the association between negative affect and bulimic symptoms: An ecological momentary assessment study. Comprehensive Psychiatry, 54(1), 61–69. 10.1016/j.comppsych.2012.05.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Keel PK, Bodell LP, Forney KJ, Appelbaum J, & Williams D (2019). Examining weight suppression as a transdiagnostic factor influencing illness trajectory in bulimic eating disorders. Physiology & Behavior, 208, 112565. 10.1016/j.physbeh.2019.112565 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Keel PK, Haedt A, & Edler C (2005). Purging disorder: An ominous variant of bulimia nervosa? International Journal of Eating Disorders, 38(3), 191–199. 10.1002/eat.20179 [DOI] [PubMed] [Google Scholar]
  24. Keel PK, Haedt-Matt AA, Hildebrandt B, Bodell LP, Wolfe BE, & Jimerson DC (2018). Satiation deficits and binge eating: Probing differences between bulimia nervosa and purging disorder using an ad lib test meal. Appetite, 127, 119–125. 10.1016/j.appet.2018.04.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Keel PK, Kennedy GA, Rogers ML, Joyner KJ, Bodell LP, Forney KJ, & Duffy ME (2022). Reliability and validity of a transdiagnostic measure of reward valuation effort. Psychological Assessment, 34(5), 419–430. 10.1037/pas0001107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kjaersdam Telléus G, Lauritsen MB, & Rodrigo-Domingo M (2021). Prevalence of Various Traumatic Events Including Sexual Trauma in a Clinical Sample of Patients With an Eating Disorder. Frontiers in Psychology, 12, 687452. 10.3389/fpsyg.2021.687452 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lie SØ, Bulik CM, Andreassen OA, Rø Ø, & Bang L (2021). Stressful life events among individuals with a history of eating disorders: A case-control comparison. BMC Psychiatry, 21(1), 501. 10.1186/s12888-021-03499-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Liebman RE, Coniglio KA, Becker KR, Tabri N, Keshishian AC, Wons O, Keel PK, Eddy KT, & Thomas JJ (2020). Behavioral inhibition moderates the association between overvaluation of shape and weight and noncompensatory purging in eating disorders. International Journal of Eating Disorders, 53(1), 143–148. 10.1002/eat.23195 [DOI] [PubMed] [Google Scholar]
  29. Madowitz J, Matheson BE, & Liang J (2015). The relationship between eating disorders and sexual trauma. Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity, 20(3), 281–293. 10.1007/s40519-015-0195-y [DOI] [PubMed] [Google Scholar]
  30. Malet-Karas A, Bernard D, Piet E, & Bertin E (2022). Disordered eating as a repercussion of sexual assault: A consequence to consider. Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity, 27(6), 2095–2106. 10.1007/s40519-021-01356-5 [DOI] [PubMed] [Google Scholar]
  31. McKenzie SJ, Williamson DA, & Cubic BA (1993). Stable and reactive body image disturbances in bulimia nervosa. Behavior Therapy, 24(2), 195–207. 10.1016/S0005-7894(05)80263-1 [DOI] [Google Scholar]
  32. Molendijk ML, Hoek HW, Brewerton TD, & Elzinga BM (2017). Childhood maltreatment and eating disorder pathology: A systematic review and dose-response meta-analysis. Psychological Medicine, 47(8), 1402–1416. 10.1017/S0033291716003561 [DOI] [PubMed] [Google Scholar]
  33. Murphy R, Straebler S, Cooper Z, & Fairburn CG (2010). Cognitive behavioral therapy for eating disorders. The Psychiatric Clinics of North America, 33(3), 611–627. 10.1016/j.psc.2010.04.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Perkins DF, & Luster T (1999). The relationship between sexual abuse and purging: Findings from community-wide surveys of female adolescents. Child Abuse & Neglect, 23(4), 371–382. 10.1016/S0145-2134(99)00009-5 [DOI] [PubMed] [Google Scholar]
  35. Preti A, Incani E, Camboni MV, Petretto DR, & Masala C (2006). Sexual abuse and eating disorder symptoms: The mediator role of bodily dissatisfaction. Comprehensive Psychiatry, 47(6), 475–481. 10.1016/j.comppsych.2006.03.004 [DOI] [PubMed] [Google Scholar]
  36. Sack M, Boroske-Leiner K, & Lahmann C (2010). Association of nonsexual and sexual traumatizations with body image and psychosomatic symptoms in psychosomatic outpatients. General Hospital Psychiatry, 32(3), 315–320. 10.1016/j.genhosppsych.2010.01.002 [DOI] [PubMed] [Google Scholar]
  37. Sysko R, Steinglass J, Schebendach J, Mayer LES, & Walsh BT (2018). Rigor and reproducibility via laboratory studies of eating behavior: A focused update and conceptual review. International Journal of Eating Disorders, 51(7), 608–616. 10.1002/eat.22900 [DOI] [PubMed] [Google Scholar]
  38. Vocks S, Legenbauer T, & Heil A (2007). Food intake affects state body image: Impact of restrained eating patterns and concerns about eating, weight and shape. Appetite, 49(2), 467–475. 10.1016/j.appet.2007.03.006 [DOI] [PubMed] [Google Scholar]
  39. Wonderlich SA, Crosby RD, Mitchell JE, Thompson KM, Redlin J, Demuth G, Smyth J, & Haseltine B (2001). Eating disturbance and sexual trauma in childhood and adulthood. International Journal of Eating Disorders, 30(4), 401–412. 10.1002/eat.1101 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data are available through the NIMH Data Archive (C2451).

RESOURCES