Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Oct 1.
Published in final edited form as: J Am Coll Health. 2022 Jul 26;72(7):2014–2021. doi: 10.1080/07448481.2022.2100707

Sexual assault predicts unhealthy weight management among college women: A longitudinal, prospective study

Larissa A McGarrity 1, Robyn L Shepardson 2, Kate B Carey 3, Michael P Carey 4
PMCID: PMC9877249  NIHMSID: NIHMS1836319  PMID: 35882060

Abstract

Objective:

To examine whether sexual assault prospectively predicts unhealthy weight management behaviors in college women.

Method:

Participants were female college students (N=483) with monthly assessments across the first year, including frequency and severity of sexual assault and unhealthy weight management behaviors.

Results:

Frequency of sexual assault prior to college predicted dieting, purging, and diet pill use, over the first year. Severity of those experiences predicted dieting and diet pill use. Frequency of sexual assault during first semester predicted purging and diet pill use during second semester. Severity predicted all three unhealthy weight management behaviors. Findings held consistently when controlling for SES, race/ethnicity, BMI, and pre-college mood, anxiety, and eating disorders, as well as unhealthy weight management behaviors during the first semester in longitudinal analyses. Reverse models were non-significant.

Conclusions:

This research documents detrimental effects of sexual assault on unhealthy weight management behaviors in college women, and highlights the importance of prevention and intervention.

Keywords: sexual assault, weight management, dieting, purging, college women

Prior studies have documented a consistent association between sexual assault and a range of negative consequences during emerging adulthood, including mental health conditions and health risk behaviors.1,2 The majority of this research has focused on associations with alcohol and substance misuse and sexual risk behaviors.3,4 As part of the efforts to elucidate the varied physical and mental health impacts of traumatic experiences on women, researchers have also begun to investigate how sexual assault uniquely influences unhealthy weight management and eating behaviors,5,6 given the high prevalence of both among young women.

Rates of sexual trauma are disturbingly high for women in emerging adulthood. Approximately 28% of college women report a history of either incapacitated or forced rape by the time they enter college.7 When a broader definition of sexual assault is considered (including unwanted sexual contact, attempted rape, and sexual coercion), 39% of women report these experiences by the time they reach college.2 A recent systematic review by Muehlenhard et al8 found that 20% of women endorse sexual assault during the college years, with the first year as a particularly high-risk period. Scholars have suggested this transition year combines exposure to high-risk environments, few safeguards for protection or support, and perpetrator targeting of young women.9

Similarly, unhealthy weight management behaviors are commonplace among young women. Peak onset of these maladaptive eating behaviors typically occurs between the ages of 16–20, when many women are entering college.10,11 Unhealthy weight management behaviors, particularly dieting, are important early indicators of the development of disordered eating.12 Previous research has documented that nearly 60% of female adolescents and young adults report restrictive dieting within the past year, and the prevalence of other unhealthy weight management behaviors (e.g., purging or diet pills) increases from around 8% to 20% in the transition from adolescence to young adulthood.13 The high prevalence of these behaviors is especially striking given that the majority of female adolescents and young adults fall into the typical, rather than overweight or obesity, body mass index range.14 Regardless of an individual’s weight status or reason for dieting, accumulating evidence suggests this behavior is largely ineffective long-term and associated with health risks.15 Potential contributing factors to increased risk of unhealthy weight management behaviors in the transition to college include “environmental pressures to conform to peer groups, heightened exposure to mass media which exacerbates internalization of the thin ideal, and increased pressure to be sexually attractive” in the first year of college.16

Large-scale, population-based studies have demonstrated consistent cross-sectional associations between sexual assault and maladaptive weight management, particularly among adolescent and young adult women.5,6 These studies have generally shown that individuals with a history of sexual assault report higher utilization of vomiting, laxative use, or diet pill use as weight management strategies. Multiple theories address the unique association between sexual assault and unhealthy weight management, with mechanisms including an attempt to exert or regain control over one’s life and body,17 emotion dysregulation and attempt at coping through health risk behaviors,3,18 internalization of negative societal messages about the female body with an associated obsession or preoccupation with body image,19,20 and self-harm as “punishment”5.

Two relevant longitudinal studies on college women have further elucidated this relationship over time.2,15 The first study focused on predictors of assault and re-assault and showed that vomiting and laxative use at baseline were associated with experiencing sexual assault 3 months later, but not with control for pre-college assault.2 Although the study also replicated the cross-sectional association between history of assault and unhealthy weight management behaviors, it left unanswered questions about the longitudinal prediction of unhealthy weight management behaviors during college from experiences of sexual assault. The second study reported a significant association between experienced and attempted rape during the transition to college and development of disordered eating pathology in the first three months.15 Attempts to suppress thoughts about the trauma were also associated with disordered eating. The literature on sexual assault and unhealthy weight management behaviors is growing, yet remains limited by incomplete and widely varying timeframes and measures of sexual assault, retrospective and cross-sectional study designs that limit understanding of effects and causality, and lack of control for potential confounding variables (such as mental health history).

The current study sought to address these methodological limitations in examination of the longitudinal association between sexual assault and unhealthy weight management behaviors in young women. This study built upon the existing literature in the following ways: (a) improved measurement of sexual assault with investigation of both frequency and severity of those experiences, (b) control for demographic covariates, mental health history, and baseline assessments of the outcome (to effectively determine direction of effects and increase confidence in causal relationships), (c) longitudinal, prospective study design with assessments monthly across the first year, allowing for examination of directional relationships, and (d) examination of purging behaviors and diet pill use, but also dieting - a more subtle, but common, behavior in adolescence and emerging adulthood that predicts both eating disorders and obesity over time.21,22 This study was also able to examine the reverse causal association (that unhealthy weight management attempts may precede and predict sexual assault). Reciprocal relationships or bidirectional effects have been documented with regard to alcohol and substance use and sexual assault in other studies.23 Although it is unlikely that the same may be true for weight management attempts, it has been proposed that overall health behavior risk profiles may increase vulnerability for assault or re-assault.2

In a prospective analysis of female college students, we tested the following hypotheses: (a) Higher frequency of pre-college sexual assault are associated with higher likelihood of unhealthy weight management behaviors in the first year of college. (b) Higher severity of pre-college sexual assault predicts higher likelihood of unhealthy weight management behaviors in the first year of college. (c) Higher frequency of sexual assault during first semester of college predicts higher likelihood of unhealthy weight management behaviors in the second semester. (d) Higher severity of sexual assault during first semester predicts higher likelihood of unhealthy weight management behaviors in the second semester. (e) First semester unhealthy weight management behaviors do not predict frequency of second semester assault. (f) First semester unhealthy weight management behaviors do not predict severity of second semester assault.

Methods

Participants

Participants were 483 women from a larger prospective study with monthly assessments of numerous health behaviors across the first year of college. The sample has been described in detail in other publications from this dataset.24 All women in the incoming first-year class at a private university in the Northeast United States were invited to participate, and 26% enrolled in the study. Participants attended face-to-face orientation sessions (groups of no more than 12), at which time study staff confirmed eligibility, explained the study in detail (including risks and benefits), and answered questions. Participants provided written informed consent. All study procedures were conducted in compliance with the host university’s Institutional Review Board. Participants received either compensation ($10–20 per survey, depending on length) or course credit.

Measures

Sexual Assault

A 20-item measure adapted by Testa et al25 from the Sexual Experiences Survey26 assessed participant experiences of sexual assault since the age of 14 prior to college (at baseline), as well as across the first year (at the end of each semester). Items measured both perpetrator tactics (including use of physical force, arguments or pressure, threats, incapacitation) and types of contact (including fondling/kissing/touching, oral sex, attempted penetration, and penetration). Participants indicated the number of times that each sexual assault experience occurred during each time frame (since age 14 prior to college, Fall semester, Spring semester), utilizing the following scale: 0=0 times, 1=1 time, 2=2 times, 3=3 times, 4=4 or more times. At baseline, instructions read “Indicate the number of times each experience has happened to you since age 14 (from your 14th birthday until today).” Example items at baseline for assessing rape across tactics included: “Has anyone ever (overwhelmed you with arguments about sex or continual pressure for sex)/(threatened to physically harm you or someone close to you)/(used physical force, such as holding you down)/(when you were incapacitated by drugs or alcohol and unable to object or consent) in order to succeed in making you have sexual intercourse when you indicated that you didn’t want to?” For the purpose of this study, variables were created to capture both quantity (number of sexual assault experiences) and severity (0=no sexual assault reported; 1=unwanted sexual contact, including kissing, touching, fondling; 2=unwanted oral sex, or attempted rape/penetration; 3=rape/penetration) of sexual assault experiences at each relevant time point. For severity ratings, participants were categorized by the most severe event reported. Severity of sexual assault categories were based on prior research documenting greater health risks associated with more severe forms of victimization.27

Unhealthy Weight Management

Participants were asked at baseline and then 12 subsequent monthly assessments across the first year about unhealthy weight management behaviors. The question “Over the last month, did you do any of the following in order to lose weight?” was asked, along with single items assessing dieting, vomiting or laxative use, and diet pill use. Answer choices were binary (yes/no). For the purpose of this study, data were aggregated by semester for each weight management behavior, such that participants who responded to all monthly assessments within that semester and denied engaging in that behavior were coded “0” and participants who endorsed engaging in that behavior at any of the time points (regardless of missing data) were coded “1”. Use of single item measures for weight management behaviors is common in the existing literature, with adequate test-retest reliability.13

Covariates

Socioeconomic status (SES) was operationalized as subjective social status of family of origin. Participants were asked the following question: “Please select the dot where you think your family stands at this time in your life, relative to other families in the United States.” and responses ranged from 1–10 with 1 being the lowest rung of the ladder and 10 being the highest rung. Previous research has documented that single item measures of subjective social status are valid as compared to objective SES indicators and useful in studies on mental and physical health.28

Participants self-reported their race/ethnicity. Given the racial/ethnic composition of the sample (with the majority of participants identifying as white), a binary variable was created to identify whether participants identified as white or as a minority racial/ethnic group.

Body mass index (BMI) was calculated based on the self-reported weight and height of participants at baseline. The formula for BMI is weight in kilograms divided by height in meters squared.29 BMI was included as a covariate in all analyses because studies have shown that questions about dieting behavior are interpreted differently based on patient weight categories,30 and due to possible associations between BMI and both sexual trauma and unhealthy weight management behaviors.

Pre-college mood, anxiety, and eating disorders were binary variables based on participant self-report of whether they had been diagnosed with a mood disorder (e.g., depression, bipolar), anxiety disorder (e.g., obsessive compulsive, panic disorder, phobia), or eating disorder (e.g., anorexia, bulimia) prior to college.

Analysis

Data were aggregated by semester and analyzed using logistic, ordinal, and Poisson regression models. Logistic models were used for prediction of binary variables (including presence or absence of purging, diet pill use, and dieting) to examine hypotheses A, B, C, and D. Poisson models were used for prediction of count variables (i.e., amount of times participants experienced sexual assault) to examine hypothesis E. Ordinal models were used for prediction of ordinal variables (i.e., severity of sexual assault) to examine hypothesis F. Covariates included in all models were SES, race/ethnicity, BMI, and pre-college mood, anxiety, and eating disorders. In longitudinal models assessing first semester behaviors predicting second semester behaviors across the first year of college, the outcome variable assessed in the first semester was included as an additional covariate. For example, in models assessing whether sexual assault in the first semester predicted diet pill use in the second semester, control was included for diet pill use in the first semester (consistent with standard longitudinal analyses to better establish direction of effect). Alpha was set a priori at .05. We report adjusted odds ratios to aid in interpretation of logistic regression results. Analytic sample sizes in the present study differ by model (ranging from 219–373) as they include women for whom adequate data were available for the variables in that model. Data were primarily lost due to the way unhealthy weight management behaviors were coded from the monthly assessments (see Measures section for description). The authors felt that it would be inappropriate to make assumptions about participants not having engaged in these behaviors if data were missing at any time point so a conservative approach was taken.

Results

Participant Characteristics

See table 1 for participant demographic characteristics as well as rates of reported sexual assault. The majority of the sample was age 18, identified race/ethnicity as White, and fell into the typical weight range by BMI category. In terms of subjective socioeconomic status of family of origin, the mean rating on the 1–10 scale was 6.23 (SD = 1.67).

Table 1.

Participant demographics and rates of sexual assault.

Percentage
AGE
18 94%
19 5%
20+ 1%
RACE/ETHNICITY
White 64%
Asian 11%
Black 10%
Latino(a) 9%
Biracial or Other 13%
SES RATING
1–3 7%
4–6 44%
7–10 50%
BMI
<18.5 (Underweight) 9%
19.5–24.9 (Typical Weight) 74%
25.–29.9 (Overweight) 14%
30+ (Obesity) 3%
SEXUAL ASSAULT
Unwanted Contact (Pre-College) 57%
Unwanted Contact (First Semester) 36%
Unwanted Contact (Second Semester) 36%
Unwanted Oral Sex or Attempted Intercourse (Pre-College) 49%
Unwanted Oral Sex or Attempted Intercourse (First Semester) 29%
Unwanted Oral Sex or Attempted Intercourse (Second Semester) 21%
Rape (Pre-College) 15%
Rape (First Semester) 9%
Rape (Second Semester) 9%
Open in a new tab

Prior to college, 10% of women reported being diagnosed with mood disorder, 9% an anxiety disorder, and 4% an eating disorder. 65% of women reported a history of at least one experience of sexual assault since the age of 14. Of the total sample, 57% reported history of unwanted contact (i.e., kissing, touching, fondling), 49% unwanted oral sex or attempted intercourse, and 15% rape. 14% of the sample reported experiencing poly-victimization across all 3 types.

Across the first year of college, 83% of women in our analytic sample reported dieting (73% of total sample), 15% use of diet pills (9% of total sample), and 14% use of laxatives or vomiting to counteract weight gain (8% of total sample) during at least one time point. 42% reported at least one experience of sexual assault during the first semester and 35% during second semester. During the first semester, 36% of the total sample reported experiencing unwanted contact, 29% unwanted oral sex or attempted intercourse, and 9% rape. During the second semester, 36% reported experiencing unwanted contact, 21% unwanted oral sex or attempted intercourse, and 9% rape. These prevalence estimates are not population-based and are reflective only of our analytic sample.

Hypothesis A: Pre-College Assault (Quantity) Predicts Unhealthy Weight Management in First Year of College

Number of pre-college sexual assault experiences significantly predicted purging behavior in college, B = .07, SE = .03, p = .03, OR = 1.07. Specifically, a greater number of sexual assault experiences prior to college was associated with increased odds of participants reporting any vomiting or laxative use for weight management across the first year. Other significant predictors in the full model included: history of pre-college eating disorder diagnosis (B = 2.13, SE = .70, p < .01, OR = 8.39), such that having a prior eating disorder diagnosis was associated with increased odds of vomiting and laxative use across the first year.

Number of pre-college sexual assault experiences significantly predicted use of diet pills in college, B = .09, SE = .03, p < .01, OR = 1.09. Specifically, a greater number of sexual assault experiences prior to college was associated with increased odds of participants reporting any diet pill use for weight management across the first year. Other significant predictors in the full model included: history of pre-college eating disorder diagnosis (B = 1.88, SE = .80, p = .02, OR = 6.56) and BMI (B = .20, SE = .06, p < .01, OR = 1.22), such that having a prior eating disorder diagnosis and higher BMI were associated with increased odds of diet pill use across the first year.

Number of pre-college sexual assault experiences significantly predicted dieting behaviors in college, B = .12, SE = .05, p < .01, OR = 1.13. Specifically, a greater number of sexual assault experiences prior to college was associated with increased odds of participants reporting any dieting for weight management across the first year. BMI was also a significant predictor (B = .44, SE = .07, p < .001, OR = 1.55), such that higher BMI was associated with greater odds of dieting across the first year.

Hypothesis B: Pre-College Assault (Severity) Predicts Unhealthy Weight Management in First Year of College

Severity of pre-college sexual assault experiences was not significantly associated with purging behaviors in college (B = .36, SE = .19, p = .06, OR = 1.43). Paralleling the results for quantity of assault and vomiting/laxative use, significant predictors in the full model included: history of pre-college eating disorder diagnosis, B = 2.20, SE = .70, p < .01, OR = 9.02.

Severity of pre-college sexual assault experiences significantly predicted use of diet pills in college, B = .43, SE = .19, p = .03, OR = 1.53. Specifically, increased severity of sexual assault prior to college was associated with increased odds of participants reporting any diet pill use for weight management across the first year. Paralleling the results for quantity of assault and diet pill use, other significant predictors in the full model included: history of pre-college eating disorder diagnosis (B = 2.12, SE = .78, p < .01, OR = 8.36) and BMI (B = .21, SE = .06, p < .001, OR = 1.23).

Severity of pre-college sexual assault experiences predicted dieting behaviors in college, B = .41, SE = .15, p < .01, OR = 1.51. Specifically, increased severity of sexual assault prior to college was associated with increased odds of participants reporting any dieting for weight management across the first year. Paralleling the results for quantity of assault and dieting, higher BMI was also a predictor, B = .45, SE = .07, p < .001, OR = 1.57.

Hypothesis C: First Semester Assault (Quantity) Predicts Unhealthy Weight Management in Second Semester

Number of sexual assault experiences during the Fall semester predicted purging behaviors during the Spring semester, B = .14, SE = .07, p = .03, OR = 1.16. Specifically, a greater number of first semester sexual assault experiences was associated with increased odds of participants reporting vomiting or laxative use during the second semester. Other predictors in the full model included: vomiting or laxative use in the first semester (B = 4.92, SE = .83, p < .001, OR = 137.19) and history of pre-college eating disorder diagnosis (B = 1.97, SE = .97, p = .04, OR = 7.14).

Number of sexual assault experiences during the Fall semester predicted diet pill use during the Spring semester, B = .11, SE = .05, p = .04, OR = 1.11. Specifically, a greater number of first semester sexual assault experiences was associated with increased odds of participants reporting diet pill use during the second semester. Other predictors in the full model included: diet pill use in the first semester (B = 2.41, SE = .77, p < .01, OR = 11.18) and BMI (B = .21, SE = .07, p < .01, OR = 1.24).

Number of sexual assault experiences during the Fall semester was not associated with dieting behaviors during the Spring semester, B = .08, SE = .06, p = .18, OR = 1.08. Predictors in this full model included: dieting in the first semester (B = 2.93, SE = .36, p < .001, OR = 18.75) and BMI (B = .20, SE = .07, p < .01, OR = 1.22).

Hypothesis D: First Semester Assault (Severity) Predicts Unhealthy Weight Management in Second Semester

Severity of sexual assault experiences during the Fall semester predicted purging behaviors during the Spring semester, B = .66, SE = .30, p = .03, OR = 1.94. Specifically, increased severity of first semester sexual assault experiences was associated with increased odds of participants reporting any vomiting or laxative use during the second semester. Predictors in the full model also included vomiting or laxative use in the first semester, B = 4.85, SE = .83, p < .001, OR = 127.42.

Severity of sexual assault experiences during the Fall semester predicted diet pill use during the Spring semester, B = .70, SE = .23, p < .01, OR = 2.02. Specifically, increased severity of first semester sexual assault experiences was associated with increased odds of participants reporting any diet pill use during the second semester. Paralleling the results for quantity of assault and diet pill use, predictors in the full model also included: diet pill use in the first semester (B = 2.23, SE = .81, p < .01, OR = 9.32) and BMI (B = .24, SE = .08, p < .01, OR = 1.27).

Severity of sexual assault experiences during the Fall semester predicted dieting behaviors during the Spring semester, B = .39, SE = .17, p = .02, OR = 1.48. Specifically, increased severity of first semester sexual assault experiences was associated with increased odds of participants reporting any dieting for weight management during the second semester. Paralleling the results for quantity of assault and dieting, predictors in the full model also included: dieting in the first semester (B = 2.96, SE = .36, p < .001, OR = 19.22) and BMI (B = .22, SE = .07, p < .01, OR = 1.24).

Hypothesis E: Reverse Causal Relationship – First Semester Unhealthy Weight Management Does Not Predict Second Semester Assault (Quantity)

Purging behaviors during the Fall semester were not significantly associated with number of sexual assault experiences during the Spring semester, B = .42, SE = .45, p = .35. Predictors of second semester sexual assault in the full model included: number of sexual assault experiences in the first semester (B = .14, SE = .02, p < .001) and history of anxiety disorder (B = 2.65, SE = .93, p < .01).

Diet pill use during the Fall semester was not associated with number of sexual assault experiences during the Spring semester, B = −.17, SE = .41, p = .68. Predictors of second semester sexual assault in the full model included: number of sexual assault experiences in the first semester (B = .13, SE = .02, p < .001) and history of anxiety disorder (B = 2.66, SE = .91, p < .01).

Dieting during the Fall semester was not associated with number of sexual assault experiences during the Spring semester, B = .12, SE = .37, p = .74. Predictors of second semester sexual assault in the full model included: number of sexual assault experiences in the first semester (B = .13, SE = .01, p < .001) and history of anxiety disorder (B = 2.48, SE = .80, p < .01).

Hypothesis F: Reverse Causal Relationship – First Semester Unhealthy Weight Management Does Not Predict Second Semester Assault (Severity)

Purging behaviors during the Fall semester were not significantly associated with severity of sexual assault experiences during the Spring semester, B = −.42, SE = .24, p = .08. Predictors of second semester sexual assault severity in the full model included: severity of sexual assault experiences in the first semester (B = .41, SE = .05, p < .001), history of anxiety disorder (B = −.51, SE = .23, p = .03), history of mood disorder (B = .46, SE = .23, p = .04), and racial/ethnic minority status (B = −.25, SE = .11, p = .03).

Diet pill use during the Fall semester was not associated with severity of sexual assault experiences during the Spring semester, B = .21, SE = .28, p = .46. Predictors of second semester sexual assault severity in the full model included: severity of sexual assault experiences in the first semester (B = .39, SE = .05, p < .001), history of anxiety disorder (B = −.49, SE = .23, p = .03), history of mood disorder (B = .49, SE = .23, p = .03), and racial/ethnic minority status (B = −.24, SE = .11, p = .03).

Dieting behaviors during the Fall semester were not associated with severity of sexual assault experiences during the Spring semester, B = .13, SE = .11, p = .27. Predictors of second semester sexual assault severity in the full model included: severity of sexual assault experiences in the first semester (B = .41, SE = .05, p < .001), history of anxiety disorder (B = −.45, SE = .22, p = .04), history of mood disorder (B = .51, SE = .20, p = .01), and racial/ethnic minority status (B = −.24, SE = .11, p = .03).

Discussion

This study addressed key methodological problems in the existing literature to build on studies examining the association between sexual assault and unhealthy weight management behaviors, using a longitudinal sample of first-year college women. Study hypotheses were largely supported for most outcomes examined. Key findings from the present study included: 1) The number of pre-college sexual assault experiences significantly predicted vomiting or laxative use, diet pill use, and dieting during the first year of college. 2) Severity of pre-college sexual assault experiences significantly predicted diet pill use and dieting during the first year of college. 3) The number of first-semester sexual assault experiences significantly predicted vomiting or laxative use and diet pill use during the second semester (controlling for first semester maladaptive weight management strategies). 4) Severity of first-semester sexual assault experiences significantly predicted vomiting or laxative use, diet pill use, and dieting during second semester (controlling for first-semester maladaptive weight management strategies). 5) Reverse models were non-significant (i.e., unhealthy weight management behaviors in the first semester did not predict sexual assault quantity or severity in the second semester in full models).

The rates of both sexual assault and unhealthy weight management behaviors reported by college women in this study were alarming. Each semester, 36% experienced unwanted sexual contact, 21–29% unwanted oral sex or attempted rape, and 9% rape. The majority (73–83%) of women reported dieting behaviors, 9–15% reported use of diet pills, and 8–14% purging behaviors (vomiting or laxative use) for weight loss, at some point during the first year. Our sexual assault prevalence estimates appear higher than previous studies,31 likely due to use of a more comprehensive definition of sexual assault that include a range of tactics, types of assault, and both attempted and completed experiences. Our weight management behavior estimates appear higher than previous studies as well,32 likely related to the coding of weight management behaviors (see measures section). Although our prevalence estimates are not population-based and should not be generalized as such, they do raise concern about better understanding these experiences and behaviors, and developing strategies for intervention.

Our main study findings in terms of the association between sexual assault and unhealthy weight management behaviors were also largely consistent with previous studies, including those documenting cross-sectional associations,5,6 limited longitudinal work,16 and absence of an effect for the opposite direction when controlling for past assault.2 It is notable that results remained significant with control for covariates (including SES, race/ethnicity, and BMI) and known correlates or risk factors of unhealthy weight management (including history of mood, anxiety, and eating disorders).33,34 In full models examining first-semester variables predicting second-semester variables, control was also included for the dependent variable at baseline. This classic approach to longitudinal analyses helps to increase overall confidence in prediction/causality rather than merely examining correlations or associations. In our models, past unhealthy weight management behaviors were strongly and consistently predictive of later behaviors, consistent with prior longitudinal work demonstrating persistence of these behaviors over time.13 The finding that sexual assault is predictive above and beyond this past behavior is particularly notable. This study was also able to examine both frequency and severity of sexual assault with a well-validated and more comprehensive measure than has been commonly used in the literature. Studies utilizing single self-report items of sexual assault may underestimate the frequency of these experiences, given research showing that definitions of assault vary widely and that many women who have experienced rape do not label their experience as such.35 Also consistent with previous research,36 history of sexual assault among these women was a robust predictor across models of re-assault.

Limitations

Limitations to the present study should be acknowledged and suggest directions for further research. First, a single item was used to measure each unhealthy weight management behavior so these results should be replicated with more comprehensive scales, a broader range of behaviors (e.g., binge eating), and other aspects of disordered eating (e.g., maladaptive cognitions). Second, all behaviors were assessed using self-report, which can be vulnerable to memory or self-presentation bias. Self-reported mental health history (control variables) likely underestimates those who have had symptoms of depression, anxiety, or eating disorders but did not receive formal treatment. Third, the parent study did not assess other types of victimization (e.g., dating violence, physical abuse, emotional or verbal harassment) so we are unable to draw conclusions about whether sexual trauma is uniquely predictive of unhealthy weight management behaviors as compared to these other types of interpersonal trauma. Previous research has had conflicting results in terms of which types of victimization are most relevant, including emotional abuse,18,37 and sexual or physical abuse.6,38 Fourth, our sample consisted of primarily white (64%) college women so caution when generalizing findings is warranted. This is a limitation especially common in the study of eating behaviors, and research with underrepresented groups should be prioritized in the field. Caution is warranted in generalizing findings beyond college students. Finally, although the data were consistent with the hypothesized direction of effect from sexual trauma to unhealthy weight management, and inconsistent with the reverse, it is most accurate to say that we failed to reject the null hypothesis and to emphasize that it is impossible to accept or confirm the null.

Conclusions

The predominant conclusion of this study is that experiences of sexual assault were prospectively associated with a range of unhealthy weight management behaviors among college women. These effects emerged even after controlling for prior eating behaviors. Findings also strongly support the direction of the association from sexual assault to unhealthy weight management, as relationships in the opposite direction were not significant (i.e., reciprocal relationships were not found). Results were relatively robust across types of maladaptive weight management behaviors, time frames (pre-college versus across the first year), and quantity versus severity of sexual assault experiences. Future longitudinal research should further establish the timeline of developing maladaptive eating behaviors after sexual assault and key timepoints for intervention.

Implications

These findings have implications for prevention and intervention efforts among college women. Campus-based advocates working with survivors of sexual assault should provide education about the range of common symptoms experienced following these experiences, including impact on body image and eating behaviors, as well as appropriate resources and referrals. Clinicians evaluating young women who have presented for sexual trauma should closely assess and monitor for unhealthy weight management behaviors over time. A brief screening tool designed for college women39 may reveal the need for services targeting maladaptive weight management behaviors. Studies have demonstrated efficacy of virtual interventions (e.g., The Body Project, cognitive behavioral therapy) in preventing the onset of unhealthy weight and eating behaviors in high risk groups.40,41 Our study findings suggest that there could be significant benefit to considering college women who have experienced sexual assault as a high risk group and including them in these kinds of prevention efforts. Similarly, in developing interventions for these eating behaviors during college, the possible contribution of prior sexual trauma should be appropriately considered. Implementing trauma-informed care in these situations may increase the efficacy of interventions to reduce maladaptive weight management behaviors.42

Our reverse models suggested that it would not be beneficial to focus sexual assault prevention efforts on those with maladaptive weight management behaviors (as these were not predictive in full models); however, focusing prevention efforts on those with a history of mood or anxiety disorder, racial/ethnic minorities, and especially those who have experienced prior assault may lower rates of assault or re-assault. Any discussion of sexual assault prevention should acknowledge that effective approaches must be comprehensive, with primary focus on perpetrators and overall campus culture that contribute to these significant risks rather than placing the emphasis on survivors in preventing assault. Overall, this research provides further evidence of the detrimental effects of sexual trauma on mental and physical health of women and highlights the importance of prevention and intervention programs.

References

  • 1.Carey KB, Norris AL, Durney SE, Shepardson RL, Carey MP. Mental health consequences of sexual assault among first-year college women. J Am Coll Health. 2018;66:480–486. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Gidycz CA, Orchowski L, King CR, Rich CL Sexual victimization and health-risk behaviors: A prospective analysis of college women. J Interpers Violence. 2008;23:744–763. [DOI] [PubMed] [Google Scholar]
  • 3.Champion HLO, Foley KL, DuRant RH, Hensberry R, Altman D, Wolfson M. Adolescent sexual victimization, use of alcohol and other substances, and other health risk behaviors. J Adolesc Health. 2004;35:321–328. [DOI] [PubMed] [Google Scholar]
  • 4.Norris AL, Carey KB, Walsh JL, Shepardson RL, Carey MP. Longitudinal assessment of heavy alcohol use and incapacitated sexual assault: A cross-lagged analysis. Addict Behav. 2019;93:198–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Ackard DM, Neumark-Sztainer D. Date violence and date rape among adolescents: Associations with disordered eating behaviors and psychological health. Child Abuse Negl. 2002;26:455–473. [DOI] [PubMed] [Google Scholar]
  • 6.Silverman JG, Raj A, Mucci LA, Hathaway JE. Dating violence against adolescent girls and associated substance use, unhealthy weight control, sexual risk behavior, pregnancy, and suicidality. JAMA. 2001;286:572–579. [DOI] [PubMed] [Google Scholar]
  • 7.Carey KB, Durney SE, Shepardson RL, Carey MP. Incapacitated and forcible rape of college women: Prevalence across the first year. J Adolesc Health. 2015;56: 678–680. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Muehlenhard CL, Peterson ZD, Humphreys TP, Joskowski KN. Evaluating the one-in-five statistic: Women’s risk of sexual assault while in college. J Sex Res. 2017;54:549–576. [DOI] [PubMed] [Google Scholar]
  • 9.Cranney S The relationship between sexual victimization and year in school in U.S. colleges: Investigating the parameters of the “red zone”. J Interpers Violence. 2015;30:3133–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Galmiche M, Dechelotte P, Lambert G, Tavolacci MP. Prevalence of eating disorders over the 2000–2018 period: A systematic literature review. Am J Clin Nutr. 2019;109:1402–1413. [DOI] [PubMed] [Google Scholar]
  • 11.Striegel-Moore RH, Dohm F, Kraemer HC, Taylor CB, Daniels S, Crawford PB, Schreiber GB. Eating disorders in white and black women. Am J Psychiatry. 2003;160:1326–31. [DOI] [PubMed] [Google Scholar]
  • 12.Stice E, Desjardins CD. Interactions between risk factors in the prediction of onset of eating disorders: Exploratory hypothesis generating analysis. Behav Res Ther. 2018;105:52–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Neumark-Sztainer D, Wall M, Larson NI, Eisenberg ME, Loth K. Dieting and disordered eating behaviors from adolescence to young adulthood: Findings from a 10-year longitudinal study. J Am Diet Assoc. 2011;111:1001–1011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Delinsky S, Wilson GT. Weight gain, dietary restraint, and disordered eating in the freshman year of college. Eat Behav. 2008;9:82–90. [DOI] [PubMed] [Google Scholar]
  • 15.Sares-Jaske L, Knekt P, Mannisto S, Lindfors O, Heliovaara M. Self-report dieting and long-term changes in body mass index and waist circumference. Obes Sci Pract. 2019;5:291–303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Collins B, Fischer S, Stojek M, Becker K. The relationship of thought suppression and recent rape to disordered eating in emerging adulthood. J Adolesc. 2014;37:113–121. [DOI] [PubMed] [Google Scholar]
  • 17.Waller G Perceived control in eating disorders: Relationship with reported sexual abuse. Int J Eat Disord. 1998;23:213–216. [DOI] [PubMed] [Google Scholar]
  • 18.Burns EE, Fischer S, Jackson JL, Harding HG. Deficits in emotion regulation mediate the relationship between childhood abuse and later eating disorder symptoms. Child Abuse Negl. 2012;36:32–39. [DOI] [PubMed] [Google Scholar]
  • 19.Petersen JL, Hyde JS. Peer sexual harassment and disordered eating in early adolescence. Dev Psychol. 2013;49:185–194. [DOI] [PubMed] [Google Scholar]
  • 20.Preti A, Incani E, Camboni MV, Petretto DR, Masala C. Sexual abuse and eating disorder symptoms: The mediator role of bodily dissatisfaction. Compr Psychiatry. 2006;47:475–481. [DOI] [PubMed] [Google Scholar]
  • 21.Cena H, Stanford FC, Ochner L, et al. Association of a history of childhood-onset obesity and dieting with eating disorders. Eat Disord. 2017;25:216–229. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Keel PK, Baxter MG, Heatherton TF, Joiner TE. A 20-year longitudinal study of body weight, dieting, and eating disorder symptoms. J Abnorm Psychol. 2007;116:422–432. [DOI] [PubMed] [Google Scholar]
  • 23.Bryan AEB, Norris J, Abdallah D, et al. Longitudinal change in women’s sexual victimization experiences as a function of alcohol consumption and sexual victimization history: A latent transition analysis. Psychol Violence. 2016;6:271–279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Fielder RL, Walsh JL, Carey KB, Carey MP. Predictors of sexual hookups: A theory-based, prospective study of first-year college women. Arch Sex Behav. 2013;42:1425–1441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Testa M, VanZile-Tamsen C, Livingston JA, Koss MP. Assessing women’s experiences of sexual aggression using the Sexual Experiences Survey: Evidence for validity and implications for research. Psychol Women Q. 2004;28:256–265. [Google Scholar]
  • 26.Koss MP, Gidycz CA. Sexual Experiences Survey: Reliability and validity. J Consult Clin Psychol. 1985;53:422–423. [DOI] [PubMed] [Google Scholar]
  • 27.Turchik JA, Hassija CM. Female sexual victimization among college students: Assault severity, health risk behaviors, and sexual functioning. J Interpers Violence. 2014;29:2439–2457. [DOI] [PubMed] [Google Scholar]
  • 28.Operario D, Adler NE, Williams DR. Subjective social status: Reliability and predictive utility for global health. Psychol Health. 2004;19:237–246. [Google Scholar]
  • 29.CDC. Body Mass Index Website. https://www.cdc.gov/healthyweight/assessing/bmi/index.html. Accessed June 15, 2022.
  • 30.Neumark-Sztainer D, Jeffrey RW, French SA. Self-reported dieting: How should we ask? What does it mean? Associations between dieting and reported energy intake. Int J Eat Disord. 1997;22:437–449. [DOI] [PubMed] [Google Scholar]
  • 31.Fedina L, Holmes JL, Backes BL. Campus sexual assault: A systematic review of prevalence research from 2000 to 2015. Trauma Violence Abuse. 2018;19:76–93. [DOI] [PubMed] [Google Scholar]
  • 32.Fayet F, Petocz P, Samman S. Prevalence and correlates of dieting in college women: A cross-sectional study. Int J Womens Health. 2012;4:405–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Stephen EM, Rose JS, Kenney L, Rosselli-Navarra F, Weissman RS. Prevalence and correlates of unhealthy weight control behaviors: Findings from the National Longitudinal Study of Adolescent Health. J Eat Disord. 2014;2:3630. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Lloyd EC, Haase AM, Zerwas S, Micali N. Anxiety disorders predict fasting to control weight: A longitudinal large cohort study of adolescents. Eur Eat Disord Rev. 2020;28:269–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Sasson S, Paul LA. Labeling acts of sexual violence: What roles do assault characteristics, attitudes, and life experiences play? Behav Soc Issues. 2014;23:35–49. [Google Scholar]
  • 36.Conley AH, Overstreet CM, Hawn SE, Kendler KS, Dick DM, Amstadter AB. Prevalence and predictors of sexual assault among a college sample. J Am Coll Health. 2017;65:41–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Faravelli S, Stojek M, Hartzell E. Effects of multiple forms of childhood abuse and adult sexual assault on current eating disorder symptoms. Eat Behav. 2010;11:190–192. [DOI] [PubMed] [Google Scholar]
  • 38.Doyle JP, Frank E, Saltzman LE, McMahon PM, Fielding BD. Domestic violence and sexual abuse in women physicians: Associated medical, psychiatric, and professional difficulties. J Wom Health Gend Base Med. 1999;8:955–965. [DOI] [PubMed] [Google Scholar]
  • 39.Graham AK, Trockel M, Weisman H, et al. A screening tool for detecting eating disorder risk and diagnostic symptoms among college-age women. J Am Coll Health. 2018;67:357–366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Ghaderi A, Stice E, Andersson G, Eno Persson J, Allzen E. A randomized controlled trial of the effectiveness of virtually delivered Body Project (vBP) groups to prevent eating disorders. J Consult Clin Psychol. 2020;88:643–56. [DOI] [PubMed] [Google Scholar]
  • 41.Taylor CB, Bryson S, Luce KH, et al. Prevention of eating disorders in at-risk college-age women. Arch Gen Psychiatry. 2006;63:881–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Brewerton TD. An overview of trauma-informed care and practice for eating disorders. J Aggress Maltreatment Trauma. 2019;28:445–462. [Google Scholar]

RESOURCES