Alcohol Use Before Sexual Violence and Cognitive Appraisals: Differential Associations With Barriers to Help-Seeking

Labeling

College women often vary in how they label an experience of rape (Wilson & Miller, 2016). Some appraise their unwanted sexual experience using labels that are more consistent with victimization (e.g., sexual assault and rape), whereas others use less severe, alternate labels (e.g., miscommunication and bad sex). Supporting this, a recent meta-analysis indicated that 62.7% of college women who have experienced rape do not label these experiences as SV (Wilson & Miller, 2016). Although most prior work has examined labeling dichotomously (i.e., whether women label or do not label SV), many women label unwanted sexual experiences on a continuum. Women may view the event as somewhat consistent with SV, use certain labels (e.g., sexual assault) but not others (e.g., rape), or be uncertain about how to define the unwanted experience (Donde et al., 2018; Peterson & Muehlenhard, 2004). These findings underscore the need to comprehensively examine the degree of labeling when considering barriers to help-seeking. If, for example, college women label an experience as less consistent with SV, they may experience more minimization barriers. That is, decreased labeling of SV may impede survivors from perceiving the event as “bad enough” to justify the use of mental health services. Relatedly, decreased labeling could prevent women from identifying the potential benefits of help-seeking as a strategy to reduce the psychological consequences of rape (Wilson & Scarpa, 2017).

Whereas lower labeling may be related to greater minimization barriers, greater labeling may be associated with a higher risk of experiencing negative treatment and social-emotional barriers. For example, women who label their unwanted sexual experiences as more consistent with SV may expect more negative treatment from providers related to concerns about larger stereotypic and inaccurate beliefs about rape (e.g., women often lie about rape), which are still endorsed by some therapists (Idisis et al., 2007). Because of this cultural milieu in which survivors are often unsupported and poorly treated, women who identify an unwanted sexual experience as SV may be more likely to anticipate that therapists would respond in similar ways, such as by blaming or not believing them (Idisis et al., 2007). Likewise, women who report greater SV labeling may perceive more social-emotional barriers. Because they self-identify as having experienced SV, these women may be more concerned that peers and others will respond in unsupportive ways that are consistent with myths about rape. Women who label SV more might also be more likely to avoid thoughts or reminders of SV to reduce psychological distress (Littleton et al., 2006), which could include disclosure to formal mental health providers.

Self-Blame

In addition to the labeling of one's SV experience, appraisals that reflect self-blame for rape may also be related to barriers to help-seeking. Although the responsibility for SV is always with the perpetrator, women often believe they should have known better or could have done more to protect themselves (Orchowski et al., 2009; Suarez & Gadalla, 2010). Such attributions of self-blame can arise from survivors’ attempts to assimilate the SV experience into preexisting beliefs regarding one's capacity for agency and control (Resick & Schnicke, 1992) and are associated with a lower likelihood of disclosure, including to mental health providers (Holland & Cortina, 2017). One way in which self-blame could hinder help-seeking is by promoting minimization barriers, in which survivors hold beliefs that they are unworthy of accessing formal support or that it is their responsibility to address any negative mental health consequences of SV (Patterson et al., 2009). Women who blame themselves might also expect more negative treatment barriers, including that providers will blame and judge them or fail to act in response to their disclosure (Kennedy & Prock, 2018). Similarly, social-emotional barriers may be higher among survivors who believe they were at fault. For example, self-blame may lead to fear of being criticized by others or getting in trouble if they were to seek formal mental health services (Patterson et al., 2009; Ullman & Lorenz, 2020). Women who experience greater self-blame related to rape may also anticipate that talking about SV would evoke greater feelings of embarrassment or shame (Kennedy & Prock, 2018) and therefore be more reticent to seek help.

Survivor Alcohol Use, Labeling, and Self-Blame

Individuals are less likely to label unwanted sexual experiences as SV when they involve alcohol consumption by the survivor (Grubb & Turner, 2012). Indeed, although most SV experiences in college involve alcohol use (Mohler-Kuo et al., 2004), societal beliefs about what constitutes rape often fail to include drinking (Ryan, 2011). Likewise, many women who consumed alcohol prior to SV do not label their own experience as rape (Kahn et al., 2003). One possible reason is that many women report they were less able to forcibly resist unwanted sexual advances because of intoxication, and this lack of forcible resistance may have precluded them from identifying the SV experience as rape due to inaccurate societal myths that “real rape” involves physical force (Kahn et al., 2003). Given the likelihood that survivor alcohol use is associated with lower labeling, it seems probable that alcohol use may also be indirectly associated with barriers to help-seeking through labeling.

Survivor alcohol use may also exacerbate inaccurate attributions of self-blame (Jaffe, Steel, et al., 2021; Peter-Hagene & Ullman, 2018). Such appraisals would be consistent with societal myths and misperceptions that women who consumed alcohol before SV are culpable for rape (Grubb & Turner, 2012), as well as findings that women who were drinking prior to SV are viewed as less credible and more blameworthy than women who had not been drinking (Grubb & Turner, 2012; Wenger & Bornstein, 2006). Consistent with this, participants who thought they consumed alcohol during a laboratory-based alcohol administration study reported greater self-blame following a hypothetical rape scenario and, in turn, were less likely to indicate that they would have reported the rape to the police (Flowe & Maltby, 2018). Because greater self-blame is likely related to both survivor alcohol use and greater barriers to help-seeking, it is possible that alcohol use is indirectly associated with each of the three barriers through self-blame. Such a pattern would suggest that women who were drinking prior to SV are more likely to blame themselves for rape and, in turn, perceive greater barriers to mental health services.

Participants

Undergraduate women (N = 141; 140 cisgender women, 1 declined to state her sex assigned at birth) were recruited from the psychology subject pool at a large Midwestern university in the United States. Participants were drawn from a larger sample who participated in a study about “life experiences” (Jaffe, Cero, et al., 2021; Jaffe, Kumar, et al., 2021) where the only inclusion criterion was being at least 19 years old (the local age of majority). To be included in the current analyses, participants had to identify as a woman (which may include both cisgender and transgender women), given that people who identify as women disproportionately experience SV victimization (Breiding et al., 2014). Participants also had to have experienced SV consistent with legal definitions of rape (see below) since the age of 18 and indicate that they did not seek mental health services following this event (“Have you seen a counselor or mental health professional since the unwanted incident?”). Of 2,033 participants who completed the baseline survey from spring 2018 to spring 2020 (data not collected in the summers), 1,783 (87.7%) provided valid data (i.e., passed 75% of attention checks). Out of those participants, 1,245 (69.8%) identified as women, of which 209 (16.8%) reported experiencing rape since age 18, and 141 (67.5%) did not seek mental health services.

Participants’ mean age was 20.48 (SD = 2.43), ranging from 19 to 35 years old. Most participants identified as heterosexual (90.1%), with 8.5% identifying as bisexual, and 1.4% identifying as “something else” or “don't know” (0.7% did not think of themselves as having a sexuality and 0.7% reported still figuring out their sexuality). Consistent with larger university demographics, participants were primarily White (89.4%), followed by Hispanic/Latinx (9.9%), Black (3.5%), Asian (3.5%), American Indian or Alaska Native (2.1%), Pacific Islander or Native Hawaiian (1.4%), and not listed (0.07%; 1 participant wrote Middle Eastern). Participants could select multiple racial and ethnic categories.

Procedure

The study was approved by the university's Institutional Review Board. Participants were recruited from the psychology subject pool and completed informed consent prior to study onset. Next, they completed online questionnaires via Qualtrics, including the self-report measures described below. Although participants were invited to complete up to two follow-up surveys depending on the time remaining in the semester (see Jaffe, Kumar, et al., 2021), current analyses focused on the baseline survey. After study completion, students were provided with information about campus counseling services and received research credit for participation.

Measures

Data Analytic Plan

We used SPSS Version 27 to examine descriptive statistics and Pearson's r correlations and Mplus Version 8.4 (Muthén & Muthén, 1998–2019) to test the hypothesized indirect effects. Given that the outcomes of barriers to help-seeking may be interrelated, but are worthy of separate examination, a separate parallel mediation model was tested for each of the three hypothesized effects: (1) the indirect effect of survivor alcohol use on minimization barriers via labeling and self-blame; (2) the indirect effect of survivor alcohol use on negative treatment barriers via labeling and self-blame; and (3) the indirect effect of survivor alcohol use on social-emotional barriers via labeling and self-blame. Across all three models, self-blame and labeling were covaried. Bias-corrected bootstrapping with 5,000 samples was used to obtain 95% confidence intervals (CIs) for the indirect effects (Preacher et al., 2007). If the CI does not include zero, an indirect effect is supported. The covariance coverage ranged from 97.9% to 100%. Participants with missing data were retained in analyses using maximum likelihood estimation (Enders, 2010). Because all possible paths were estimated in the model, all models were saturated. Therefore, traditional measures of global fit were not interpreted. For each barrier, the amount of variance explained by the model (R²) was examined.

Descriptive Statistics

Descriptive statistics and correlations are provided in Table 1. All variables were within acceptable ranges for skewness and kurtosis (Kline, 2015). As expected, alcohol use prior to SV was associated with greater self-blame. Unexpectedly, survivor alcohol use was associated with greater labeling. Alcohol use prior to SV was not related to any of the three barriers to help-seeking. Lastly, greater labeling and self-blame were each significantly associated with less minimization barriers, and greater negative treatment and social-emotional barriers.

OPEN IN VIEWER

Table 1.

Descriptive Statistics and Correlations.

	n	n (%) or Mean (SD)	Range	1	2	3	4	5	6
1. Alcohol use	140	68 (48.6%)	0, 1	—	.21*	.22**	−.02	−.08	.04
2. Labeling	141	4.04 (1.65)	1-7		—	.35***	−.48***	.17*	.29***
3. Self-blame	141	3.10 (1.23)	1-5			—	−.21*	.29***	.34***
4. Minimization barriers	140	5.38 (3.28)	0-12				—	−.11	-.06
5. Negative treatment barriers	140	1.69 (2.89)	0-15					—	.71***
6. Social–emotional barriers	138	6.01 (6.61)	0-32						—

Note. *p < .05; **p  < .01; ***p < .001.

Parallel Mediation Models

Across all parallel mediation models, the paths from the independent variable (survivor alcohol use) to the mediators (labeling and self-blame), as well as the association between the mediators, are the same. These models revealed that survivor alcohol use was significantly associated with greater labeling (b = 0.70, SE = 0.27, p < .01, β = .21) and greater self-blame (b = 0.53, SE = 0.20, p < .01, β = .22). Greater labeling was significantly associated with greater self-blame (b = 0.60, SE = 0.17, p < .001, β = .31). Direct and indirect effects with the dependent variable in each model (barriers to help-seeking) are described below.

Minimization Barriers

As shown in Figure 1, greater labeling was significantly associated with less minimization barriers (b = −0.95, SE = 0.14, p < .001, β = −.48). However, self-blame was not significantly associated with minimization barriers (b = −0.15, SE = 0.23, p = .51, β = −.06). Alcohol use prior to SV and minimization barriers were no longer directly associated when controlling for labeling and self-blame (b = 0.64, SE = 0.51, p = .21, β = .10). Lastly, there was an indirect effect of alcohol use to less minimization barriers via greater labeling (b = −0.66, 95% CI [−1.27, −0.19], β = −.10), but not self-blame (b = −0.08, 95% CI [−0.43, 0.14]), indicating that greater labeling (but not self-blame) explained the link between survivor alcohol use and less minimization barriers. The overall model explained 23.9% of the variance in minimization barriers. Together, findings suggest that survivors who had been drinking prior to the SV event were more likely to label their unwanted experience as SV and, in turn, endorsed fewer minimization barriers (i.e., beliefs that the assault was not “serious enough” to use mental health services).

OPEN IN VIEWER

Figure 1.

Path analyses for minimization barriers. Note. Unstandardized coefficients (SE) are reported for each path. Labeling and self-blame (mediators) were covaried. An indirect effect of alcohol use prior to the SV event on minimization barriers was present via labeling (b = −0.66, 95% CI [−1.27, −0.19], β = −.10), but not self-blame (b = −0.08, 95% CI [−0.43, 0.14]). The model explained 23.9% of the variance in minimization barriers. *p < .05; **p <.01; ***p < .001.

Negative Treatment Barriers

As shown in Figure 2, greater self-blame was significantly associated with greater negative treatment barriers (b = 0.67, SE = 0.18, p < .001, β = .28). However, labeling was not significantly associated with negative treatment barriers (b = 0.19, SE = 0.16, p = .25, β = .11). There was a remaining direct association between alcohol use prior to SV and decreased negative treatment barriers when controlling for labeling and self-blame (b = −0.93, SE = 0.47, p < .05, β = −.16). Lastly, there was an indirect effect of alcohol use to greater negative treatment barriers via greater self-blame (b = 0.35, 95% CI [.11, .75], β = .06), but not labeling (b = 0.13, 95% CI [−.05, .53]), indicating that greater self-blame (but not labeling) explained the association between survivor alcohol use and greater negative treatment barriers. The model explained 10.9% of the variance in negative treatment barriers. These findings suggest that survivors who had been drinking were more likely to blame themselves for SV and, in turn, endorsed greater negative treatment barriers (i.e., beliefs that they would be treated negatively by mental health providers if they were to seek help).

OPEN IN VIEWER

Figure 2.

Path analyses for negative treatment barriers. Note. Unstandardized coefficients (SE) are reported for each path. Labeling and self-blame (mediators) were covaried. An indirect effect of alcohol use prior to the SV event on negative treatment barriers was present via self-blame (b = 0.35, 95% CI [.11,.75], β = .06), but not labeling (b = 0.13, 95% CI [−.05, .53]). The model explained 10.9% of the variance in negative treatment barriers. *p < .05; **p <. 01; ***p < .001.

Social-Emotional Barriers

As shown in Figure 3, greater self-blame (b = 1.52, SE = 0.39, p < .001, β = .28) and labeling (b = 0.80, SE = 0.36, p < .05, β = .20) were each significantly associated with greater social-emotional barriers. Survivor alcohol use and social-emotional barriers were no longer directly associated when controlling for labeling and self-blame (b = −0.93, SE = 0.98, p = .34, β = −.07). There were indirect effects of survivor alcohol use to greater social-emotional barriers via greater labeling (b = 0.56, 95% CI [0.07, 1.60], β = .04) and greater self-blame (b = 0.80, 95% CI [0.26, 1.69], β = .06), indicating that both greater self-blame and greater labeling explained the association between survivor alcohol use and greater social-emotional barriers. The model explained 14.8% of the variance in social-emotional barriers. Thus, survivors who had been drinking were more likely to label their unwanted experience as SV and blame themselves, which, in turn, were each related to greater social-emotional barriers (i.e., beliefs that they would experience emotional and social consequences if they were to seek help from a mental health provider).

OPEN IN VIEWER

Figure 3.

Path analyses for social-emotional barriers. Note. Unstandardized coefficients (SE) are reported for each path. Labeling and self-blame (mediators) were covaried. An indirect effect of alcohol use on social-emotional barriers was present via labeling (b = 0.56, 95% CI [.07, 1.60], β = .04) and self-blame (b = 0.80, 95% CI [.26, 1.69], β = .06). The model explained 14.8% of the variance in social-emotional barriers. *p < .05; **p < .01; ***p < .001.

Labeling

Unexpectedly, survivor alcohol use was associated with greater SV labeling, such that women who had consumed alcohol prior to SV were more likely to identify and label the unwanted sexual experience as terms consistent with violence, including rape, sexual assault, and nonconsensual. This finding is contrary to prior work showing that alcohol-involved SV might be difficult for women to identify as sexual assault or rape because it differs from common social myths about what constitutes “real rape” (e.g., involves the use of physical force; Kahn et al., 2003). It is possible that this pattern of associations could have been a product of our selection of a sample of women who did not use services if decisions about help-seeking are independently driven by assault characteristics (i.e., alcohol use) and labeling. However, another possibility could be increased societal awareness in recent years following the #MeToo movement (Jaffe, Cero, et al., 2021), including about the high cooccurrence between alcohol and SV. Moreover, colleges and universities are now mandated to implement prevention programs to address high rates of SV (Campus Sexual Violence Elimination Act, 2011), which often include depictions of survivor alcohol use in hypothetical rape scenarios (Thompson et al., 2021). The educational value of these programs is well-documented (Jouriles et al., 2018) and may help college women incorporate alcohol use into their scripts for SV, thereby increasing the likelihood of labeling unwanted sexual experiences as rape. Although such a trend would be positive, greater SV labeling was associated with greater self-blame, suggesting that enhanced awareness of the cooccurrence between alcohol use and SV may not be sufficient to inhibit feelings of self-blame among rape survivors who had been drinking.

When examining our overall models, results revealed that greater labeling mediated associations between survivor alcohol use and less minimization barriers, as well as associations between survivor alcohol use and greater social-emotional barriers. Together, findings suggest that although women who endorse greater labeling of alcohol-involved SV may increasingly view mental health services as a strategy they could use to cope with SV, they may still be hesitant to use these services due to concerns about other negative consequences. These women might be more attentive to risks associated with loss of confidentiality, such as providers reporting SV to other parties (e.g., police and Title IX administrators) or perpetrator retaliation. Similarly, women who more strongly identify alcohol-involved SV as victimization may expect that disclosure will be more emotionally difficult, thereby reducing their willingness to speak with a mental health provider. In contrast, women who labeled their unwanted experience in ways that were less aligned with SV may expect less of these consequences but remain uncertain as to whether what they experienced was “serious” (i.e., minimization). In other words, the extent to which women label SV may be related to whether social-emotional or minimization barriers are most salient in explaining why they did not seek help. Notably, alcohol-involved SV was particularly relevant for social-emotional barriers because of greater labeling, whereas SV that did not involve alcohol was associated with greater minimization barriers via lower labeling.

In contrast, labeling was not associated with negative treatment barriers at the bivariate level and did not account for associations between survivor alcohol use and negative treatment barriers. Consequently, the ways in which survivors identify SV experiences seem to be unrelated to their expectations of providers. Moreover, the model examining predictors of negative treatment barriers explained the least amount of variance, suggesting that negative treatment barriers could be better accounted for by other variables that were not assessed here. Factors that play a role in fostering negative treatment barriers beyond labeling, such as past experiences with mental health providers (Ullman & Lorenz, 2020), should be explored.

Self-Blame

Findings regarding associations between alcohol use and our second cognitive appraisal, self-blame, were consistent with our hypotheses. That is, women who consumed alcohol prior to being raped attributed more responsibility for SV to themselves. Although the responsibility for SV is always with the perpetrator, survivors of alcohol-involved SV may blame themselves for drinking too much or not doing more to protect themselves, fostering inaccurate beliefs that their actions contributed to rape. Given the upsetting nature of such attributions, alcohol use prior to SV is an important risk factor to consider when seeking to alleviate psychological distress among survivors.

When examining each of the three mediation models involving self-blame, results revealed that greater self-blame mediated associations between survivor alcohol use and negative treatment barriers, as well as associations between survivor alcohol use and greater social-emotional barriers. Women who had been drinking prior to SV and blame themselves may be more likely to not seek services because they are worried that they would be treated poorly by providers and would experience negative outcomes resulting from service use. If women who were drinking prior to SV internalize societal beliefs that place blame on survivors, women may expect that others will respond in a similar manner, such as with hostile or judgmental reactions that place blame on aspects of their behavior at the time of SV. Similarly, they may expect providers and peers to excuse or minimize the perpetrator's actions if they see themselves as more responsible for SV. In addition to these concerns about how others might respond, women may anticipate that breaches in confidentiality would have more negative consequences (e.g., getting in trouble for drinking) if they have internalized blame. Relatedly, the process of disclosure to a formal provider may be viewed as more distressing if women expect to feel high levels of shame or embarrassment. Consequently, efforts to reduce self-blame following alcohol-involved SV could help address fears related to help-seeking.

However, self-blame did not account for associations between survivor alcohol use and minimization barriers. Although we expected that women would be more likely to believe that what happened to them was not important enough to seek help when they had internalized blame for SV, bivariate findings suggest that these women were less likely to experience minimization barriers. It is possible that self-blame is related to greater psychological distress following rape, intensifying the perceived need for mental health services. However, findings suggest that self-blame is less relevant when considered in context with labeling among women who had been drinking prior to SV. Whereas labeling was directly related to minimization barriers following alcohol-involved SV, associations between self-blame and minimization barriers might vary as a function of other factors. For instance, women who blame themselves for alcohol-involved SV might experience less minimization barriers if they use less effective coping strategies (e.g., using disengagement rather than reappraisal; Clements & Ogle, 2009; Littleton et al., 2006). Such ineffective coping skills could impede women's capacity to regulate negative emotions associated with self-blame, thereby amplifying the perceived value of services to mitigate distress.

Limitations

Although the findings of this study advance current knowledge regarding the complex nature of help-seeking barriers among women who have experienced rape, several limitations should be noted. First, the retrospective, cross-sectional design limits our ability to draw conclusions about temporal ordering. Although all mediators and outcome variables were linked to a single prior experience of rape (i.e., survivors were asked to report about each construct in relation to their most upsetting experience), it is possible that barriers associated with the decision not to seek mental health services could influence labeling and appraisals of blame. Future studies should use longitudinal designs to examine these and related questions, such as how cognitive appraisals might precede and predict interest in and barriers to help-seeking over time. In addition, women may have experienced other SV events that were not examined in the current analysis. If women experienced multiple victimizations, it is possible that they would be most likely to label and seek help related to their most upsetting experience. Future research could address this question by examining barriers to help-seeking across all possible SV experiences.

Second, our sample consisted of undergraduate women who were primarily White, cisgender, and heterosexual. Although these individuals are at high risk for SV (Fedina et al., 2018), findings may not generalize to men, individuals who are gender diverse or gender nonbinary, or cisgender women with varied intersectional identities (e.g., same-aged peers who are not attending college, older women, racial/ethnic minorities, and sexual minorities). Indeed, these individuals may experience even greater rates of barriers to help-seeking because of cultural stereotypes about victim worthiness or anticipated discrimination. The absence of transgender women from our sample, in particular, is a limitation given the greater risk for SV among transwomen (Coulter et al., 2017), as well as discriminatory barriers to service use (e.g., rigid intake questionnaires that misgender help-seekers; Todahl et al., 2009). Another understudied population in SV research is cisgender and transgender men, who report that they did not seek help due to fears that providers would not take them seriously or would perceive them as weak or unmasculine (Donne et al., 2018). Relatedly, Black women may be less likely to seek services if they expect providers to hold negative and inaccurate stereotypes about Black woman sexuality or disclose to the police (Tillman et al., 2010). Thus, work is needed to test the associations examined here using more diverse samples.

Finally, we examined labeling using an aggregated score that captured several different labels for SV (i.e., rape, sexual assault, and not consensual) to account for the varied ways in which survivors may choose to define SV. Some women, for example, label SV as sexual assault but not as rape, whereas others identify with both labels (Donde et al., 2018). However, it is possible that each label could have unique implications, such as being differentially associated with barriers to help-seeking, which should be examined in future work. Similarly, we examined alcohol use as a binary variable in order to compare participants who had and had not been drinking prior to their unwanted sexual experience. However, it is possible that there are also differences in cognitive appraisals and, in turn, barriers to help-seeking based on the amount of alcohol that a survivor consumed or how intoxicated they felt at the time of SV. Future studies could examine such variables (e.g., number of drinks, subjective intoxication, and perceived impairment) to determine whether labeling and self-blame might increase as a function of the amount of alcohol consumed, thereby further exacerbating certain barriers to help-seeking. Similarly, future studies could use experimental methodologies, such as alcohol administration, to examine whether acute intoxication impacts cognitive appraisals and perceived barriers to help-seeking using methods such as hypothetical vignettes to maximize internal validity.

Implications and Future Directions

Limitations notwithstanding, the current findings have implications for broader efforts to reduce barriers to help-seeking among undergraduate SV survivors. Although not all survivors will want to utilize mental health services, it is important to reduce barriers to help-seeking as a means of facilitating women's ability to use these services if they wish. Doing so would help empower survivors to make the choice that is best for them without being impeded by barriers. First, to overcome associations between labeling and minimization barriers among women whose SV experience included drinking, programming could incorporate explicit references to alcohol as a prevalent characteristic of SV on college campuses to decrease the likelihood that women who are interested in mental health services feel unable to go because they do not think what happened to them was worthy of formal help-seeking. Second, mental health and advocacy services could use inclusive language (e.g., “serving those who have experienced unwanted sexual activity”) to decrease the perception that an experience is not “serious” or “severe” enough to use services among women who label SV less. Third, because greater labeling was related to greater social-emotional barriers, focusing on the confidential nature of services could help assuage fears about adverse consequences. Colleges often fail to provide sufficient transparency regarding Title IX policies, such as mandatory reporting requirements for employees, which can leave students unsure about where and to whom they can disclose SV without triggering a report to the university (Holland et al., 2018, 2021). Because of this, services that are confidential and do not require a mandated report under university Title IX policy should emphasize this feature to increase transparency and accessibility. Mental health services could emphasize that college women will not get in trouble or be reported for underage drinking and that providers, in most cases with legal adults, will not disclose SV to other campus or community services unless survivors request that information to be shared.

Relatedly, findings underscore the importance of targeting self-blame among women who had been drinking before SV as a means of lessening negative treatment and social-emotional barriers. Campaigns designed to increase campus awareness could emphasize that experiences of SV are the sole responsibility of the perpetrator and that drinking does not mean the survivor is to blame. Moreover, mental health providers and other campus personnel (e.g., resident assistants, faculty, and administrators) might benefit from additional training to reduce the likelihood that they respond to survivors in negative ways that affirm inaccurate beliefs about rape. Doing so could help create a campus climate where women feel empowered to share such experiences without fear of social reprisal or blame. Further, programs could evaluate and implement strategies to decrease the stigma associated with the use of mental health services. Such strategies could include increased outreach to spread awareness about the nature and key features of mental health services (e.g., confidential, collaborative, and supportive) to buffer against concerns that could prevent women from seeking services. For instance, providers could emphasize that sessions are confidential, and survivors will not have to share any information they do not wish to disclose.

The current findings also suggest several future directions for research, such as exploring other cognitive appraisals that could impact barriers to mental health services. For instance, women who view the world as less benevolent (e.g., bad events are common; Janoff-Bulman, 1989) may be more likely to anticipate the negative consequences of help-seeking. Similarly, other assault characteristics (e.g., use of force and relationship with the perpetrator) may be related to how women perceive their SV experiences (Jaffe, Cero, et al., 2021) and, in turn, these barriers.

Finally, future work could examine potential moderators of the associations found here, such as rape myth acceptance (i.e., inaccurate beliefs about the causes of SV; Suarez & Gadalla, 2010). It is possible, for example, that the relation between survivor alcohol use and self-blame is stronger when women report more beliefs that their actions can cause SV or that perpetrators do not mean to commit harm. Contextual factors, including perceived social support from other sources (e.g., friends and family), access to adaptive emotion regulation and coping strategies, and the quality of prior experiences with mental health services, may also play a role in moderating the associations found here. An examination of these factors could help identify risk and resiliency factors that could be targeted to reduce barriers to help-seeking among college women.

Together, findings reveal that barriers to help-seeking vary based on the presence of alcohol consumption prior to SV and the extent to which women label and blame themselves for unwanted sexual experiences. Whereas lower labeling of SV seems to be most important when considering reasons for not seeking services that are linked with perceptions that SV was not important enough to justify formal support (minimization), self-blame was uniquely associated with concern about poor treatment by providers. Moreover, both greater labeling and self-blame were associated with greater barriers related to social and emotional costs that could result from help-seeking. Targeting these cognitive appraisals may help reduce barriers to help-seeking so that survivors feel able and empowered to access mental health services if they choose to do so.