Article # To Tell or Not to Tell Violence Against Women 2015, Vol. 21(9) 1145–1165 © The Author(s) 2015 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1077801215590672 vaw.sagepub.com SAGE Carrie M. Carretta¹, Ann W. Burgess², and Rosanna DeMarco³ ## Abstract The underreporting of rape is well known; however, there is less information on women who fail to disclose to anyone. This online study suggests that 24% of 242 women who were non-disclosing compared with those who had disclosed were significantly less likely to seek treatment for emotional injuries. Also, almost two thirds of non-disclosing women believed that the abuse was their fault versus 39.1% of women with prior disclosure. Of clinical interest is that regardless of disclosure pattern, there was no significant difference in reports of depression, anxiety, or posttraumatic stress disorder (PTSD), and the majority of respondents endorsed support for online counseling over telephone or individual contact. ## Keywords | | | | :--- | :--- | | rape, disclosure, depression, anxiety, PTSD | | ## Background For centuries, rape thrived on prudery and silence. The silence lifted when consciousness-raising (CR) groups became the major organizing tool of the re-emergence of the women’s rights movement in the late 1960s. These CR groups involved informal groups of women discussing their experiences with incest, child, adolescent, and adult rape. Prior to that time, women who disclosed a sexual assault to law enforcement risked censure, scorn, indifference, or loss of credibility (Ledray, Burgess, & Giardino, 2011). 1Rutgers, The State University of New Jersey, Newark, USA 2Boston College, Chestnut Hill, MA, USA 3University of Massachusetts Boston, USA Corresponding Author: Carrie M. Carretta, Assistant Professor/Research Faculty at Rutgers, The State University of New Jersey, 180 University Avenue, Ackerson Hall, Room 224, Newark, NJ 07102, USA. Email: carrie.carretta@sn.rutgers.edu; carrie.carretta@gmail.com 3502-014 Page 1 of 21 EFTA_00001434 EFTA00156775 1146 Violence Against Women 21(9) Early published evidence by Smith et al. (2000) indicated restricted disclosure of women who reported being raped. They gathered representative data from 3,220 Wave II respondents from the National Women's Study telephone survey regarding the length of time women who were raped before age 18 delayed disclosure, to whom they disclosed, and variables that predicted disclosure. There were 288 women who reported at least one rape prior to their 18th birthday. Fully 28% of child rape victims reported that they had never told anyone about their child rape prior to the research interview and 47% did not disclose for over 5 years post rape. Close friends were the most common confidants. Younger age at the time of rape, family relationship with the perpetrator, and experiencing a series of rapes were associated with disclosure longer than 1 month. Shorter delays were associated with stranger rapes. Logistic regression revealed that the age at rape and knowing the perpetrator were independently predictive of delayed disclosure. Relevant to these findings, Fisher, Daigle, Cullen, and Turner (2003) identified key factors that contributed to rape reporting included the impact of self-blame, the seriousness of the incidents, type of victim-offender relationships, certain victim characteristics (e.g., age, income level, education level, race), and the contextual characteristics of the crimes. Starzynski, Ullman, Filipas, and Townsend (2005) emphasized that deciding whom to tell about sexual assault is an important and potentially consequential decision for sexual assault survivors. A diverse sample of adult sexual assault survivors in the Chicago area was surveyed about sexual assault experiences, social reactions received when disclosing assault to others, attributions of blame, coping strategies, and posttraumatic stress disorder (PTSD). Women disclosing to both formal and informal support sources experienced more stereotypical assaults, had more PTSD symptoms, engaged in less behavioral self-blame, and received more negative social reactions than those disclosing to informal support sources only. Research shows that survivors with lower levels of posttraumatic stress or depressive symptoms are less likely to seek help from formal social systems (Lewis, 2005; Starzynski et al., 2005). In addition, survivors who blamed themselves for causing the rape were less likely to disclose the rape to formal social systems (Starzynski et al., 2005). Although studies have shown that survivors with less severe psychological symptomatology are less likely to seek assistance, it is still unclear what prevents these survivors from seeking help. In a mail survey with 155 respondents studying how social reaction to rape disclosure affects sexual assault victims, Ullman (1996) found that negative social reactions were strongly associated with increased psychological symptoms, while most positive social reactions were unrelated to adjustment (Ullman, 1996). The only social reactions related to better adjustment were being believed and being listened to by others. Wolitzky-Taylor et al. (2011) interviewed a national sample of 2,000 college women about rape experiences in 2006 and found only 11.5% of college women in the sample reported their most recent/only rape experience to authorities, with only 2.7% of rapes involving drugs and/or alcohol reported (Wolitzky-Taylor et al., 2011). Minority status (i.e., non-White race) was associated with lower likelihood of report- 3502-014 Page 2 of 21 EFTA_00001435 EFTA00156776 Carretta et al. 1147 ing, whereas sustaining injuries during the rape was associated with increased likelihood of reporting. Currently, there is developing published evidence on rape reporting and disclosure. Using a prospective study to identify predictors of sexual assault disclosure, Orchowski and Gidycz (2012) examined the responses of 374 support providers and learned women most often disclosed a sexual assault to a female peer supporting the findings over 10 years ago by Smith et al. (2000). Despite the feminist movement of the 1970s, which marked the beginning of the era of rape reform in the United States, to fast-forward to 2013, two findings that affect a victim's mental health have not changed. First, sexual assault remains the most widely underreported violent crime and second, victims typically do not seek help after coercive sexual encounters (Fisher et al., 2003; Siegel, Golding, Stein, Burnam, & Sorenson, 1990). Second, statistics from the National Violence Against Women Survey (NVAWS) indicate that only 19.1% of the women and 12.9% of the men who were raped since their 18th birthday reported their rape to the police (Tjaden & Thoennes, 2006). Indeed, the underreporting of sexual assault persists in bearing the infamous label of “the hidden crime,” and poses serious problems on an individual and societal level (Grohol, 1997). Parallel with the incidence of rape being far more extensive than reported in official statistics is the fact that the large majority of rapists are never apprehended. In 2007, there were 90,427 incidents of rape reported to law enforcement that resulted in only 23,307 arrests or 25.8% of reported cases (U.S. Department of Justice, 2008). Victimization data show a higher number of rapes and sexual assaults—191,670 (Catalano, 2006)—which means that potentially more than half of the rapes and sexual assaults go unreported (and therefore unpunished) to law enforcement (Fisher et al., 2003). Clearly, the vast majority of rapists are never brought to justice as FBI clearance rates for rape average about 50% per year. From a public policy perspective, official estimates of the incidence and prevalence of sexual assault that are used for planning program initiatives are likely underestimated; therefore, individuals and areas that are at high risk for sexual assault are likely failing to receive adequate attention. In addition, the failure to report precludes the arrest of offenders, which limits the degree to which the criminal justice system can serve as a deterrent to sexual assault crimes (Fisher et al., 2003). ## Rape and Self-Disclosure—Keeping a Secret With Silence The issue of self-disclosure—that information about oneself that a person is willing to reveal to others—is an important area of clinical inquiry. Rape traditionally has not been a socially acceptable issue for disclosure. In decades past, it has often been seen as something that lessened the worth of the victim and that was the victim's fault. Thus, Irving Goffman's classic analysis of stigma and the management of spoiled identity are particularly useful in analyzing the disclosure of a rape. Goffman (1963) uses the term "stigma" to indicate "an attribute that is deeply discrediting" in a certain 3502-014 Page 3 of 21 EFTA_00001436 EFTA00156777 1148 Violence Against Women 21(9) social context (p. 3). He distinguishes between cases where the stigma is known to others already or evident immediately (the individual is discredited), and cases where the stigma is not known by others and not immediately perceivable (the individual is discreditable). As Goffman states, when an individual's differentness is not immediately apparent, and is not known beforehand . . . then . . . the issue is . . . that of managing information about his failing. To tell or not to tell; to display or not to display; to let on or not to let on; to lie or not to lie; and in each case, to whom, how, when, and where (p. 40). The issue becomes “the management of undisclosed discrediting information about self” (p. 42). Secrets, especially those involving incest, rape, and abortion, are closely aligned with non-disclosure. Georg Simmel (1906) defines secrets as “consciously willed concealment” (p. 449) that involve a tension that when revealed breaks its power that can result in positive or negative outcomes. Secrecy sets barriers. A secret disclosed may advance to betrayal. Wolff (1950) describes a secret being surrounded by temptation and possibility of betrayal, and the external danger of being discovered is interwoven with the internal danger of giving oneself away. As long as the rape remains cloaked in secrecy, the victim only has to manage that information internally; once the rape is disclosed, the victim has to manage the external reactions. Jack (1991, 1999) and Jack and Ali (2010) describe the phenomenon of “silencing the self” as a behavior common to women where information is withheld in the context of all types of relationships. Whether a relationship is experienced with violence or through rape or without these experiences, Jack contends through her Theory of Silencing the Self that women do not share certain thoughts or feelings that would contradict what others expect of them because it jeopardizes relationships with others and how they are “seen.” Congruent with the social stigma that surrounds disclosure, women avoid conflict and silence their voice which often leads to a loss of self as well as feelings of shame and anger. Ironically, avoiding conflict and abiding by societal expectations are found to be protective and normative in many cultures, and yet Jack (1991) found in her original work that the very secret or silence kept was strongly correlated with clinical depression. Jack and Dill (1992) identified four particular sub-concepts of silencing the self-behaviors from the qualitative analyses of data she collected through a large longitudinal study that included conversations with women talking about their lives and depression. The four self-silencing sub-concepts/behaviors are called (a) Silencing the Self, (b) the Divided Self, (c) Care as Self-Sacrifice, and (d) the Externalized Self. The first sub-concept/sub-scale, Silencing the Self, for which the theory was named described how women often do not ask directly for what they need or tell others what they are feeling. The second sub-concept/sub-scale, the Divided Self, described how women present a compliant exterior to the public when they actually feel hostile and angry. The third sub-concept/sub-scale, Care as Self-Sacrifice, described how women put the feelings and needs of another before their own. The fourth sub-concept/ 3502-014 Page 4 of 21 EFTA_00001437 EFTA00156778 Carretta et al. 1149 sub-scale, the Externalized Self, described how women judge themselves by external standards. In the end, the “self” for women is focused in a context of others’ needs, and it is the “self-in-relation” that drives behavior choices to disclose or not to disclose (Jack, 1991; Jordan, Kaplan, Miller, Stiver, & Surrey, 1991). The issue of rape disclosure and moving away from silencing one's voice has psychosocial and clinical importance. One model of rape counseling is to encourage clients to report and talk to members of their social network as support for experiencing a stressful situation. The dilemma, however, is that the decision to tell or not to tell someone about a rape tends to be a difficult one and many women will experience decisional conflict that is related to the uncertainty regarding outcomes that will result from the choice. Regret of disclosure emerges as a potential outcome (Marchetti, 2012). Not to disclose can be viewed as self-protective as the individual has control of the information and preserves relationships without conflict (Jack, 1991). Telling others dilutes the control. We were interested in studying patterns of rape disclosure as basic to assessing coping and adaptation to a traumatic event. This article presents data from a larger rape study with the intent to provide specific implications for counseling victims regarding disclosure of the information. Thus, the research questions for this study were as follows: Research Question 1: What was the pattern of disclosure for rape using an anonymous web-based survey?
Research Question 2: What was the symptom response based on whether the respondent had disclosed prior to the survey or disclosed at the time of survey?
Research Question 3: What type of follow-up was preferred by respondents?
## Method ## Design This study utilized a descriptive cross-sectional design. Participants completed the study via REDCap Survey, a web-based, online survey tool. Online surveys have been established as an effective means of obtaining a large sample of rape victims (Littleton, 2007, 2010). ## Setting Data were collected via an online survey. A web-based procedure was chosen as it has several benefits. First, the use of a web-based survey has been established as an effective means of obtaining a large sample of rape victims (Littleton, 2007, 2010). Furthermore, the use of an online study allows for elimination of missing data by prompting participants to address non-completed items. Finally, this methodology was selected because it offers participants the ability to complete study instruments at their convenience, offers privacy and confidentiality at the time of participation, and affords the participant an opportunity for safe disclosure. 3502-014 Page 5 of 21 | EFTA_00001438 | | :--- | :--- | This is a simple Markdown document with no headings, paragraphs, or tables. It contains just one line of text. EFTA00156779 1150 Violence Against Women 21(9) A script was available online as soon as the participant accessed the study link via REDCap Survey. Potential participants were screened online. Once an exclusion criterion was met, no other information was gathered. If deemed eligible, the subjects were provided with informed consent. After they read and acknowledged understanding by answering three questions covering material contained within the consent, the participant was allowed to proceed to access the study packet online. All information was collected via participant self-report. Participants were allowed to save responses online and return to finish the packet at their convenience. If the participant elected to log off and log back on to complete the study packet, the first screen reiterated the fact that nobody would be able to contact them for any reason. All data collected from participants were kept online. Questions were created by the principal investigator to determine whether the participant had previously (before answering this survey) disclosed that they had an unwanted sexual experience (completed rape), to whom they disclosed if they affirmed disclosure, and their preferences for follow-up. The questions relative to prior disclosure were contained in the beginning of the survey and were as follows: (a) Is this the first time you are disclosing that you had an unwanted sexual experience? Yes/no; (b) If you have told one or more people about this incident, whom did you tell? Please check all that apply; (c) If you checked “other” above, please fill in the relationship you have with the person you told about the most recent incident of unwanted sexual contact. Please do not put in a personal name but only identify your relationship with that person. Branching logic was employed, so that participants would only see Questions 2 and 3 if they answered “no” to Question Number 1. The question relative to follow-up preferences was contained toward the end of the survey and stated: For experiences such as the one I had, I feel more comfortable disclosing the situation: (a) online anonymously with no way for anyone to re-contact me, (b) online with a way that someone could follow-up with me in the future, (c) in person face-to-face, (d) on the telephone anonymously with no way for anyone to re-contact me, and (e) on the telephone with a way that someone could follow-up with me in the future. ## Sample This study included a convenience sample of 242 adult female victims of completed rape drawn from the population of females aged 18-64 in the United States and internationally. The subjects recruited into the study met the following inclusion criteria: (a) between the ages of 18-64, (b) ≤5 years since their most recent incident of rape, (c) ability to understand English, (d) no recent report of psychosis, (e) the ability to complete study instruments, and (f) female gender. Individuals were excluded if they met the following exclusion criteria: (a) most recent incidence of rape happened while participant was <18 years of age, (b) unable to understand the informed consent as evidenced by incorrectly answering three, and (c) questions designed to determine understanding the content of the study. Determination of an incidence of rape for inclusion was screened for using the Sexual Experiences Scale Short Form Victimization (Koss et al., 2007; Koss & Gidycz, 1985). 3502-014 Page 6 of 21 EFTA_00001439 EFTA00156780 Carretta et al. 1151 The sample was recruited in a variety of ways. First, recruitment was accomplished using email messages inviting participation in the survey. Emails were sent out through ResearchMatch as part of an opt-in list of individuals who had previously given their contact information for that purpose, as well via the clinical trials registry maintained by the sponsoring University. A description of the study along with a link was provided in the email. This link led to the dedicated study website, specifically designed to provide comprehensive information on the study, a toll-free telephone number to contact a live person if the potential participant so chose, and a link that would provide direct access to the study itself. Responses went directly into the REDCap survey system, designed and maintained by Vanderbilt University. This provided a tracking mechanism for responses, prevented the release of any information and/or data to an outside server, and increased response rates. Additional methods included the following: informative advertisements placed on national screening and online support websites, and in domestic violence shelters, rape crisis centers, offices of psychiatrists, and psychotherapists, local emergency room departments, primary care office lobbies, and public venues such as college bulletin boards, grocery stores, bathroom stalls, libraries, social media sites such as Facebook, and police departments. Other techniques included posting informative public service announcements on local radio channels and direct marketing of the study online to organizations in which the principal investigator is affiliated. Specialist health care providers, home health agencies, church groups, and support groups may also have referred participants to the study based on flyers supplied to their organizations. Strategies to enhance participant recruitment and retention included ensuring anonymity, with no way to link any participant to any particular response, and the ability to complete the study packet in more than one sitting. All information was collected via participant self-report, and Institutional Review Board (IRB) approval was given by the University of Medicine and Dentistry of New Jersey. Participants were allowed to save responses online and return to finish the packet at their convenience. All data collected from participants were kept online. Answers to survey questions determined whether the participant had previously (before answering this survey) disclosed that they had an unwanted sexual experience, to whom they disclosed if they affirmed disclosure, and their preferences for follow-up. Branching logic was employed, so that participants would only see Questions 2 and 3 if they answered “no” to Question Number 1. Also used in this data analysis were questions related to current mental health and three standard measurement tests for anxiety (State Trait Anxiety Inventory—trait portion only), depression (Beck Depression Inventory II; BDI-II), and PTSD (Posttraumatic Stress Diagnostic Scale; PDS). Participants were also asked questions related to their insurance status at the time of their most recent assault and questions related to medications they currently take for anxiety, depression, or sleep disturbance. There was no compensation for study completion. A total of 384 completed the study consent form and at least some portion of the study. Of those, 242 (63%) completed all the study instruments sufficiently for inclusion in the analysis of the research questions. There were no statistically significant differences between the completers and non-completers on any demographic factor. 3502-014 Page 7 of 21 EFTA_00001440 EFTA00156781 1152 Violence Against Women 21(9) The 242 participants ranged from 18-56 years in age with a median age of 27 years (25th-75th Interquartile Range (IQR): 23.8/33.3). The sample was primarily Caucasian ($n = 218$, 90%), with the remaining identifying themselves as African American ($n = 18$, 7%) or other ($n = 6$, 3%). The majority of the sample reported being single/not partnered ($n = 185$, 76%). Participants lived in all regions of the United States; Northeast ($n = 41$, 17%), Southeast ($n = 18$, 8%), Midwest ($n = 40$, 17%), South ($n = 97$, 41%), and West ($n = 20$, 9%) with 9% reporting living outside the United States ($n = 20$). Although the majority of the sample was well educated and reported having at least a bachelor’s ($n = 100$, 41%) or a master’s degree ($n = 54$, 23%), they were less affluent, with 70% of the sample reporting incomes of US$60,000 or less (range US$100,000). The majority of the sample ($n = 171$, 71%) reported having no children or having any religious preference ($n = 146$, 60%). Those reporting having non-governmental health insurance Point of Service (POS), Preferred Provider Organization (PPO), Health Maintenance Organization (HMO) was slightly higher ($n = 143$, 59%) than those having insurance that was government subsidized (Medicaid, Medicare, MediCal; $n = 99$, 41%). The majority of the sample (96%) denied living with their abuser ($n = 232$). ## Analysis Frequency distributions summarized the number of participants who reported nondisclosure before the survey and those who had disclosed prior to the survey. Cross-tabulations were constructed to determine the percentages of individuals who affirmed first-time disclosure with reporting of follow-up preferences. The chi-square test of independence was used to test for differences in the distributions. ## Findings The demographic characteristics of those citing first-time disclosure and those reporting having disclosed previously are summarized in Table 1. Statistically significant differences between the two groups were observed in ages of the participants and presence of children. A higher proportion of those stating first-time disclosure reported having children ($n = 24$ of $58, 41\%$) than in the group citing prior disclosure ($n = 47$ of $137, 34\%$). Those who admitted to first-time disclosure were, on average, older than those who cited previously disclosing. While not statistically significant ($p = .055$), within the group citing first-time disclosure approximately half ($n = 30$ of $58, 51.7\%$) reported having some form of governmental subsidized insurance, while a considerably smaller respective proportion ($n = 69$ of $184, 37.5\%$) was seen in the group citing prior disclosure. There were no statistically significant differences between the first-time and nonfirst-time respondents in terms of type of unwanted experience, nor for relationship between the perpetrator and victim (see Table 2). Finally, there was no statistically significant difference between the respondent groups in terms of type of follow-up preferred ($p = .153$). The majority of participants, 3502-014 Page 8 of 21 | EFTA_00001441 | | :--- | :--- | This is a simple Markdown document with no headings, paragraphs, or tables. It contains just the text "EFTA_00001441". EFTA00156782 Carretta et al. 1153 Table 1. Descriptive Statistical Summaries of Consented Individuals First Time Disclosing and Not First Time Disclosing Study Variables—Demographics.
First time disclosed(n=58)Not first disclosure(n=184)
Characteristicn(%)n(%)p value
Race.582
Caucasian51(87.9)167(90.8)
African American6(10.3)12(6.5)
Other1(1.7)5(2.7)
Education.774
12th grade/GED5(8.6)10(5.4)
Some college—Did not graduate13(22.4)34(18.5)
2-year degree7(12.1)19(10.3)
4-year degree22(37.9)78(42.4)
Graduate degree11(19.0)43(23.4)
Marital status.635
Single/not partnered43(74.1)142(77.2)
Married/partnered15(25.9)42(22.8)
Residence area.702
Northeast7(13.0)34(18.7)
Southeast6(11.1)12(6.6)
Midwest10(18.5)30(16.5)
South24(44.4)73(40.1)
West3(5.6)17(9.3)
Outside the United States4(7.4)16(8.8)
Children.021
Yes24(41.4)47(25.5)
No34(48.6)137(74.5)
Health insurance type.055
Governmental(Medicare,Medicaid,etc.)30(51.7)69(37.5)
Non-governmental28(48.3)115(62.5)
Religious preference.218
Roman Catholic14(24.1)24(13.0)
Protestant11(19.0)36(19.6)
Jewish3(5.2)4(2.2)
Muslim0(0.0)1(0.5)
Buddhist0(0.0)3(1.6)
No preference30(51.7)116(63.0)
Annual household income.553
Less thanUS$25,00014(24.1)44(25.1)
US$26,001-US$40,00012(20.7)48(27.4)
US$40,001-US$60,00014(24.1)29(16.6)
(continued) Page 9 of 21 | EFTA_00001442 | | :--- | :--- | This is a simple Markdown document with no headings, paragraphs, or tables. It contains just the text "EFTA_00001442". EFTA00156783 1154 Violence Against Women 21(9) Table I. (continued)
CharacteristicFirst time disclosed(n=58)Not first disclosure(n=184)p value
n(%)n(%)
US$60,001-US$80,0001(1.7)12(6.9)
US$80,001-US$100,0006(10.3)13(7.4)
More than US$100,0006(10.3)14(8.0)
Prefer not to answer5(8.6)15(8.6)
Currently lives with abuser.324
Yes4(6.9)7(3.8)
No54(93.1)177(96.2)
Median(IRQ)Median(IRQ)
Age.003
Age(years)31.5(23.0,42.3)27.0(24.0,32.0)
Note. With the exception of age, $ \chi^{2} $ Tests of Independence were used to test for differences between the respondent groups. A Mann-Whitney Test was used for that respective test for age. GED = general education diploma. IRQ = Interquartile Range. Table 2. Descriptive Statistical Summaries of Consented Individuals First Time Disclosing and Not First Time Disclosing—Type of Unwanted Experience and Perpetrator/Victim Relationship.
First time disclosed(n=58)Not first disclosure(n=184)
Characteristicn(%)n(%)p value
Type of unwanted experience.186
Forcible rape28(48.3)117(63.6)
Pressured sex11(19.0)29(15.8)
Sex stress4(6.9)7(3.8)
Multiple types15(25.9)31(16.8)
Perpetrator/victim relationship.091
Intimate partner26(44.8)71(38.6)
Non-intimate known28(48.3)78(42.4)
Stranger4(6.9)35(19.0)
Total58(100)184(100)
Note. $\chi^2$ Tests of Independence were used to test for differences between the disclosure groups. both those who had previously disclosed ($n = 123, 70.3\%$) and those had who admitted to first-time disclosure ($n = 46, 79.3\%$), reported they preferred online follow-up to both the face-to-face and telephone options (Table 3). 3502-014 Page 10 of 21 EFTA_00001443 EFTA00156784 Carretta et al. 1155 Table 3. Descriptive Statistical Summaries of Consented Individuals First Time Disclosing and Not First Time Disclosing Study Variables—Disclosure and Follow-Up.
CharacteristicFirst time disclosed(n=58)Not first disclosure(n=175)p value
n(%)n(%)
Follow-up preferences.153
Online46(79.3)123(70.3)
Telephone6(10.3)14(8.0)
Face-to-face6(10.3)38(21.7)
Note: $\chi^2$ Tests of Independence were used to test for differences between the disclosure groups. Table 4. Descriptive Statistical Summaries of Consented Individuals First Time Disclosing and Not First Time Disclosing Study Variables—Medication Use.
First time disclosed(n=58)Not first disclosure(n=184)p value
Characteristicn(%)n(%)
Depression medication.571
Yes16(27.6)58(31.5)
No42(72.4)126(68.5)
Anxiety medication.574
Yes13(22.4)48(26.1)
No45(77.6)136(73.9)
Sleep medication.635
Yes15(25.9)42(22.6)
No43(74.1)142(77.2)
Note. $\chi^2$ Tests of Independence were used to test for differences between the disclosure groups. ## Post Hoc Analysis of Disclosure Groups Current use of medication for depression, anxiety, and sleep disturbances for those who had previously disclosed and those who had not is summarized in Table 4. There were no statistically significant differences in the rates of use of the types of medication between the groups. Follow-up with providers for physical and emotional injuries was evaluated for those who had previously disclosed and those who had not (summaries in Table 5). Results indicated that there was a statistically significant difference between the groups in rates of seeking treatment for emotional injuries with both medical providers ($p = .003$) and non-medical therapist/counselors ($p < .001$). In both cases, a higher proportion of those admitting to first-time disclosure reported never seeking treatment 3502-014 Page 11 of 21 EFTA_00001444 EFTA00156785 1156 Violence Against Women 21(9) Table 5. Descriptive Statistical Summaries of Consented Individuals First Time Disclosing and Not First Time Disclosing Study Variables—Rape Resource Utilization.
CharacteristicFirst time disclosed(n=58)Not first disclosure(n=184)p value
n(%)n(%)
Medical practitioner for physical injuries.055
Never53(91.4)138(75.0)
Once or twice4(6.9)38(20.7)
3-5 times1(1.7)4(2.2)
More than 5 times0(0.0)4(2.2)
Total58(100)184(100)
Medical practitioner for emotional injuries.003
Never52(89.7)119(64.7)
Once or twice2(3.4)24(13.0)
3-5 times2(3.4)12(6.5)
More than 5 times2(3.4)29(15.8)
Total58(100)184(100)
Non-medical therapist/counselor for emotional injuries<.001
Never46(79.3)78(42.4)
Once or twice5(8.6)21(11.4)
3-5 times4(6.9)12(6.5)
More than 5 times3(5.2)73(57.8)
Total58(100)184(100)
Lawyer for injuries.218
Never56(96.6)160(87.0)
Once or twice1(1.7)9(4.9)
3-5 times0(0.0)5(2.7)
More than 5 times1(1.7)10(5.4)
Total58(100)184(100)
Called police.009
Yes5(8.6)45(24.5)
No53(91.4)139(75.5)
Total58(100)184(100)
for emotional injuries from a medical provider (52 of 58, 90%) or a therapist/counselor (46 of 58, 79%) than those who had previously disclosed (65% and 42%, respectively). The overwhelming majority of those citing first-time disclosure ($n = 46$, 70.3%) reported that they had never seen a non-medical therapist/counselor for emotional injuries, whereas the majority (57.8%) of those citing prior disclosure reported seeing a therapist/counselor more than 5 times. As expected, given that one group cited no prior disclosure, there was a statistically significant difference in reporting the assault to police. 3502-014 Page 12 of 21 EFTA_00001445 EFTA00156786 Carretta et al. 1157 Table 6. Descriptive Statistical Summaries of Consented Individuals First Time Disclosing and Not First Time Disclosing Study Variables—Blame.
CharacteristicFirst time disclosed(n=58)Not first disclosure(n=184)p value
n(%)n(%)
Feels good about self.206
Yes26(44.8)100(54.3)
No32(55.2)84(45.7)
Feels abuse was their fault.001
Yes37(63.8)72(39.1)
No21(36.2)112(60.9)
Table 7. Differences Between Groups for Disclosure Follow-Up Preferences and Outcomes (Depression and Anxiety).
First time disclosed (n = 58)Not first disclosure (n = 184)F(df = 1,240)p valueη²
M (SD)M (SD)
BDI-II22.7 (15.0)21.1 (13.8)0.53.466<.01
STAI-Y (trait)52.4 (12.3)50.9 (14.1)0.54.465<.01
Note. MANOVA was used to test for differences in depression and anxiety among the groups. Wilks's lambda = .998, F(1, 240) = 0.286, p = .751. BDI-II = Beck Depression Inventory II. STAI-Y = State Trait Anxiety Inventory–Form Y. The difference in the rates of feeling good about oneself between the groups was not statistically significant (45% vs. 54%), yet there were statistically significant differences in the reported belief that the abuse was their fault with 63.8% of those reporting first-time disclosure believing that the abuse was their fault versus 39.1% of those with prior disclosure (Table 6). Further analysis was completed to determine whether differences exist in rape trauma presentation/symptomatology (depression, anxiety) and diagnosis of PTSD among women who have and have not disclosed the event. Descriptive summaries of the two groups are presented in Table 7. There were no statistically significant differences between the groups in terms of depression ($p = .466$) or anxiety ($p = .465$; Table 7). In addition, there were similar proportions of those who met the criteria for a diagnosis of PTSD ($p = .481$) within each of the groups (Table 8). Most respondents (76%) in this study had disclosed an unwanted sexual experience that happened within the previous 5 years. Of clinical interest, however, 24% had never disclosed until asked on this survey. Of the 184 who had previously disclosed, persons told included friends; medical professionals; family members; spouse or partner; police, coworkers; clergy; academic staff; domestic violence; rape crisis, and hotline staff; with one woman disclosing to a local newspaper. 3502-014 Page 13 of 21 EFTA_00001446 EFTA00156787 1158 Violence Against Women 21(9) **Table 8. Summaries for Disclosure and PTSD.**
First time disclosed(n=58)Not first disclosure(n=179)
n(%)n(%)p value
PDS.481
With PTSD29(50)80(45)
Without PTSD29(50)99(55)
Note. PTSD = posttraumatic stress disorder; PDS = Posttraumatic Stress Diagnostic Scale. ## Discussion The decisional conflict around disclosing a rape either to others or to police has a long history. That almost a quarter (24.5%) of this sample reported to police is closer to the 19% suggested by Tjaden and Thoennes (2006) but less than the 47% reported by Catalano et al. (2009) (Frueh et al., 2000). Not only is rape a seriously underreported crime, it is also an undertreated crime. Only a small number of women seek treatment, with one fifth or 21% sought treatment from a medical provider for an emotional injury and almost half (49%) from a counselor or a therapist. This pattern is somewhat different than that suggested by Amstadter, McCauley, Ruggiero, Resnick, and Kilpatrick (2008), who reported that 38% sought treatment from a medical professional and 54% from a mental health specialist. Medication use in our study was low in contrast to Smith et al. (2005) who reported that visits to providers to obtain prescriptions for anti-depressants rose dramatically between 1995 and 1996 from 13.8 visits to 35.5 visits (Smith et al., 2005). Plichta and Falik (2001) report a significant relationship between intimate partner violence and taking medication for depression and anxiety (Plichta & Falik, 2001). We found almost identical rates for those reporting crimes in this study perpetrated by an intimate partner ($n = 97$) and non-intimate known ($n = 106$). There were significantly less reported incidents by a stranger in this sample ($n = 40$). This finding agrees with most prior studies (Basile, Chen, Black, & Saltzman, 2007; Johnson, Zlotnick, & Perez, 2008; Littleton, 2007, 2010; Plichta & Falik, 2001; Tjaden & Thoennes, 2006) but disagrees with the findings by other researchers (Frazier, 2003; Resick, Jordan, Girelli, Hutter, & Marhoefer-Dvorak, 1988), where about half of the participants were found to have been raped by a stranger. There were 30% ($n=74$) of women who reported taking anti-depressant medication with 25% ($n=61$) taking medication for anxiety and 24% ($n=57$) for sleep disturbances. These findings could be due to the low rate of treatment for emotional injuries. However, almost half (48%) reported that they do not feel good about themselves since their most recent abuse incident, and 45% said they feel the most recent incident of abuse was their fault. This pattern is consistent with other studies specific to rape, citing self-blame as significantly related to psychological distress (Frazier, 2003; Koss, Figueredo, & Prince, 2002; Najdowski & Ullman, 2009). 3502-014 Page 14 of 21 EFTA_00001447 EFTA00156788 Carretta et al. 1159 There are several limitations that are noteworthy. First, the sample did not use random sampling and was comprised of a convenience sample of adult participants who self-reported one or more incidents of rape within the past 5 years. This approach limits generalizability of the study to those participants who were aware of the study based on the limited recruitment mechanisms employed and decided to participate, and the findings cannot necessarily be generalized to survivors of other possible traumatic experiences. Second, this study was a cross-sectional design, and therefore no causal inferences can be made. Third, the instruments used in this study were not necessarily specific to rape, and thus may have more limited ability to assess certain symptoms or outcomes specific to an experience of rape such as fear of sexual contact. Another example exists related to the measure of PTSD. Although this measure mentions the concept of rape as one possible traumatic experience, it cannot be determined by virtue of the questionnaire if the diagnosis of PTSD is solely or most significantly related to the rape experience. Furthermore, the measure of PTSD was distinctly different from the measures of depression and anxiety, in that the measure of PTSD did not measure the continuum of symptoms. It is likely that more and stronger associations would have been found if the measure had allowed for measurement on a continuum. Fourth, the preference for online follow-up (as opposed to face-to-face or telephone) may be an artifact of the chosen methodology. That is to say, women more likely to self-select for participation in an online study regarding sexual experiences and disclosure may also be more likely to (a) be frequent Internet users and (b) prefer e-based communication. Finally, this study did not ask individuals the reason they chose to disclose to certain persons. Ullman and Filipas’s (2001) study on 323 sexual assault victims reported that disclosing the sexual assault to more persons was related to more negative and positive reactions. Given the equivocal findings, it is strongly recommended that this question be asked in future research. ## Rape-Related Dynamics Carter-Snell and Jakubec (2013) conducted an in-depth analysis of 100 data-based articles (of a total 2,116) on interpersonal violence to determine the relative impact of selected risk and resiliency factors pertaining to mental health impacts. As mental health counselors can do little about risk factors after the assault (e.g., severity of violence, prior trauma), our focus is on secondary prevention and identification of resilience factors. Our study found that irrespective of disclosure, victims of rape do not readily seek treatment for psychological or symptom remediation when we know that silencing themselves in addition to the experience of trauma is highly correlated with clinical depression (Jack, 1991; Jack & Ali, 2010). In trying to explain this lack of victim help-seeking behavior, several researchers have put forth suggestions. Koss (1994) posited that interviewer effects and other factors such as others overhearing an interview may be responsible for victims’ unwillingness to disclose (Koss, 1994). Campbell, Dworkin, and Cabral (2009) focus on the negative mental health effects 3502-014 Page 15 of 21 EFTA_00001448 EFTA00156789 1160 Violence Against Women 21(9) of rape, instead of the recovery aspect, and consider the role of personality characteristics, preexisting mental health conditions, biological/genetic factors, use of force and/or threats, and substance use not examined in previous models (Campbell et al., 2009). Campbell, Wasco, Ahrens, Sefl, and Barnes (2001) and Ledray (1998) suggest that the avoidance of treatment could be due to fear that counseling could result in (a) further incidence of abuse (e.g., perpetrator becomes aware that victim is disclosing) or (b) re-traumatization based on having to recount the story over and over again (e.g., to multiple medical practitioners, law enforcement; Campbell et al., 2001; Ledray, 1998). Jack (1991) would support the notion that women in violent and non-violent situations struggle to be direct, open, and honest with what they need and feel because of the more dangerous feeling of losing relationships with others if they did so. The researchers argue that the impact of disclosure by their autonomous voice is integral to victims’ post-assault psychological healing, and that should victims of rape blame themselves, they may not disclose the event to anyone. They continue to suggest, as does Kilpatrick et al. (1992), that failure to disclose is probably resulting in inadequate treatment and that failure to disclose then denies them opportunities for support (Grohol, 2011). A major finding in this study was the importance of self-blame as diagnostic of the non-disclosing group. Self-blame reflects a psychosocial mechanism of self-criticism and low self-evaluation in which the individual accepts personal responsibility for negative events. Janoff-Bulman’s (1979) classic study of self-blame in rape victims distinguishes two types of self-blame—behavioral and characterological. Behavioral self-blame is control related, involves attributions to a modifiable source (one’s behavior), and is associated with a belief in the future avoidance of a negative outcome. Characterological self-blame is esteem related, involves attributions to a relatively non-modifiable source (one’s character), and is associated with a belief in personal deservingness for past negative outcomes. Self-blame is another way of translating Jack and Dill’s (1992) findings that women often put others before themselves (care as self-sacrifice). In her study of 38 rape crisis centers, behavioral self-blame, and not characterological self-blame, emerged as the most common response of rape victims to their victimization, suggesting the victim’s desire to maintain a belief in control, particularly the belief in the future avoidability of rape. Given that our study found that the percentages of those with and without PTSD in both the disclosure and non-disclosure groups were almost identical, we suggest attention be given to the power of self-blame, secrecy, and non-disclosure as self-protective mechanisms. The lack of significant differences between the disclosure groups suggests that rape trauma is present irrespective of disclosure, and that disclosure in itself is not cathartic to the point that rape survivors experience symptom remission. But self-blame appears to be an incapacitating factor in the recovery process. This self-blame finding could be attributed to the fact that those who have previously disclosed may have sought professional treatment, and thus may have worked toward resolution of self-blame. 3502-014 Page 16 of 21 EFTA_00001449 EFTA00156790 Carretta et al. 1161 ## Implications for Practice Carter-Snell and Jakubec (2013) argue that effective secondary prevention of adverse mental health consequences following rape depends upon the identification of resilience factors. This would allow the professionals to draw on these strengths to promote the client and to shape interventions in a manner informed by evidence. An example would be to help the individual to reframe events to reduce self-blame, or to identify individuals or agencies that may provide positive reactions and supportive resources. In addition, Carter-Snell and Jubenec (2013) recommend studying the mental health impact of education programs for the community, police, and health professionals related to the use of responses such as psychological first aid and positive responses to rape disclosures. Rape involves concern regarding disclosure, secrecy and self-silencing, stigma and the organizing emotion of self-blame. In a sense, this requires counselors to assume a role as a clinical detective in assessing for the red flags with non-disclosing clients. Inquiry about any type of unwanted sexual experience needs to be considered of all clients with the knowledge that it may take some time for an affirmative answer to surface. For those clients who do disclose, we recommend that this issue of disclosure and self-blame be part of a counseling plan that is aimed at helping victims talk through the positives and negatives of disclosure and to help the victim predict those they tell (or have told) will be supportive and understanding, or blaming. The steps that counselors and therapists can take for counseling the victim on the issue of self-disclosure include the following: 1. Gather information from the victim to help make a prediction whether those told will be supportive or not. Inquire about the person's prior reaction to stressful news. 2. Have the victim predict the person's reaction. 3. Weigh the advantages of telling with the disadvantages of telling. 4. Support the victim's decision whichever side she or he wishes to take. Talk through what is anticipated in terms of support as well as if the person told turns out to blame or discredit the victim. Be sure the victim can handle both reactions. 5. Request that the victim report back the reaction to the counselor to provide support for whichever way the reaction went. Additional counseling will be needed if the person blamed the victim and was not supportive. ## Summary Disclosure of unwanted sexual experiences remains a major problem. To date, there is no study within the past 10 years that has attempted to update incidence and prevalence. Moreover, the reports that do exist present divergent findings. To pursue development of studies aimed at testing prolific treatment interventions, we must first glean 3502-014 Page 17 of 21 EFTA_00001450 EFTA00156791 1162 Violence Against Women 21(9) a more accurate and concrete understanding of the true depth of the number of survivors and also begin to identify more acceptable methods for disclosure. Web-based anonymous surveys have demonstrated effectiveness in other populations. The finding that regardless of the disclosure pattern, the majority of respondents supported online counseling was impressive. The overwhelming majority of participants in both groups cited that online follow-up was preferred to either telephone or face-to-face contact. Brief and colleagues (2013) were able to successfully recruit 600 Operation Enduring Freedom (OEF) and Operation Iraqi Freedom (OIF) veterans into an alcohol and PTSD treatment study online in about 6 weeks. The outcomes showed favorable effects on (a) drinking days, (b) percent heavy drinking days, (c) average drinks per drinking day, and (d) PTSD symptomatology. The advantage of this approach is the potential for an incredible reach to those in rural areas, to those unable or unwilling to combat the stigma, and to those who live in areas with few mental health resources. Given this study's findings that regardless of disclosure pattern, individuals prefer the use of the Internet to traditional counseling modalities. Thus, it is reasonable to assert that this method could be optimal for providing the most cohesive and accurate estimates to date from a broad, diverse population. ## Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. ## Funding The author(s) received no financial support for the research, authorship, and/or publication of this article. ## References 3502-014 Page 18 of 21 | EFTA_00001451 | | :--- | :--- | This is a simple Markdown document with no headings, paragraphs, or tables. It contains just one line of text. EFTA00156792 Carretta et al. 1163 Catalano, S. M. (2006). *Criminal victimization, 2005* (NCJ 214644). Washington, DC: Bureau of Justice Statistics, Office of Justice Programs. 3502-014 Page 19 of 21 EFTA_00001452 EFTA00156793 1164 Violence Against Women 21(9) 3502-014 Page 20 of 21 EFTA_00001453 EFTA00156794 Carretta et al. 1165 3502-014 Page 21 of 21 | EFTA_00001454 | | :--- | :--- | This is a simple Markdown document with no headings, paragraphs, or tables. It contains just the text "EFTA_00001454". EFTA00156795