Disclosure of Child Sexual Abuse by Adolescents

Abstract

This qualitative study aimed to study the process of disclosure by examining adolescents from the general population who had experienced child sexual abuse (CSA). Twenty-six sexually victimized adolescents (23 girls, 3 boys; age: 15-18 years) participated in a qualitative face-to-face in-depth interview on different aspects of disclosure. A qualitative content analysis was conducted following Mayring and using the qualitative data analysis program Atlas.ti. In addition, quantitative correlation analyses were calculated to identify factors associated with disclosure. Less than one third of participants immediately disclosed CSA to another person. In most cases, recipients of both immediate and delayed disclosure were peers. More than one third of participants had never disclosed the abuse to a parent. Main motives for nondisclosure to parents were lack of trust or not wanting to burden the parents. Factors that correlated positively with disclosure were extrafamilial CSA, single CSA, age of victim at CSA, and having parents who were still living together. Negative associations with disclosure were found for feelings of guilt and shame and the perpetrator’s age. Many adolescent survivors of CSA have serious concerns about disclosure to their parents and consider friends as more reliable confidants. These findings have two main implications for prevention: (1) In order to facilitate disclosure to parents, the strengthening of the child–parent relationship should be given specific attention in prevention programs, and (2) prevention programs should aim at teaching adolescents how they can help a victim if they become a recipient of disclosure.

Keywords

child sexual abuse disclosure adolescents sexual maltreatment

It is widely acknowledged that a child’s self-disclosure of sexual abuse is a prerequisite for intervention and the provision of adequate support to the child (Paine & Hansen, 2002). It is further believed that early disclosure decreases the risk of serious long-term consequences for mental health and the likelihood that the perpetrator will victimize other children (Arata, 1998; Fontes, 1993). However, research has shown that disclosure is not always followed by a stop of sexual victimization and that a considerable proportion of children do not receive the necessary support to cope with the abuse (e.g., Hershkowitz, Lanes, & Lamb, 2007; Roesler & Wind, 1994). Differences in the provision of support may explain why results on the relationship between disclosure and mental health outcomes are inconsistent (Arata, 1998). Some studies found a positive effect of disclosure (e.g., Arata, 1998), other studies found no association (e.g., Ruggiero et al., 2004), and other findings suggested that disclosure can even have negative repercussions on mental health (Berliner & Conte, 1995; Lamb & Edgar-Smith, 1994).

Most studies on the disclosure of child sexual abuse (CSA) either investigated retrospective reports by adults (London, Bruck, & Ceci, 2005; Roesler & Wind, 1994) or examined alleged victims who were evaluated in forensic or clinical interviews (e.g., Devoe & Coulborn Faller, 1999; Malloy, Lyon, & Quas, 2007). With both of these approaches there are major methodological problems. Whereas retrospective accounts by adults are believed to underestimate disclosure rates due to recall bias, studies with children who undergo forensic or clinical evaluation examine cases reported to authorities, which comprise only a minority of all cases of sexually victimized children (London, Bruck, Wright, & Ceci, 2008). Furthermore, studies with alleged children are unsuitable for estimating disclosure rates, since most of these children disclose CSA prior to the referral. What these studies usually focus on instead are characteristics of the disclosure process before or during the investigative interview (e.g., Devoe & Coulborn Faller, 1999; Hershkowitz et al., 2007).

Studies with sexually victimized adolescents recruited from the general population have been suggested to be the most accurate way to examine disclosure of CSA (Crisma, Bascelli, Paci, & Romito, 2004; London et al., 2008; Priebe & Svedin, 2008). Because the CSA is closer in time, recall bias in adolescent samples can be expected to be less than in adult samples. Furthermore, compared to younger children, adolescents have the ability to participate in studies without their parents’ knowledge and consent. The inclusion of children who have been victimized by a parent or who have not yet disclosed the abuse is only possible if parental consent is not required for study participation. However, despite their importance, there are few studies on disclosure that included adolescents (e.g., Crisma et al., 2004; Kogan, 2004; Priebe & Svedin, 2008). Whereas these studies have reported disclosure rates as high as 80%, results also indicated that most victims disclosed only after a considerable delay (e.g., Kogan, 2004). Older age and female gender of victim, perpetrator being a stranger, extrafamilial CSA, the victim’s parents living together, and positive parental bonding were shown to facilitate disclosure (Kogan, 2004; Priebe & Svedin, 2008). Results with regard to recipients of disclosure are inconsistent. Whereas Crisma et al. and Priebe and Svedin reported that peers most often become confidants of CSA, Kogan found that parents and friends were about equally important as recipients of disclosure. Other recipients of disclosure identified were siblings, professionals, and other relatives. Whether young survivors not only disclose CSA but also report the abuse to the police has not been investigated among adolescents . However, retrospective studies with adults showed consistently that reporting to the police is an exception rather than a common step in the disclosure process (e.g., Hanson, Resnick, Saunders, Kilpatrick, & Best, 1999).

Although the CSA literature has focused considerable attention on motives for nondisclosure or delayed disclosure of CSA (Paine & Hansen, 2002), there are only few studies investigating this issue. A qualitative telephone interview study by Crisma et al. (2004) is the only investigation that examined impediments to disclosure reported by adolescents recruited from the general population. The main reasons for nondisclosure to parents included the child’s feeling that the parents could not be relied on, fear of being blamed or not believed, feelings of shame, and not wanting to burden parents. Smith and Cook (2008) interviewed young adults and found that denying (e.g., wanting to forget about CSA), fear of parental sanctions, or fear that parents would report the CSA to the police were also common motives for nondisclosure toward parents. Similar motives for nondisclosure were reported by Jensen, Gulbrandsen, Mossige, Reichelt, and Tjersland (2005), who analyzed psychotherapy sessions with sexually victimized children and their parents, and Roesler and Wind (1994), who examined adult women. However, previous studies focused mainly on nondisclosure to parents and did not specifically examine the children’s reasons for nondisclosure to other people known to them, such as friends, relatives, or professionals.

Due to a lack of comprehensive studies on disclosure in adolescent samples, we conducted qualitative face-to-face in-depth interviews with sexually victimized adolescents from the general population. We chose a qualitative research design, which we combined with quantitative analyses. Qualitative interviews have been shown to be an important subsidiary research method to quantitative CSA research (e.g., Lamb & Edgar-Smith, 1994; Smith & Cook, 2008) because they enable a more direct and deeper approach to the survivors’ subjective experiences than quantitative surveys (Crisma et al., 2004). This study focused on the following research questions:

Research Question 1: How many of the sexually victimized adolescents interviewed had disclosed the abuse to someone or reported the abuse to the police?

Research Question 2: Whom do sexually victimized children and adolescents inform about the abuse?

Research Question 3: What are children’s and adolescents’ motives for not disclosing or delaying disclosure?

Research Question 4: Are there any factors that are associated with disclosure and reporting, such as the characteristics of the abuse (e.g., relationship to perpetrator)?

Method

Procedure

The study was approved by the local research ethics committee. Numerous efforts were made to recruit adolescents who had experienced CSA, including (a) placing the study link on 18 websites for adolescents and victims of sexual assaults; (b) distributing study flyers in school classes that participated in an associated school survey on the prevalence of CSA in Switzerland; (c) distributing study flyers to services that provide sex education, counseling services for adolescents and victims of sexual violence, the child protection team at University Children’s Hospital Zurich, pediatricians, local youth clubs, and self-defense classes; (d) publishing a call for study participation twice in a daily newspaper, and (e) placing our study link on internet-based social networks (such as Facebook).

The recruitment material (e.g., flyer, link on websites) contained a written description of several types of CSA (e.g., exhibitionism, rape). Adolescents were asked whether they had experienced one or more of the listed events or had experienced some other type of sexual violence or harassment. Furthermore, the study objectives and procedures were described and anonymous participation guaranteed. Adolescents were also informed that their travel expenses would be reimbursed and that they would receive two cinema tickets as a thank-you for their participation.

For practical reasons, recruitment focused on the Canton of Zurich,¹ Switzerland. The inclusion criterion for age was set to 15 to 18 years. At age 15 to 18, parental consent was not a prerequisite for participation based on Swiss law.

Adolescents who were interested in participating in the interviews were asked to contact one of the researchers (VS or MAL) via phone or email. If the authors were emailed, a telephone appointment was arranged. At the first telephone contact, adolescents were given comprehensive information about study participation, and this same information was sent in writing via mail or email.

Most interviews were conducted at University Children’s Hospital Zurich by one of the researchers (VS). Two participants preferred to be interviewed at home. Before the interview began, participants were informed about the interview procedure and reassured that they were allowed to take a break or stop the interview whenever they wanted. Participants then signed an informed consent form. The interviews lasted on average 2 hours (range: 1-3 hr). After the interview, participants were offered short-time counseling or a referral to a support service for victims of sexual assault to access psychosocial support if needed.

Participants

Twenty-six adolescents participated in the study: 14 responded to the call in the daily newspaper, 4 learned about the study via the study flyer, 3 saw the study link on a website of professional services, 3 were referred by the child protection team of the University Children’s Hospital Zurich, and 2 were encouraged to participate by acquaintances who knew of the study. Twenty-three (88.5%) participants were girls, and three (11.5%) were boys. The age of participants ranged from 15.4 to 18.3 years (M = 17.0 years). Twenty-two were Swiss, and four were of foreign nationality. Ten participants were still in school, 15 in an apprenticeship or other vocational training, and one was still looking for an apprenticeship position. Socioeconomic status (SES) scores were low for 7 participants (26.8%), middle for 15 participants (57.7%), and high for 4 participants (15.8%).

All participants contacted the researchers because of a single type of CSA that they had been subjected to. In all but one case, this was the most serious event of sexual violence that they had experienced. However, on average, participants had experienced 2.6 additional types of CSA. The most serious types of CSA experienced by participants ranged from sexual harassment to completed rape. More than half of the participants had experienced contact CSA without penetration, and more than one third had been raped (see Table 1). Eight participants had experienced intrafamilial CSA, and six had been sexually assaulted by a stranger. One half of the sexual assaults had been committed by adolescent perpetrators. All perpetrators were male, and the age of participants at the time of CSA ranged from 3 to 17 years (M = 11.7 years).

Table 1.

Types of Child Sexual Abuse (CSA) and Relationship to Perpetrator

	N
Type of CSA
Contact without penetration	14
Penetration	9
Attempted penetration	2
Noncontact	1
Singular CSA	9
Repeated CSA	17
Perpetrators
Male	26
Female	0
Unknown adolescent	6
(School) friend	5
Biological father	4
Partner of mother	3
Boyfriend	2
Friend of parents	2
Colleague at work	1
Uncle	1
Caretaker in children’s home	1
Unknown adult	1

Interview Structure and Measures

The first part of the interview comprised standardized questions and measures on family situation, sociodemographic data, sexual victimization, general and mental health, and feelings of guilt and shame. To collect the quantitative data presented in this article, the following standardized measures and questions were applied:

Sexual victimization

After the participants were asked to tell the interviewer what kind of CSA they had experienced, a German translation of the Sexual Assault Module of the Juvenile Victimization Questionnaire (JVQ; Hamby, Finkelhor, Ormrod, & Turner, 2004) was administered. This allowed us to collect detailed information about the experienced CSA in a standardized way and helped us to identify additional events of sexual violence that the adolescents had experienced.

The Sexual Assault Module of the JVQ consists of a checklist of seven different types of sexual victimization (sexual assault by known adult, nonspecific sexual assault, sexual assault by peer, rape, flashing/sexual exposure, verbal sexual harassment, and statutory rape). If a child experienced a certain type of sexual violence, the module asks additional questions on specific characteristics of the assault (e.g., number of times the child was victimized, the child’s age at beginning and end of the assault, and the child’s relationship to the perpetrator). The JVQ showed good reliability and validity in a U.S. national random sample of 10- to 17-year-old adolescents (Finkelhor, Hamby, Ormrod, & Turner, 2005).

We produced an authorized translation of the original English version of the JVQ following Mallinckrodt’s guidelines (Mallinckrodt & Wang, 2004). This procedure included the following steps: (a) two independent translations from English to German by native speakers of the target language, (b) creation of a German consensus version, (c) back translations of the consensus version into English by an independent native speaker, (d) the consensus German version and the back translated English version were reviewed by the original author of the JVQ, and (e) questions from the original author were reviewed to produce the final German version. Finally, the JVQ authors approved the final translated version.

Socioeconomic data

Socioeconomic status (SES) was calculated based on paternal occupation and maternal education on a 6-point scale. SES scores ranged from 2 to 12 points and were classified in three social classes: 2-5 = low SES, 6-8 = middle SES, and 9-12 = upper SES. This measure was proven to be a reliable and valid indicator of SES in Switzerland (Landolt, Nuoffer, Steinmann, & Superti-Furga, 2002). In addition, participants were asked about their educational training, nationality, and whether their parents were living together or were separated or divorced.

Questions on feelings of guilt and shame

For each of experienced event of sexual violence (as assessed by the JVQ), participants were asked whether they had ever felt guilty for or ashamed of the experienced abuse (yes/no).

The second part of the interview consisted of a qualitative half-standardized interview guide with questions on disclosure and received support. In this article, we present only data on or related to disclosure. Data on disclosure was collected by the following interview questions:

- Did you talk to anyone about the sexual abuse? Did you ask anyone for help?

- Who did you talk to? Who did you ask for help? (parents, peers, relatives, counseling services for victims of CSA, other persons?)

- Who knows about the sexual abuse? (parents, peers, relatives, counselor, other persons?)

- Have you reported the sexual abuse to the police?

- If you did not disclose the abuse/not ask for help, what were your reasons?

- Would you have liked to talk to other people about the abuse but you did not? Why?

- Would you have liked more or fewer people to have learned about the sexual abuse?

- Is there anything else that was important to you with regard to disclosure of the abuse? Is there anything else that you would like to mention?

Data Analyses

Qualitative analysis

After the half-standardized qualitative parts of the interviews were transcribed, a qualitative inductive content analysis following Mayring was conducted by using the qualitative data analysis software Atlas.ti version 5.2 (www.atlasti.com). Mayring’s qualitative research approach is one of the best-established qualitative methodologies in social research in the German-speaking countries in Europe. Based on the research questions, material in written form is analyzed by stepwise inductive building of categories (codes), to which statements in the texts are then assigned. The process of categorization and interpretation proceeds close to the material and is often not theory driven. After 10% to 50% of the material is analyzed, the categories are reevaluated and, if necessary, revised. They can also be summarized into larger categories (families). At the end of the categorization process, interrater reliability of categories is checked. If interreliability is satisfactory, quantitative analyses of categories can be conducted to test research questions (Mayring, 2008). It is the quantification of qualitative results that distinguishes Mayring’s methodological approach from other qualitative analysis techniques such as the interpretative phenomenological analysis by Smith, Flowers, and Larkin (2009), which focuses more on illustrative description and interpretation of key phenomena found in the material.

Figure 1 shows the process of inductive categorization and content analysis performed in this study. Disclosure was defined as an active process during which a participant told a confidant about CSA. If somebody learned about the CSA through someone other than the participant, it was not regarded as disclosure.

Figure 1.

Process of inductive categorization and content analysis in the study (Mayring, 2008)

In accordance with the research questions, the following aspects of disclosure were of interest:

Whether participants disclosed CSA or not, and whether they disclosed immediately after the abuse or only after a delay. Because aspects of disclosure were assessed in a qualitative way, the data did not allow a full breakdown of the time between CSA and disclosures. However, the interview data revealed whether participants disclosed the abuse to someone on the same day (within 24 hr) or after a longer time.

To whom participants disclosed the event (at initial disclosure and following initial disclosure).

Motives for delayed disclosure or nondisclosure.

Whether participants reported the CSA to the police.

The coding of the interviews was carried out by one researcher (VS). Following this process, two researchers (VS, MAL) jointly grouped codes for recipients and motives for disclosure into larger categories (families; see Tables 2 and 3). Figure 2 shows the process of summarization of codes into families for the category “recipient = parents.” Intercoder reliability was tested following the recommendations of Lombard, Snyder-Duch, and Bracken (2010): Two researchers (TM, US), neither of whom had been involved in the analysis or in the construction of the categories, were given a random sample of 20% of the statements from the interviews that had previously been coded by the first author. After they were instructed regarding the definition of the categories, they were asked to assign the statements to the categories. The three code assignment ratings (VS & MAL, TM, and US) achieved actual agreement of 100% with both Krippendorff’s alpha and Cohen’s kappa being 1.0, which indicated excellent intercoderreliability.

Table 2.

Rates of Disclosure by Recipients of Disclosure

	Immediate disclosure (N/%)	Delayed disclosure (N/%)	Disclosure in total^a (N/%)
Confidants
Peers	6 (23.1)	16 (61.5)	25 (96.2)
Parent(s)	5 (19.2)	11 (42.3)	16 (61.5)
Intimate partners	—	7 (26.9)	7 (26.9)
Professionals^b	—	6 (23.1)	6 (23.1)
Siblings	3 (11.5)	2 (7.7)	6 (23.1)
Interviewer	—	26 (100)	26 (100)
Teachers	—	5 (19.2)	5 (19.2)
Other relatives	1 (3.9)	3 (11.5)	4 (15.4)
Superiors	—	2 (7.7)	2 (7.7)
Clergy person	—	1 (3.9)	1 (3.9)

Total N is not always consistent with the sum of the two columns “immediate and delayed disclosure” because in some interviews time point of disclosure did not become clear.

(Mental) health professionals, social workers, counseling services (not included are psychotherapists to whom participants were referred after disclosure).

Table 3.

Motives for Nondisclosure or Delayed Disclosure by Groups of Eligible Recipients

Motives for delayed or nondisclosure to	Denying	So as not to burden person/person emotionally unstable	Lack of trust	Shame/stigmatization	No comprehension of what had happened	Fear of disbelief	Fear of perpetrator	Fear of parental sanctions	So as not to destroy one’s family	Intimacy
	(N/%)	(N/%)	(N/%)	(N/%)	(N/%)	(N/%)	(N/%)	(N/%)	(N/%)	(N/%)
Peers	—	3 (11.5)	—	—	—	—	—	—	—	1 (3.8)
Parent(s)	1 (3.8)	6 (23.1)	6 (23.1)	3 (11.5)	3 (11.5)	3 (11.5)	2 (7.7)	3 (11.5)	2 (7.7)	1 (3.8)
Siblings	—	—	—	—	—	—	—	—	1 (3.8)	—
Teachers^a	—	—	—	—	—	1 (3.8)	1 (3.8)	—	—	—
Superiors/work	—	—	—	—	—	1 (3.8)	1 (3.8)	—	—	1 (3.8)
General nondisclosure^b	9 (34.6)	—	—	4 (15.4)	3 (11.5)	1 (3.8)	2 (7.7)	—	1 (3.8)	—
Summarized index^c	9 (34.6)	8 (30.8)	6 (23.1)	5 (19.2)	5 (19.2)	5 (19.2)	4 (15.4)	3 (11.5)	3 (11.5)	3 (11.5)

aIn one case nondisclosure referred to a social (education) worker in a children’s home.

Refers to statements when participants talked about motives for delayed disclosure or nondisclosure in general (not related to specific persons).

Number of participants who mentioned the particular motives, independently of what group of people nondisclosure/delayed disclosure referred to. The summarized index does not correspond with the sum of values in each column because multiple naming was possible (e.g., a participant mentioned the same motive for nondisclosure to her parents and for delayed disclosure to her teacher).

Figure 2.

Grouping of codes into family “recipient = parent”

Quantitative analyses

Basic analysis of frequencies of the disclosure data (e.g., disclosure rates) were calculated using Atlas.ti. All other quantitative analyses were performed using the statistical package SPSS for Windows, release 18.0 (SPSS Inc, Chicago, IL). For the sample description, tables of frequencies and descriptive statistics (e.g., means) were used. To identify factors associated with disclosure and reporting rates, Fisher’s exact tests were performed. Significance of results was tested with two-tailed tests, and p value < .05 was considered significant. In addition, effect sizes d according to Cohen (1988) were calculated. An effect size of 0.2 is considered to be small, an effect size of 0.5 to be moderate, and an effect size of 0.8 to be large (Cohen, 1988).

The following variables were included to test for associations: immediate disclosure (yes/no), disclosure to parents (yes/no), report to the police (yes/no), singular versus repeated CSA, nonpenetrative versus penetrative CSA, extrafamilial versus intrafamilial CSA (defined as CSA committed by a person belonging to the core family), age of study participant at CSA (<12 vs. >11 years), age of perpetrator (<18 vs. >17 years), feelings of guilt (yes/no), feelings of shame (yes/no), and relationship situation of biological parents (living together/separated).

Both quantitative and qualitative analyses were performed with reference to the most serious CSA that participants had experienced. Due to the small number of male participants, an analysis of gender differences was not feasible.

Results

The results are presented following the research questions.

Number of Disclosing Adolescents and Reports to the Police

Eight (30.1%) participants disclosed the abuse immediately (within 24 hr) after the event; 17 (65.4%) delayed disclosure. One participant had not disclosed the abuse prior to the interview. Delays in disclosure ranged from several days to several years. Most participants described disclosure as a positive experience providing relief. Disclosure was also often a prerequisite for receiving support and for disrupting the abuse.

P 13²: After the sexual assault I didn’t tell anybody about what had happened. But then I talked to a friend and afterwards I felt much better.

I:And you told your mother about the abuse six months afterwards?

P6:No, even longer, after a year and a half. . . . At first I didn’t know how to tell her but then suddenly I told her. . . . She packed his stuff and threw him out of the flat. . . . We went to the police and afterwards I got therapy.

Six participants also reported the CSA to the police prior to the interview. One girl informed the police about the abuse after the interview. All participants who made a report to the police had been assisted and accompanied by a parent.

P22:

I went to a friend and asked if I could use her mobile. I called my parents. Yeah, and then, my mother came and picked me up. And then . . ., she said, “And now we’re going to the police. And then we went to the police.”

Recipients of Disclosure

Table 2 provides an overview of rates of immediate and delayed disclosure according to recipients. Independent of whether disclosure was immediate or delayed, disclosures were most often made to peers, followed by to parents. About a third of the participants disclosed the abuse to their peers but not to their parents. For example one girl, who had been abused by an acquaintance of her mother’s, talked to her friend and her boyfriend about the abuse but was too anxious to inform her parents:

And you never talked to anybody about the abuse?

P5:

Yes, I spoke to a friend, and once to my boyfriend, my ex-boyfriend.

[. . .]

And you couldn’t tell your parents what had happened?

P5:

No.

You still haven’t talked to them?

P5:

No.

Immediate disclosures were also made to siblings in three cases and to a relative in one case, and intimate partners, helping professionals, and teachers proved to be important recipients of delayed disclosure for some participants. A girl who had been sexually assaulted by a school friend said,

P2:

I suppressed it for about two hours. I went to the driving lesson. I was very angry and sad. Then my brother picked me up and then I almost collapsed. I told him everything.

Another girl said,

P17:

I talked to my teacher. I couldn’t concentrate anymore in school, my marks got worse and worse. My teacher asked me what the matter was with me. I thought about telling a long time. But then I just told him.

Peers were also most often the first recipients of disclosure (n = 13, 50%), whereas parents were the first to be informed in only one-third of cases (n = 9, 34.6%).

Was she ( a clergy person) the first person you disclosed yourself to?

P21:

No. That was a good friend.

And this was before you talked to your mother?

P21:

Yes.

Two (7.7%) participants first disclosed to an intimate partner, one (3.9%) to a helping professional, and another one (3.9%) to a sibling.³ On average, participants self-disclosed CSA to three groups of people (M = 3.0, SD = 1.5). Peers who were disclosed to were more often female than male (19 female vs. 10 male recipients; gender of recipient was unclear in four cases). Furthermore, mothers were more often recipients of disclosure than were fathers; 7 participants informed only their mothers about CSA, and 9 participants informed both parents. None of the participants self-disclosed only to their fathers.

Motives for Delayed or Nondisclosure

Table 3 shows categories of reasons for delayed disclosure or nondisclosure to which at least three participants could be allocated as well as frequencies of reasons with regard to the potential recipients. The most frequently mentioned motive (reported by more than a third of the participants) was denying the CSA. This category comprised statements that suggested that the participant wanted to forget about the CSA, psychologically suppressed the CSA, or could not remember the CSA for a certain period of time after the incident. These reasons were most often mentioned when participants talked about nondisclosure in general rather than in relation to a particular person.

Why do you think, didn’t you tell anybody about the abuse earlier?

P6:

I just wanted to forget about it. And then, of course, I tried to hide it in order to forget about it.

P22:

And then I would say I forgot about it. Better to say I psychologically suppressed it. And then I went to school again.

The second most frequently mentioned reason for delayed or nondisclosure was that participants did not want to burden potential recipients with such information or that participants saw the potential recipients as emotionally too instable to confide in. This motive was primarily mentioned for nondisclosure to parents. A girl who had been sexually assaulted by a school mate disclosed to her parents only after a delay because she was anxious that her mother would not be able to cope with the fact that her daughter had almost been raped.

P2:

However, I didn’t want to tell my parents about it. I knew that my mum would not have been able to deal with it.

The narrative of another girl shows the ambivalent feelings that children can have toward disclosure: This girl had been abused by a perpetrator who had also abused her sister; her sister had disclosed the abuse to their mother. The girl knew that she would feel relieved if her mother knew about the abuse, but anxiety about causing her mother distress prevented her from disclosing.

P5:

On the one hand I know I would feel much better but I also don’t want to talk to her, because it would be extremely difficult for her. . . . Yeah, I mean I see how distressed my mother is and that’s why I don’t want to tell her about this too. I don’t know what would happen if I would say to her “Mom, the same thing happened to me too.”

Some ant to burden a friend.

P9:

I cried a lot and I just like didn’t have anybody to talk to. . . . I had a friend. . . . She came from a caring parents’ home. I didn’t want to destroy that.

Another prevalent motive with regard to nondisclosure to parents was lack of trust. Almost a quarter of the participants said that they did not tell their parents because they did not have a sufficiently close or trusting relationship with them:

P3:

I mean I had somehow never had a confidant, but it would have been better for me. . . . It was just like that my parents never—I mean had never cared for me. . . . ; our relationship had never been good. . . . It had never been close.

P11:

I can’t talk to my mother. She’s so withdrawn and incommunicative. It’s just the whole relationship with my mother.

Even if participants did not state specifically that they would doubt the reliability of their relationship with their parents, they often mentioned anxieties indicating that many were unsure of whether parents would be on their side. Feelings of shame, fear of stigmatization, fear of disbelief, and fear of parental sanctions (e.g., because participants had sex) were the most frequently anxieties that prevented participants from disclosure to parents.

Why didn’t you tell her sooner?

P21:

I just felt so ashamed.

Why do you think you didn’t you tell your mother . . . . ?

P26:

Later I thought that she wouldn’t believe me anyway, and that she would think that I saw things, that I was crazy.

P19:

Yes, I thought that they (parents) would get angry with me.

Okay. Because you had sex?

P19:

Mmh, yes.

Some participants were also afraid of the perpetrator. A girl who was maltreated by a teacher both physically and sexually for several years was threatened by him: he told her that he would kill her if she told anybody.

Do you know why you didn’t tell anybody?

P26:

I was scared of him. He said that he would kill me if I told anyone.

In intrafamilial abuse, adolescents were also often afraid that disclosure would disrupt their family or disrupt their mother’s intimate relationship with the perpetrator. A girl who had been sexually harassed by her mother’s boyfriend was anxious that she and her mother would not be able to make their living because her mother only received a disability pension.

P8:

There was always the anxiety that they would break up with each other, and that my mum and I would not be able to live on our own.

Finally, two other prevalent concerns that prevented participants from disclosing were noncomprehension of what had happened to them and regarding CSA as too intimate to be talked about. Noncomprehension of CSA was mentioned mainly when participants were preadolescent at the time of CSA.

Why were you not able to talk to anybody?

P5:

. . . because I did not understand what actually had happened.

P2:

And I mean, it was difficult for me. I mean I’m at an age at which you don’t want to reveal your whole love life to your parents.

You mean at an age when you don’t want them to know about your private life?

P2:

Yes, exactly.

Factors Associated With Disclosure and Reporting

Table 4 lists the results of tested associations between disclosure and reporting rates and characteristics of CSA, feelings of guilt and shame, and the relationship situation of the participants’ parents. Significant results (Fisher’s exact test) or results with a Cohen’s d of at least 0.5 (indicating a moderate association) are printed in bold. Associations were mainly found with immediate disclosure (within 24 hr). Immediate disclosure was more likely when the perpetrator was not a family member, when the CSA happened only once, when the victim was older than 12 years at the beginning of CSA, when the perpetrator was a minor, when the participant did not feel guilty or ashamed of the abuse, and when the participant’s parents were still living together. Participants whose parents were not separated were also more likely to disclose the abuse to their parents. No associations were found with regard to reporting rates.

Table 4.

Results of Quantitative Analyses of Factors That Are Related to Disclosure (Fisher’s Exact Tests, Effect Sizes)

	Variables related to disclosure
Characteristics of child sexual abuse (CSA)	Immediate disclosure (yes/no)	Disclosure to parents (yes/no)	Report to police (yes/no)
Intrafamiliar CSA (yes/no)	p = 0.03, d = 1.0	p = 1.00, d = 0.0	p = 0.63, d = 0.3
Age at CSA (< vs. >11 years)	p = 0.02, d = 1.1	p = 1.00, d = 0.2	p = 0.38, d = 0.4
Penetrative CSA (yes/no)	p = 0.42, d = 0.4	p = 0.68, d = 0.3	p = 0.64, d = 0.4
Singular CSA (yes/no)	p = 0.08, d = 0.9	p = 0.69, d = 0.2	p = 0.63, d = 0.4
Age of perpetrator (< vs. >17 years)	p = 0.20, d = 0.7	p = 1.00, d = 0.0	p = 1.00, d = 0.0
Relationship situation of parents
Parents living together (yes/no)	p = 0.007, d = 1.3	p = 0.005, d = 1.4	p = 1.00, d = 0.1
Guilt/shame
Feelings of guilt	p = 0.08, d = 0.8	p = .69, d = 0.2	p = 1.00, d = 0.0
Feelings of shame	p = 0.15, d = 0.7	p = 1.00, d = 0.1	p = 1.00, d = 0.2

Discussion

This study examined aspects of disclosure of CSA in a nonclinical sample of sexually victimized adolescents by means of qualitative face-to-face in-depth interviews. The main objectives of the study were to examine the number of disclosing adolescents, who recipients of disclosure are, children’s and adolescents’ motives for nondisclosure and delayed disclosure, and factors that are related to disclosure.

The results indicate that less than a third of sexually victimized children and adolescents disclose the abuse to somebody within 24 hr. Consistent with clinical experience (Paine & Hansen, 2002) and previous findings from quantitative research (Hershkowitz et al., 2007; Kogan, 2004), the majority of participants disclosed CSA with a delay that ranged up to several years. This finding suggests that participants had to surmount certain barriers that hindered disclosure.

A further main result of our study is that peers are considerably more often recipients of disclosure than the victims’ parents. This indicates that adolescents had more concerns about telling their parents about the abuse than about telling a friend. The findings on motives for nondisclosure or delayed disclosure confirmed this assumption. Reasons mentioned for delayed or nondisclosure to parents indicate that the young people often doubted the trusting nature or reliability of their relationship with parents. Specifically, many participants said that they did not trust their parents, felt ashamed, or were anxious that their parents would not believe them or would even punish them. Furthermore, approximately a quarter of participants regarded their parents as not being emotionally stable enough to cope with the sexual abuse. In most of the interviews, nondisclosure to parents became the main issue because participants obviously felt distressed by it. Most of the young people had a strong wish to talk to their parents about their frightening sexual experiences, but many did not feel able to do so.

These findings correspond with the finding by Priebe and Svedin (2008) that peers were most often the confidants of disclosure and support Priebe and Svedin’s hypothesis that victims more frequently expect unsupportive reactions from their parents than from their friends. The results are also in line with Crisma et al. (2004), who reported that the main impediments of adolescents disclosing to their parents were fear of accusation and unstable family relationships. Hershkowitz et al. (2007) examined 30 alleged child victims by means of a forensic interview, and the results indicated that many children are afraid that they would face negative reactions from their parents if they disclosed. That study also found that the children’s anxieties were often reasonable, since the majority of the parents participating in the study in fact reacted in an unsupportive way.

Based on our and previous findings, it is reasonable to conclude that a reliable relationship with parents may be one of the most important predictors of disclosure to parents, which in turn may be a highly predictive determinant of whether the CSA comes to a halt. However, the fact that peer disclosure is highly prevalent may not only reflect children’s mistrust toward parents. It also needs to be considered that peer relationships become more important during adolescence, which is in line with gradual parental detachment during this developmental period (Priebe & Svedin, 2008). Friends are significant confidants of adolescents generally; therefore, it is not surprising that they play an important role in the disclosure process.

This study not only investigated impediments to disclosing to parents but also to disclosing in general and to other people. Main motives for general nondisclosure included denying the CSA (such as repression of memories), feelings of shame or fear of stigmatization, and the victim’s lack of comprehension at the time of CSA that she or he was actually experiencing sexual abuse. These findings confirm the clinical experience of practitioners who have previously reported that the child’s feelings of shame, some victims’ tendency to psychologically suppress CSA, and a lack of comprehension in younger children all help the perpetrator hide the abuse (Paine & Hansen, 2002). Perpetrators often use feelings of shame as an active strategy to keep the children silent (Paine & Hansen, 2002).

As a final research question, this study examined factors related to disclosure. Since no previous qualitative study investigated factors associated with disclosure, the results of our study were compared with findings from quantitative studies. However, due to the differences in methodology, the validity of such comparisons is always limited.

In accordance with previous research (e.g., Kogan, 2004, Smith et al., 2000), immediate disclosure was positively related to extrafamilial CSA and age at the beginning of CSA (i.e., older children were more likely to disclose immediately). Whereas younger children are believed to be incapable of comprehending that they were sexually assaulted, intrafamilial CSA is thought to be more stigmatizing and, thus, inhibits the child’s disclosure (Kogan, 2004). In addition, intrafamilial victims are often worried that disclosure will lead to disruptions of family relationships or separation of their parents (Kogan, 2004). Also in line with previous research (e.g., Hershkowitz et al., 2007; Kogan, 2004), immediate disclosure followed more often after single CSA than it did after repeated CSA and if the perpetrator was a minor. Whereas immediate disclosure may have hindered the perpetrator from repeating the CSA, the association with age of the perpetrator is at least partly to be explained that minor perpetrators were only involved in extrafamilial CSA.

Immediate disclosure was also associated with feelings of shame or guilt. This relationship supports the finding that some participants reported feelings of shame or guilt as a motive for nondisclosure. As a final result, participants whose parents were still living together at study participation were more likely to disclose CSA within 24 hours and to disclose CSA to their parents than were participants whose parents were divorced or separated. Kogan (2004) found the same association in sexually victimized adolescents. Considering the above-mentioned finding that nondisclosure toward parents is often hindered by impaired quality of the parent–child relationship, this finding again supports the hypothesis that disrupted family relationships or family distress in general can encourage a child to keep the CSA a secret.

Limitations

Some limitations of this study need to be mentioned. Although our study sample was considerably large for qualitative research, samples in qualitative research are never representative and thus bear the risk of selection bias. For example, the finding that all but one participant had previously disclosed the CSA to somebody before their participation in this study suggests that nondisclosing adolescents were less willing to participate. Another bias could have been caused by the fact that about one third of participating adolescents sought advice from the interviewer and took us up on our offer of short-term counseling. It can be assumed that specifically adolescents who were looking for support participated in the study.

Whereas our sample was considerably large for a qualitative analysis, it was rather small for a correlational analysis. However, since the chance that statistical tests are significant declines with smaller sample (Bortz & Lienert, 2008), the associations found are an underestimation rather than an overestimation of the true associations. This was also supported by the considerably large effect sizes among some of the tested associations with disclosure, even though statistical tests did not reach significance. Most of our findings also correspond with previous research and clinical experience (Paine & Hansen, 2002), which is indicative of the good validity of our results.

A further shortcoming of our sample was the low participation rate of male adolescents. The participation of three boys did not allow us to analyze any gender differences regarding disclosure. It is known from previous research and practice that boys are more hesitant to disclose CSA than are girls due to fears of being regarded as homosexual (Paine & Hansen, 2002) and in self-defence of their self-image as strong and invulnerable men (Richter-Appelt, 2002). Due to these lower disclosure rates, boys may be less likely to participate in CSA research than girls are. A further drawback of our study was that the duration of delay of disclosure was not assessed, as this was not possible due to the qualitative methods used. Precise time of disclosure after assault, which is typically made to several people at different points in time, requires a quantitative assessment by standardized questions.

Despite these limitations, our study is unique in the field of CSA. As the first study of its kind to interview a considerably large sample of adolescent survivors of CSA in face-to-face interviews, the results provide qualitative in-depth insight into various aspects of disclosure of CSA. The study adds important new data to the findings of Crisma et al. (2004), who interviewed adolescent survivors of CSA by telephone. Whereas telephone interviews may result in higher participation rates because of more perceived anonymity, face-to-face interviews are regarded as the preferred and most valid method in interview research because they allow closer access to participants (King & Horrocks, 2010).

Directions for Future Research and Implications for Practice

The results of this study have some implications for future research and clinical practice. Specifically, additional studies are needed that investigate aspects of disclosure of CSA in population-based samples of adolescents and apply both quantitative and qualitative research methods. Particularly, victims’ motives not only for nondisclosure but also for disclosure need to be examined. Whereas our content analysis according to Mayring (2008) was useful for identifying the most prevalent concerns and anxieties of sexually victimized children about disclosure, our approach did not allow in-depth analysis of the participant’s decision process. For a more in-depth analysis of when, why and under what circumstances children disclose or do not disclose, other qualitative analysis tools, such as the interpretative phenomenological analysis (Smith et al., 2009), might be used.

Future research should also focus on investigating predictors of disclosure. Previous researchers have mainly investigated associations between disclosure and characteristics of CSA. Family background variables have only rarely been included in analyses. Our results showed that the victim’s relationship to her or his parents, and the parents’ emotional stableness as judged by the victim, might be two of the most important predictors of the disclosure process. Future studies should examine whether our results can be confirmed.

Some recommendations for practice can also be inferred from our results. Although current CSA prevention programs focus specifically on facilitation of disclosure to parents, they neglect the fact that many children do not tell their parents because they do not trust them. Many prevention booklets advise parents to talk openly with their children about CSA. However, open communication about CSA in the family may result in disclosure only, if the child regards the quality of the relationship with her or his parents as secure enough. Therefore, the programs should focus more on the strengthening of the child–parent relationship, for example, by offering courses in parenting.

Our results also showed that school personnel (e.g., teachers, school social workers) were rarely told about CSA. This may suggest that current prevention activities in the schools are not efficient enough to facilitate disclosure in school. Particularly if children do not see a way to disclose CSA to their parents, they should be given the alternative to disclose the abuse to an adult confidant at school.

However, although parents and teachers may be more competent to disrupt CSA, participants regarded friends as the most reliable confidants. Specifically in intrafamilial CSA, friends might be particularly important recipients of disclosure, because one parent is often involved in CSA. Therefore, prevention programs for adolescents must be developed that advise young people on how to react if a friend discloses CSA. The programs should be implemented in school education and also via internet services for young people, which could be an important platform to address disclosure of CSA. Second prevention strategies should always be culture sensitive and should be of low threshold in order to reach as many young people as possible.

Finally, special attention should been given to CSA among peers. Recent studies suggest (e.g., Krahe, 2009) that sexual violence among adolescents is increasing. Adolescent perpetrators committed about half of the sexual abuses in this study. Therefore, it is important that peer abuse is discussed with young people in the schools. Adolescents need to be confirmed in a self-confident belief that sexual violence committed by a peers is just as unacceptable as CSA committed by adults. Young people must be informed about contact persons at their schools (e.g., social workers, teachers) and services where they can find help if they become a victim of peer abuse. Disclosure of CSA can only be facilitated if survivors feel confident that disclosure will lead to some type of support.

Footnotes

Acknowledgements

We are most grateful to the study participants, who made this research possible. We would also like to thank Noemi Landolt for assisting us with the transcriptions of the interviews, and Jürg Streuli for advising us with regard to qualitative research methods.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Data collection was part of the Optimus study, which was initiated and funded by the UBS Optimus Foundation; analyses of data and writing of the article were funded by the Margrit Egnér Foundation and the Walter Häfner Foundation.

Notes

Bios

Verena Schönbucher has a PhD in psychology. She worked as a research fellow at the University Children’s Hospital Zurich (Switzerland), at the Department for Sex Research and Forensic Psychiatry of the University Hospital Hamburg-Eppendorf (Germany), and at the London Metropolitan University (Child and Woman Abuse Studies Unit). Currently, she works as a scientific collaborator and as a clinical psychologist at the University Hospital Zurich as well as at the University Children’s Hospital Zurich. Her main research interests are sexual violence and sexual abuse, psychosocial aspects of disorders of sex development, and female sexuality.

Thomas Maier, MD, is head of the psychiatric Services of the Canton St. Gallen-North, Switzerland, and lecturer at the University of Zurich. He is a board psychiatrist and a trained psychotherapist and psychotraumatologist. From 2003 to 2010 he headed the outpatient unit for the care of victims of war and torture at University Hospital Zurich. He has published on posttraumatic stress, obsessive-compulsive disorders, and transcultural psychiatry, and also on complexity and nonlinear dynamics in psychiatry and on limitations of evidence-based psychiatry.

Meichun Mohler-Kuo has a ScD in maternal and child health. She is the unit leader of mental health and social epidemiology at the Institute of Social and Preventive Medicine, University of Zurich. She has research experience in the areas of child and adolescent health, substance abuse, exposure to violence, and sexual violence.

Ulrich Schnyder, MD, is a psychiatrist and licensed psychotherapist. He is professor of psychiatry and psychotherapy, and head of the Department of Psychiatry and Psychotherapy, University Hospital Zurich, Switzerland. Originally trained as a general practitioner, he then specialized in psychiatry and psychotherapy. His clinical activities include emergency psychiatry, psychotherapy, consultation-liaison psychiatry, and psychotraumatology. His research activities are currently focused on the psychosocial consequences of accidental injuries, neurobiological aspects of PTSD, psychotherapy and pharmacotherapy for PTSD, and resilience to stress. He is past president of the European Society for Traumatic Stress Studies (ESTSS), past president of the International Federation for Psychotherapy (IFP), and immediate past president of the International Society for Traumatic Stress Studies (ISTSS).

Markus A. Landolt, PhD, is head of pediatric psychology at the University Children’s Hospital, Zurich and associate professor at the Institute of Psychology, University of Zurich, Switzerland. He has conducted a lot of research studies and published numerous papers on a variety of topics, including child psychotraumatology and pediatric psychology. Besides his research activities, he is trained in family therapy and cognitive-behavioral therapy and has been working for more than 20 years as a pediatric psychologist. He was the recipient of the Falk-von-Reichbach Award from the German Society of Traumatic Stress Studies in 2005 and was named a Fellow of the American Psychological Association in 2010. He is also a board member of the Children’s Research Center at the University of Zurich and a member of the Editorial Board of the European Journal of Psychotraumatology.

References

Arata

C. M.

(1998). To tell or not to tell: Current functioning of child sexual abuse survivors who disclosed their victimization. Child Maltreatment, 3, 63-71.

Berliner

Conte

J. R.

(1995). The effects of disclosure and intervention on sexually abused children. Child Abuse & Neglect, 19, 371-384.

Bortz

Lienert

G. A.

(2008). Kurzgefasste Statistik für die klinische Forschung [Summarily statistics for clinical research]. Heidelberg, Germany: Springer.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

Crisma

Bascelli

Paci

Romito

(2004). Adolescents who experienced sexual abuse: Fears, needs and impediments to disclosure. Child Abuse and Neglect, 28, 1035-1048.

DeVoe

E. R.

Coulborn Faller

(1999). The characteristics of disclosure among children who may have been sexually abused. Child Maltreatment, 4, 217-227.

Finkelhor

Hamby

S. L.

Ormrod

Turner

(2005). The Juvenile Victimization Questionnaire: Reliability, validity, and national norms. Child Abuse & Neglect, 29, 383-412.

Fontes

L. A.

(1993). Disclosures of sexual abuse by Puerto Rican children: Oppression and cultural barriers. Journal of Child Sexual Abuse, 2, 21-35.

Hamby

S. L.

Finkelhor

Ormrod

Turner

(2004). The Juvenile Victimization Questionnaire (JVQ): Administration and scoring manual. Durham, NH: Crimes Against Children Research Institute.

10.

Hanson

R. F.

Resnick

H. S.

Saunders

B. E.

Kilpatrick

D. G.

Best

(1999). Factors related to the reporting of childhood rape. Child Abuse & Neglect, 23, 559-569.

11.

Hershkowitz

Lanes

Lamb

M. E.

(2007). Exploring the disclosure of child sexual abuse with alleged victims and their parents. Child Abuse & Neglect, 31, 111-123.

12.

Jensen

T. K.

Gulbrandsen

Mossige

Reichelt

Tjersland

O. A.

(2005). Reporting possible sexual abuse: A qualitative study on children’s perspectives and the context for disclosure. Child Abuse & Neglect, 29, 1395-1413.

13.

King

Horrocks

(2010). Interviews in qualitative research. London: Sage.

14.

Kogan

S. M.

(2004). Disclosing unwanted sexual experiences: Results from a national sample of adolescent women. Child Abuse and Neglect, 28, 147-165.

15.

Krahe

(2009). Sexuelle Aggression und Opfererfahrung unter Jugendlichen und jungen Erwachsenen. Prävalenz und Prädiktoren [Sexual aggression and victimization among adolescents and young adults: Prevalence and Precdictors]. Psychologische Rundschau, 60, 173-183.

16.

Lamb

Edgar-Smith

(1994). Aspects of disclosure. Mediators of outcome of childhood sexual abuse. Journal of Interpersonal Violence, 9, 307-326.

17.

Landolt

M. A.

Nuoffer

J. M.

Steinmann

Superti-Furga

(2002). Quality of life and psychological adjustment in children and adolescents with early treated phenylketonuria can be normal. Journal of Pediatrics, 140, 516-521.

18.

Lombard

Snyder-Duch

Bracken

C. C.

(2010). Practical resources for assessing and reporting intercoder reliability in content analysis research projects. Retrieved from http://astro.temple.edu/~lombard/reliability/

19.

London

Bruck

Ceci

S. J.

(2005). Disclosure of child sexual abuse. What does the research tell us about the ways that children tell? Psychology, Public Policy and Law, 11, 194-226.

20.

London

Bruck

Wright

D. B.

Ceci

S. J.

(2008). Review of the contemporary literature on how children report sexual abuse to others: Findings, methodological issues, and implications for forensic interviews. Memory, 16, 29-47.

21.

Mallinckrodt

Wang

(2004). Quantitative methods for verifying semantic equivalence of translated research instruments: A Chinese version of the Experiences in Close Relationships Scale. Journal of Counseling Psychology, 51, 368-379.

22.

Malloy

L. C.

Lyon

Quas

J. A.

(2007). Filial dependency and recantation of child sexual abuse allegations. Journal of American Child and Adolescent Psychiatry, 46, 162-170.

23.

Mayring

(2008). Qualitative Inhaltsanalyse. Grundlagen und Techniken [Qualitative Content Analysis. Basics and Techniques] (10th ed.). Basel, Switzerland: Beltz.

24.

Paine

M. L.

Hansen

D. J.

(2002). Factors influencing children to self-disclose sexual abuse. Clinical Psychology Review, 22, 271-295.

25.

Priebe

Svedin

C. G.

(2008). Child sexual abuse is largely hidden from the adult society. An epidemiological study of adolescents’ disclosures. Child Abuse & Neglect, 32, 1095-1108.

26.

Richter-Appelt

(2002). Verführung, Trauma, Missbrauch [Seduction, Trauma, Abuse]. Giessen, Germany: Psychosozial-Verlag.

27.

Roesler

T. A.

Wind

T. W.

(1994). Telling the secret. Adult women describe their disclosures of incest. Journal of Interpersonal Violence, 9, 327-338.

28.

Ruggiero

K. J.

Smith

D. W.

Hanson

R. F.

Resnick

H. S.

Saunders

B. E.

Kilpatrick

D. G.

. (2004). Is disclosure of childhood rape associated with mental health outcome? Result from the National Women’s Study. Child Maltreatment, 9, 62-77.

29.

Smith

S. G.

Cook

S. L.

(2008). Disclosing sexual assault to parents. The influence of parental messages about sex. Violence Against Women, 14, 1326-1348.

30.

Smith

J. A.

Flowers

Larkin

2009. Interpretative phenomenological analysis: Theory, method and research. London: Sage.

31.

Smith

D. W.

Letourneau

E. J.

Saunders

B. E.

Kilpatrick

D. G.

Resnick

H. S.

Best

C. L.

(2000). Delay in disclosure of childhood rape: results from a national survey. Child Abuse & Neglect, 34, 273-287.