Share Your Articles

Child Abuse & Neglect 32 (2008) 1095–1108

Contents lists available at ScienceDirect

Child Abuse & Neglect

Child sexual abuse is largely hidden from the adult society An epidemiological study of adolescents’ disclosures�

Gisela Priebe a,∗, Carl Göran Svedin b

a Department of Child and Adolescent Psychiatry, IKVL, Lund University, Sofiavägen 2 D, S-221 41 Lund, Sweden b Department of Child and Adolescent Psychiatry, IMK, Faculty of Health Sciences, Linköping University, Linköping, Sweden

a r t i c l e i n f o

Article history: Received 6 February 2007 Received in revised form 1 April 2008 Accepted 17 April 2008

Keywords: Child sexual abuse Disclosure Adolescence Gender

a b s t r a c t

Objectives: The aim of this study was to investigate disclosure rates and disclosure patterns and to examine predictors of non-disclosure in a sample of male and female adolescents with self-reported experiences of sexual abuse. Method: A sample of 4,339 high school seniors (2,324 girls, 2,015 boys) was examined with a questionnaire concerning sexual experiences in this study with a focus on disclosure of sexual abuse (non-contact, contact or penetrating abuse, and including peer abuse). Results: Of the sample, 1,505 girls (65%) and 457 boys (23%) reported experience of sexual abuse. The disclosure rate was 81% (girls) and 69% (boys). Girls and boys disclosed most often to a friend of their own age. Few had disclosed to professionals. Even fewer said that the incident had been reported to the authorities. Logistic regression showed that it was less likely for girls to disclose if they had experienced contact sexual abuse with or without penetration, abuse by a family member, only a single abuse occasion or if they had perceived their parents as non-caring. Boys were less likely to disclose if they studied a vocational program, lived with both parents or had perceived their parents as either caring and overprotective or non-caring and not overprotective. Conclusions: Disclosing sexual abuse is a complex process. Much is hidden from the adult society, especially from professionals and the legal system. Since peers are the most com- mon receivers of abuse information, programs for supporting peers ought to be developed. Differences in disclosure patterns for girls and boys indicate that a gender perspective is helpful when developing guidelines for professionals. Practice implications: Professionals, especially in the school system, need to be more aware of the finding that few sexually abused children seek help from professionals or other adults and that support offers should be directly addressed not only to the vulnerable young persons themselves but also to peers who wish to help a friend.

© 2008 Elsevier Ltd. All rights reserved.

Introduction

Most studies of disclosure of sexual abuse during childhood either focus on children’s disclosure in a professional setting such as in a forensic or clinical interview (Berliner & Conte, 1995; Bradley & Wood, 1996; DeVoe & Faller, 1999; Jensen, Gulbrandsen, Mossige, Reichelt, & Tjersland, 2005; Sjöberg & Lindblad, 2002; Sorensen & Snow, 1991; Svedin & Back, 2003)

� The authors would like to acknowledge the Committee into knowledge concerning sexual exploitation of children in Sweden (S 2003:5) at the Swedish Social Ministry, the Crime Victim Compensation and Support Authority in Sweden and Her Majesty Queen Silvia’s Jubilee Foundation for their financial support to the project.

∗ Corresponding author.

0145-2134/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.chiabu.2008.04.001

1096 G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108

or on population-based retrospective reports from adults (Arata, 1998; Collings, 1995; Finkelhor, Hotaling, Lewis, & Smith, 1990; Hanson, Resnick, Saunders, Kilpatrick, & Best, 1999; Roesler, 1994; Ruggiero et al., 2004; Smith et al., 2000; Somer & Szwarcberg, 2001; Tang, 2002).

An advantage of population-based studies is that they can collect information even from participants who never disclosed their experience of sexual abuse prior to their participation in the study and who have never been in contact with the professional system. This may be of great interest for public health policy and for support services intended for sexually abused persons. The disclosure rates in the above named retrospective studies with adults are between 31% and 41% for disclosure during childhood and between 58 and 72% for lifetime disclosure. Peer abuse (peer defined as a person not more than five years older than the victim) is included in some of these studies, while it is not specified in other studies and excluded in one (Arata, 1998).

Although it can be expected that there is less recall bias in adolescent retrospective studies since the self-reported sexual abuse is closer in time there are surprisingly few retrospective studies with adolescent participants. It is difficult to compare the few available studies since, for example, Kogan (2004) only included girls, Edgardh and Ormstad (2000) included both boys and girls while others do not present separate results for boys and girls (Fergusson, Lynskey, & Horwood, 1996; Helweg- Larsen & Larsen, 2006; Kellogg & Huston, 1995). The lowest disclosure rate (56%) is reported for boys (Edgardh & Ormstad, 2000) while the other studies present disclosure rates ranging between 74% and 87%. All adolescent studies included both adult and peer abuse.

Disclosing adolescents usually talk to friends or parents (Edgardh & Ormstad, 2000; Fergusson et al., 1996; Kellogg & Huston, 1995). Friends as recipients of disclosure are of increasing importance for adolescents while younger children disclose more often to an adult (Kogan, 2004). Few disclosing adolescents, between 3% and 13%, had talked to an adult professional (Edgardh & Ormstad, 2000; Kellogg & Huston, 1995; Kogan, 2004). This is in line with findings from retrospective studies with adults (Arata, 1998; Collings, 1995; Hanson et al., 1999; Smith et al., 2000; Tang, 2002).

Few retrospective studies of adolescents have analyzed variables other than disclosure rates and recipients of disclosure. Kogan (2004) investigated how survivor characteristics, abuse characteristics and family context attributes were related to the timing and the recipient of disclosure. The identity of the recipient of the disclosure (adult, peer only, none) was related to age of onset, penetration, fear for one’s life during the abuse, injury during the abuse, family structure, and the age differential between the victim and the perpetrator while the timing of disclosure was related to age of onset, a known perpetrator, a familial relationship with the perpetrator, and a history of drug abuse in the household. The disclosure recipient or disclosure timing were not associated with race/ethnicity, parental education, household income and house- hold alcohol abuse. Kellogg and Huston (1995) studied the reasons for disclosure or non-disclosure. In general, fear and embarrassment were the most common reasons for delay or non-disclosure, but there were also differences between ethnic groups.

Disclosure of sexual abuse is usually regarded as beneficial to the exposed child (Arata, 1998; Paine & Hansen, 2002). However, previous studies with children in forensic samples (Elliott & Briere, 1994; Nagel, Putnam, Noll, & Tricket, 1997), undergraduate female students (Sinclair & Gold, 1997) and adult women (Ruggiero et al., 2004) have not shown any direct link between disclosure and a positive mental health outcome. Association between current perceived mental health and disclo- sure and the role of general parental support for young persons’ disclosure has not yet, to our knowledge, been investigated in population-based studies with adolescents.

Studies about the role of abuse-specific parental support in relation to disclosure have shown that the child’s willingness to disclose in a professional context increased when the mother was supportive and believed in the child’s disclosure (Elliott & Briere, 1994; Lawson & Chaffin, 1992). General parental support and care seem to be a good predictor for recovery after sexual abuse (Lynskey & Fergusson, 1997; Spaccarelli & Kim, 1995). Sinclair and Gold (1997) found that general parental support was not related to withholding disclosure from others, but it is unclear if parental support at the time of the study or during childhood was measured.

Previous studies of adolescents about disclosure of sexual abuse include female-only samples or mixed samples that either include small groups of boys who reported sexual abuse or that do not show separate results for boys or girls. As group differences between boys and girls concerning prevalence and sexual abuse characteristics are well-documented (Watkins & Bentovim, 1992), it can be expected that there are differences concerning disclosure rates, disclosure patterns and predictors for disclosure or non-disclosure, too. There is a need for studies that include and show sufficient numbers of both boys and girls who report sexual abuse.

This study is intended to contribute data about disclosure of sexual abuse from a large population-based sample of adolescents with self-reported experiences of child sexual abuse including peer abuse. The specific aims of this study were

• to investigate disclosure rates and disclosure patterns associated with recipients of disclosure, abuse characteristics, socio- demographic variables, perception of parents when growing up and current perceived mental health, separate for boys and girls,

• to examine predictors of non-disclosure, separate for boys and girls.

No specific hypotheses are stated as the approach is basically exploratory. All variables in the analysis are expected to be associated with disclosure.

G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108 1097

Method

Participants

This study was a part of the Swedish base for comparative studies under the aegis of the Baltic Sea Regional Study on Adolescent Sexuality and the aim was to obtain a representative sample of high school seniors (3rd year high school students) in each country. Following the common guidelines the capital (Stockholm), one large port (Malmö) and some smaller cities (Luleå, Haparanda and Falköping) were chosen in Sweden. All 3rd year students (high school seniors) in all high schools in these cities were included in the initial group. Ninety-eight percent of all students finishing grade 9 in the Swedish compulsory school system enter the high school system (grades 10–12) each year. According to official statistics, about 90% of Swedish 18 year olds are enrolled in high school, 2% are studying in other education alternatives such as university and 8% are not studying (Sweden’s Statistical Database, 2003).

In the high schools of the participating cities 10,751 high school seniors were registered. At the time of the implementation of the study in the year 2003, there were 17 different national educational programs which can be classified as either academic or vocational at high school in Sweden. According to the sampling plan, only whole classes were selected, representing 50% of all students at each of the national educational programs in each town. If there was only one class in a program in a town, the whole class was selected. This resulted in a selection of 5,623 students (52.3% of the enrolled students), and 4,377 of them choose to participate in the study. Thirty-eight questionnaires were excluded as being incompletely filled-in. The final number of participants was consequently 4,339 (n = 2,324 girls and n = 2,015 boys), resulting in a response rate of 77.2%. The mean age of the participants was 18.15 years (SD = .74). In this study a subsample is used that consists of all participants who reported experience of sexual abuse and who answered to questions about disclosure of the abuse (n = 1,493, mean age 18.17 years, SD = .68).

Procedure

The director of the entire school system in each participating community was asked to grant permission to present the research project to the principal of each high school. Once permission was given by the principal of a school, all students in the selected classes were asked for and gave consent to participate based on their consideration of oral and written information. One assistant from the research group visited each class to distribute the questionnaires and then collect them after the students had finished them. In order to ensure that the students could not influence each other, they completed the questionnaires at the same time in the classroom. If the classroom was too small to guarantee privacy another room was chosen. The anonymous questionnaires were placed in unmarked envelopes, sealed by the students, and collected by the omnipresent research assistant. The students were given oral and written information about where to get counseling if participation had caused feelings of distress.

Measures

The self-report questionnaire used in the study was based on a Norwegian survey of young people’s attitudes towards sexuality and sexual abuse (Mossige, 2001). Questions from other Nordic surveys concerning young people’s sexual experi- ences (Edgardh, 2001; Hammarén & Johansson, 2001, 2002; Tambs, 1994) and questions especially formulated for the study were added. The survey included questions about background variables, consensual sexuality, sexual abuse experiences, own sexual abusive behavior, sexual attitudes, experiences with pornography and experiences with sexual exploitation (to sell sex for compensation). There were in total 65 questions. The participants needed between 30 and 60 minutes to complete the questionnaire. In the analyses for this paper, variables about background and sexual abuse experiences were used.

A wide definition of sexual abuse including lifetime prevalence of non-contact abuse, contact abuse without penetration and penetrating abuse by both adult and peer offenders was used. The participants were asked if they had been exposed to any of the following against their will: (1) non-contact abuse “somebody exposed him/herself indecently towards you,” (2) contact abuse “somebody has pawed you or touched your body in an indecent way,” “you masturbated somebody else,” (3) penetrative abuse (not specifying if for example fingers or devices were used) “you have had sexual intercourse,” “you have had oral sex,” “you have had anal sex.”

Participants who had reported any of the above mentioned experiences of sexual abuse were asked to answer a question about how many times they had been exposed against their will and to describe the abuse characteristics on the first abuse occasion (participant’s age, offender’s age and gender, relation to the offender, victim or offender on alcohol or drugs, kinds of persuasion/pressure/force, including physical force or physical violence, used by the offender). Finally, two questions about disclosure of any abuse occasion were asked. The first question was “Could you talk to someone about what happened?” with “yes” or “no” as possible answers. The second question was “If yes, whom did you talk to?” It was possible to check several of the eight different alternatives for an answer (see Table 2) and to indicate if the incident was reported to social authorities or police.

The survey included questions about background variables such as gender, the educational program, immigrant status (first or second generation), family structure and parents’ socio-economical status. The International Socioeconomic Index (ISEI) was used to classify the occupational status of the parent or parents (Ganzeboom, de Graaf, & Treiman, 1992). The ISEI

1098 G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108

has values ranging from 0 to 90 and measures the attributes of occupation that convert a person’s education into income and status.

The Mental Health scale consisted of six slightly modified items from the SCL-90 (Derogatis, 1977) about symptoms of anxiety and depression experienced during the preceding week (see Table 4). Each question was scored from 1 (does not correspond at all) to 4 (corresponds exactly) and consequently the total score of the Mental Health scale ranged from 6 to 24. The internal consistency, assessed using Cronbach’s alpha, was .83. A cut off was set at the 80th percentile received by the participants in the total sample, resulting in a cut off point of ≥19.

Nine of 25 items from the Parental Bonding Instrument (Parker, 1990; Parker, Tupling, & Brown, 1979) were included in the questionnaire. The Parental Bonding Instrument measures two fundamental parenting dimensions, care and overprotection. Five items from the care dimension (item 2, 5, 6, 14 and 18 in the original scale, Parker et al., 1979) and four items from the overprotection dimension were used (item 9, 13, 15 and 23 in the original scale, Parker et al., 1979). Each question was scored from 1 (does not correspond at all) to 6 (corresponds exactly). The score for the caring dimension ranged between 5 and 30, high values indicating high care, and between 4 and 24 for the overprotection dimension, high values indicating high overprotection. Cronbach’s alpha was .70 for the care dimension and .75 for the overprotection dimension. Cut off values based on the data from the total sample were ≤20 for the care score (at the 25th percentile) and ≥16 for the overprotection score (at the 80th percentile). The cut off values were used in the calculation of the four types of parental bonding (Table 4).

The data for girls and boys were analyzed separately. This does not mean that girls and boys as a group are assumed to have the same experiences concerning disclosure and related variables. When differences between girls and boys are described in the paper, usually the term “gender” is used instead of “sex” in order to not confuse sex and sexual activity.

Ethical considerations

The study was approved by the Regional Ethics Review Board in Lund.

Results

Sexual abuse rates

Of the total sample of 2,324 girls and 2,015 boys, 65% of the girls and 23% of the boys reported some form of sexual abuse experience. Of the girls who reported experiences of sexual abuse, 10.0% reported non-contact abuse, 69.2% contact abuse without penetration and 20.8% penetrating abuse. Non-contact abuse was reported by 18.4% of the boys who reported sexual abuse, while 57.3% reported contact abuse and 24.3% penetrating abuse (Table 1). Participants with experiences of different kinds of abuse were categorized according to the most severe kind of abuse they had reported.

Disclosure rates and recipients of disclosure

Table 1 shows that out of the 1,962 participants who reported experience of sexual abuse, 261 (17.3%) of the girls and 208 (45.5%) of the boys did not answer the questions about disclosure. A data analysis of these non-completers was done by carrying out a logistic regression analysis separately for girls and boys including abuse severity, parents’ employment, family structure, educational program, perceived mental health and immigrant background as independent variables and non-completers as dependent variable. Boy non-completers were about two times more likely to have one or both parents unemployed than completers (aOR = 2.18, CI = 1.49–3.19). Girl non-completers were significantly more often enrolled in a vocational educational program (aOR = 1.40, CI = 1.05–1.86) and reported less often exposure to penetrating abuse compared to completers (aOR = .40, CI = .23–.68). Concerning all other variables in the analyses, there were no significant differences between completers and non-completers.

All participants had very low numbers of missing answers for questions not related to sexual abuse, for example back- ground variables, perception of parents when growing up and questions about perceived mental health (0–2.8% of those who had not answered to the disclosure questions and .4–1.8% of those who had answered to the disclosure questions).

In the following, results from those adolescents who completed the questions about disclosure are presented. Table 2 shows that girl completers reported that they had talked to somebody about the sexual abuse significantly more often than boys. Most of them named one or two of the recipients listed in Table 2 (girls, 75.9%, boys 78.1%). Six or more different recipients were marked by 2.1% of the disclosing girls and 4.8% of the disclosing boys. Both girls and boys mentioned most often a friend of their own age as the person they had disclosed to.

Table 2 also shows that few young persons (8.3%) had talked to a professional about the abuse. Professionals included teachers, social workers, nurses or other persons working professionally with children and adolescents. These persons are required by Swedish law (mandatory reporting) to report to the social authorities all cases of child sexual abuse that come to their attention. A disclosure made to a professional resulted significantly more often (33.9%) in a report to the social authorities or the police, compared to a disclosure made to someone other than a professional, 4.4% (OR = 11.17, CI = 7.10–17.58). In all, 6.8% of the adolescents answered that the incident had been reported to the social authorities or police.

Table 3 shows that the severity of the sexual abuse seemed to influence the young person’s decision to at least disclose the experience to someone. The severity also affected disclosure to a member of their families, to a professional, and if the abuse was reported to the social authorities or to the police.

G .P

rieb e,C

.G .Sved

in /

C h

ild A

b u

se &

N eglect

3 2

(2 0

0 8

) 10

9 5

– 110

8 10

9 9

Table 1 Participants: completers and non-completers.

Type of sexual abuse Could talk to someone about the sexual abuse

Girls, n = 1,505 Boys, n = 457

All, n (%) Completers, n (%) Non-completers, n (%) OR (CI 95%) All, n (%) Completers, n (%) Non-completers, n (%) OR (CI 95%)

Non-contact 150 (10.0) 116 (9.3) 34 (13.0) 1.46 (.97–2.19) 84 (18.4) 45 (18.8) 39 (18.7) 1.05 (.65–1.68) Contact 1,042 (69.2) 848 (68.2) 194 (74.3) 1.35 (1.00–1.83) 262 (57.3) 152 (61.0) 110 (52.9) .72 (.49–1.04) Penetrating 313 (20.8) 280 (22.5) 33 (12.7) .50 (.34–.74) 111 (24.3) 52 (20.9) 59 (28.4) 1.50 (.98–2.30)

Total 1,505(100) 1,244 (100) 261 (100) 457 (100) 249 (100) 208 (100)

1100 G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108

Table 2 Disclosure of sexual abuse, rates (n = 1,493).

Could talk to . . .a Girls, n = 1,244 Boys, n = 249

n (%) n (%) OR (CI 95%)

Nobody 231 (18.6) 77 (30.9) 1.96 (1.45–2.66)

Somebody 1,013 (81.4) 172 (69.1) .51 (.38–.69) Friend of my age 781 (62.8) 111 (44.6) .48 (.36–.63) Mother 351 (28.2) 43 (17.3) .53 (.37–.75) Father 160 (12.9) 34 (13.7) 1.07 (.72–1.60) Sibling 187 (11.6) 29 (11.6) .75 (.49–1.13) Other person 128 (10.3) 48 (19.3) 2.08 (1.45–3.00) Professional 117 (9.4) 7 (2.8) .28 (.13–.61) Adult relative or friend 111 (8.9) 33 (13.3) 1.56 (1.03–2.36)

The incident was reported to social authorities or police 91 (7.3) 11 (4.4) .59 (.31–1.11)

a Several alternatives possible.

The more severe the sexual abuse was, the more seldom both girls and boys had talked to mother, father or a sibling. Disclosure to a professional was associated with higher rates of more severe abuse (contact abuse with or without penetration compared to non-contact abuse) for girls (OR = 3.12, CI = 1.13–8.61), but with lower rates of more severe abuse for boys (OR = .15, CI = .03–.71). Talking to a professional was also associated with being abused by an older person (age difference ≥5 years vs. <5 years) for girls (OR = 1.65, CI = 1.10–2.49), while there was no such association for boys.

For boys, all but one of the reports to social authorities or police concerned non-contact abuse. For girls, the proportion of reported incidents was highest for non-contact abuse, but even cases of more severe abuse were reported.

Of the disclosers, 42.6% of the boys and 37.9% of the girls mentioned “friend of my own age” as the only recipient. A friend of one’s own age may include both peer(s) and romantic partner. Disclosure to a friend of one’s own age was associated with higher rates of more severe abuse (non-contact abuse vs. contact abuse with or without penetration; girls OR = 1.92, CI = 1.22–3.03, boys OR = 2.58, CI = 1.04–6.43) and being abused by a peer (age difference ≥5 years vs. <5 years; girls OR = 2.35, CI = 1.80–3.05, boys OR = 2.40, CI = 1.22–4.73).

Univariate analyses

Table 4 shows the associations between disclosure and abuse characteristics, socio-demographic variables, parental bonding and mental health.

Non-disclosing girls reported more often penetrating abuse, less frequent abuse, and abuse by a family member, a relative or a friend, compared to disclosing girls. They also reported less frequently that the perpetrator had used alcohol or drugs at the first abuse occasion. Non-disclosing girls were more often first or second generation immigrants compared to native Swedes.

Girl non-disclosers had significantly lower scores on the care dimension and higher scores on the overprotection dimen- sion on the Parenting Bonding Instrument (t-test care: t(df) = 5.74(1235), p < .001, overprotection: t(df) = −2.91(1235), p = .007). Girl non-disclosers more often perceived their parents as less caring in combination with lower overprotection or higher overprotection, than girl disclosers.

Non-disclosing boys were sexually abused more often by a family member, a relative or a friend than boy disclosers. Boy non-disclosers were also more often studying at a vocational program and lived more often together with both parents than disclosing boys.

Boy non-disclosers had significantly lower scores on the care dimension and higher scores on the overprotection dimen- sion on the Parenting Bonding Instrument (t-test care: t(df) = 2.18(246), p = .030, overprotection: t(df) = −2.00(246), p = .046). Non-disclosing boys perceived their parents during their childhood more often as less caring in combination with lower overprotection, than boy disclosers.

Adolescents who reported sexual abuse experiences reported significantly more symptoms at the Mental Health Scale, indicating that their perceived mental health was poorer compared to adolescents without these experiences (girls: M(SD): abused 15.38 (4.68), not abused 14.00 (4.68), t (df) = −6.79 (2298), p < .001; boys: M(SD): abused 14.21 (4.71), not abused 12.65 (4.37), t (df) = −6.31 (703.20), p < .001). The effect sizes were moderate (girls: .30, boys: .34). Adolescents who reported sexual abuse had also significantly more often a total score above the cut off in the Mental Health Scale (girls: OR = 1.60, CI = 1.30–1.97; boys: OR = 2.10, CI = 1.60–2.77). In general, non-disclosers reported more symptoms than disclosers (girls: M(SD): discloser 15.27 (4.71), non-discloser 16.26 (4.7), t (df) = −2.89 (1232), p = .004; boys: M(SD): discloser 13.97 (4.55), non-discloser 15.53 (4.80), t (df) = −2.45 (246), p = .015). The effect sizes were low for girls (.21) and moderate for boys (.34). On the other hand, there were no significant differences between disclosers and non-disclosers when the results were analyzed separately for non-contact abuse, contact abuse and penetrating abuse. Disclosers and non-disclosers did not differ concerning a total score above the cut off in the Mental Health Scale (Table 4).

G .P

rieb e,C

.G .Sved

in /

C h

ild A

b u

se &

N eglect

3 2

(2 0

0 8

) 10

9 5

– 110

8 110

1

Table 3 Severity of the sexual abuse and disclosure rates (n = 1,493).

Could talk to . . .a Girls, n = 1,244 Boys, n = 249

Non-contact, n = 116 Contact, n = 848 Penetrating, n = 280 p Non-contact, n = 45 Contact, n = 152 Penetrating, n = 52 p n (%) n (%) n (%) n (%) n (%) n (%)

Nobody 6 (5.2) 153 (18.0) 72 (25.7) <.001b 10 (22.2) 47 (30.9) 20 (38.5) .226b

Somebody Friend of my age 77 (66.4) 539 (63.6) 165 (58.9) .267b 20 (44.4) 74 (48.7) 17 (32.7) .135b

Mother 76 (65.5) 232 (27.4) 43 (15.4) <.001b 21 (46.7) 20 (13.2) 2 (3.8) <.001b

Father 42 (36.2) 102 (12.0) 16 (5.7) <.001b 18 (40.0) 14 (9.2) 2 (3.8) <.001b

Sibling 35 (30.2) 125 (14.7) 27 (9.6) <.001b 11 (24.4) 17 (11.2) 1 (1.9) .003b

Other person 12 (10.3) 85 (10.0) 31 (11.1) .882b 7 (15.6) 31 (20.4) 10 (19.2) .770b

Professionalc 4 (3.4) 61 (7.2) 52 (18.6) <.001b 4 (8.9) 2 (1.3) 1 (1.9) .035d

Adult relative or friend 18 (15.5) 72 (8.5) 21 (7.5) .029b 10 (22.2) 19 (12.5) 4 (7.7) .099b

The incident was reported to social authorities or police 17 (14.7) 47 (5.5) 27 (9.6) <.001b 10 (22.2) 0 1 (1.9) <.001b

a Several alternatives possible. b �2 . c “Professional” includes teacher, social worker, nurse or some other person who professionally works with children and youth. d Fisher’s exact test.

1102 G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108

Table 4 Disclosure and abuse characteristics, socio-demographic variables, parental bonding and mental health.

Variable Girls, n = 1,131–1,244 Boys, n = 227–249

Discloser, n (%) Non-discloser, n (%) Discloser, n (%) Non-discloser, n (%)

Abuse characteristics, ever Severity

Non-contact 110 (10.9) 6 (2.6)*** 35 (20.3) 10 (13.0) Contact 695 (68.6) 153 (66.2)** 105 (61.0) 47 (61.0) Penetration 208 (20.5) 72 (31.2)*** 32 (18.6) 20 (26.0)

Frequency Once 394 (39.3) 94 (41.8) 78 (47.0) 27 (38.6) 2–5 times 437 (43.6) 107 (47.6) 67 (40.4) 33 (47.1) >5 times 172 (17.1) 24 (10.7)* 21 (12.7) 10 (14.3)

Abuse characteristics, first abuse occasion Relation to perpetrator

Stranger 619 (63.0) 89 (41.4)*** 103 (64.0) 30 (43.5)** Family/relative 35 (3.6) 17 (7.9)*** 2 (1.2) 4 (5.8)* Friend/acquaintance 329 (33.5) 109 (50.7)*** 56 (34.8) 35 (50.7)*

Offender on alcohol/drugs No 561 (61.4) 149 (68.7) 88 (55.0) 46 (64.8) Yes 353 (38.6) 68 (31.3)* 72 (45.0) 25 (35.2)

Victim on alcohol/drugs No 787 (78.9) 180 (80.0) 110 (67.1) 52 (72.2) Yes 210 (21.1) 45 (20.0) 54 (32.9) 20 (27.8)

Age difference <5 years 397 (39.2) 97 (42.0) 94 (54.7) 41 (53.2) ≥5 years 449 (44.3) 102 (44.2) 61 (35.5) 25 (32.5) Unknown 167 (16.5) 32 (13.9) 17 (9.9) 11 (14.3)

Victims age ≤15 years No 379 (38.7) 88 (39.6) 92 (56.1) 34 (50.7) Yes 601 (61.3) 134 (60.4) 72 (43.9) 33 (49.3)

Physical violence No 791 (78.1) 172 (74.5) 149 (86.6) 68 (88.3) Yes 222 (21.9) 59 (25.5) 23 (13.4) 9 (11.7)

Socio-demographic variables First or second generation immigrants

No 659 (65.8) 134 (58.8) 105 (61.4) 44 (57.9) Yes 343 (34.2) 94 (41.2)* 66 (38.6) 32 (42.1)

Educational program Academic 702 (69.3) 152 (65.8) 111 (64.5) 36 (46.8) Vocational 311 (30.7) 79 (34.2) 61 (35.5) 41 (53.2)**

Family structure Living with both mom and dad 555 (55.2) 123 (53.5) 97 (56.4) 53 (70.7) Not living with both mom and dad 450 (44.8) 107 (46.5) 75 (43.6) 22 (29.3)*

Parents’ socioeconomic status (ISEI) 1–25th percentile 245 (25.3) 56 (26.7) 53 (32.7) 16 (24.6) 26–50th percentile 250 (25.8) 66 (31.4) 44 (27.2) 22 (33.8) 51–75th percentile 295 (30.5) 55 (26.2) 33 (20.4) 21 (32.3) 76–100th percentile 178 (18.4) 33 (15.7) 32 (19.8) 6 (9.2)

Both parents employed Yes 671 (66.7) 140 (60.9) 112 (66.3) 46 (60.5) No 335 (33.3) 90 (39.1) 57 (33.7) 30 (39.5)

Parental bonding High care/low overprotection 649 (64.5) 108 (46.8)*** 99 (57.6) 28 (36.8)* High care/high overprotection 111 (11.0) 25 (10.8) 21 (12.2) 13 (17.1) Low care/low overprotection 164 (16.3) 57 (24.7)*** 34 (19.8) 26 (34.2)** Low care/high overprotection 82 (8.2) 41 (17.7)*** 18 (10.5) 9 (11.8)

Mental health scale ≥19 No 724 (72.0) 152 (66.7) 141 (82.0) 56 (73.7) Yes 282 (28.0) 76 (33.3) 31 (18.0) 20 (26.3)

Note: *p < .05, **p < .01, ***p < .001, bivariate logistic regression cOR, see also Tables 5a and 5b.

G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108 1103

Table 5a Summary of logistic regression analysis (LR backward) for variables predicting non-disclosure of sexual abuse, girls.

Variable cOR Step 1 Final step (step 5)

aOR CI (95%) pa aOR CI (95%) pa

Sexual abuse category .022 .031 Non-contact (ref) Contact 4.04 4.35 1.53–12.38 .006 4.07 1.44–11.52 .008 Penetrating 6.35 4.30 1.44–12.82 .009 3.87 1.31–11.43 .014

Frequency of the sexual abuse <.001 .001 Once (ref) 2–5 times 1.03 .90 .63–1.27 .532 .92 .66–1.31 .655 More than 5 times .59 .32 .18–.57 <.001 .34 .19–.60 <.001

Relation to the offender <.001 <.001 Stranger (ref) Family/relative 3.38 2.92 1.46–5.82 .002 3.02 1.54–5.92 .001 Friend/acquaintance 2.30 1.89 1.32–2.70 .001 1.94 1.37–2.75 <.001

Offender on alcohol or drugs .73 .75 .53–1.08 .119 – – – Immigrant 1.35 .94 .66–1.33 .722 – – – Educational program 1.17 .88 .62–1.26 .493 – – – Family structure 1.07 1.00 .71–1.39 1.0 – – –

Parental bonding .002 .001 High care, low overprotection (ref) High care, high overprotection 1.35 1.15 .67–2.00 .611 1.14 .66–1.96 .641 Low care, low overprotection 2.09 2.02 1.35–3.04 .001 2.04 1.36–3.05 .001 Low care, high overprotection 3.01 2.02 1.21–3.37 .007 2.04 1.23–3.37 .005

Notes: Description of the variables included in the analysis: disclosure of sexual abuse – yes = 0, no = 1 (dependent variable); sexual abuse category – non- contact abuse = 0 (ref), contact abuse = 1, penetrating abuse = 2; frequency of the sexual abuse – once = 0 (ref), 2–5 times = 1, more than five times = 2; relation to the offender at first abuse occasion – stranger = 0 (ref), family/relative = 1, friend/acquaintance = 2; offender on alcohol or drugs at first abuse occasion – no = 0, yes = 1; immigrant – no = 0, yes = 1; educational program – academic = 0, vocational = 1; family structure – living with both mom and dad = 0, not living with both mom and dad = 1; parental bonding – high care, low overprotection (optimal bonding) = 0 (ref), high care, high overprotection (affectionate constraint) = 1, low care, low overprotection (absent/weak bonding) = 2, low care, high overprotection (affectionless control) = 3.

a Wald statistic.

Multivariate analyses—predictors of non-disclosure

As described above, a number of variables were associated with disclosure versus non-disclosure, when examined sepa- rately. These variables were four sexual abuse characteristics (sexual abuse category, frequency of the sexual abuse, relation to the offender at first abuse occasion and offender on alcohol or drugs at first abuse occasion), three socio-demographic variables (immigrant, educational program and family structure) and parental bonding. A logistic regression analysis (LR backward) was conducted to examine if the associations remained when controlled for the other variables.

Tables 5a and 5b show that four variables remained significant for girls and three for boys each as overall predictors of disclosure versus non-disclosure in the final logistic regression model. Parental bonding was the only common variable for both genders. Specifically, girls were less likely to disclose when (a) they had been exposed to contact or penetrating abuse compared to non-contact abuse, (b) they had been exposed to a single abuse occasion compared to more than five abuse occasions, (c) the offender at first/only abuse occasion had been a family member/relative or friend/acquaintance compared to a stranger and (d) they had perceived their parents as non-caring when growing up. Boys were less likely to disclose when (a) they participated in a vocational educational program, (b) they were living with both mom and dad and (c) they had perceived their parents as either caring and overprotective or non-caring and not overprotective compared to caring and not overprotective.

Discussion

The aim of this study was to investigate disclosure rates and disclosure patterns and to examine predictors of non- disclosure in a sample of male and female adolescents with self-reported experiences of sexual abuse. The main results from this study may be summarized in six main findings.

Disclosure rate

The disclosure rate (81% for girls and 69% for boys) was high in our study of adolescents compared to retrospective studies of adults presenting childhood disclosure rates between 31% and 42% (Arata, 1998; Finkelhor et al., 1990; Hanson et al., 1999; Smith et al., 2000; Tang, 2002). Studies of adolescents show high rates similar to those in our study (Edgardh & Ormstad, 2000; Fergusson et al., 1996; Helweg-Larsen & Larsen, 2006; Kellogg & Huston, 1995; Kogan, 2004). One explanation of

1104 G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108

Table 5b Summary of logistic regression analysis (LR backward) for variables predicting non-disclosure of sexual abuse, boys.

Variable cOR Step 1 Final step (step 5)

aOR CI (95%) pa aOR CI (95%) pa

Sexual abuse category .736 Non-contact (ref) Contact 1.57 1.49 .52–4.26 .457 – – – Penetrating 2.19 1.57 .45–5.49 .483 – – –

Frequency of the sexual abuse .669 – Once (ref) 2–5 times 1.42 1.38 .66–2.92 .396 – – – More than 5 times 1.38 1.05 .37–3.00 .925 – – –

Relation to the offender .156 – Stranger (ref) Family/relative 6.87 5.99 .74–48.24 .093 – – – Friend/acquaintance 2.15 1.54 .77–3.10 .225 – – –

Offender on alcohol or drugs .66 .63 .32–1.27 .195 – – – Immigrant 1.16 .81 .39–1.68 .570 – – – Educational program 2.07 2.96 1.47–5.97 .002 3.20 1.65–6.21 .001 Family structure .54 .38 .18–.79 .009 .43 .22–.86 .017

Parental bonding .050 .013 High care, low overprotection (ref) High care, high overprotection 2.19 2.43 .93–6.35 .070 2.76 1.12–6.80 .028 Low care, low overprotection 2.70 2.76 1.24–6.12 .013 3.20 1.49–6.91 .003 Low care, high overprotection 1.77 1.10 .32–3.77 .880 1.24 .41–3.71 .704

Notes: Description of the variables included in the analysis: disclosure of sexual abuse – yes = 0, no = 1 (dependent variable); sexual abuse category – non- contact abuse = 0 (ref), contact abuse = 1, penetrating abuse = 2; frequency of the sexual abuse – once = 0 (ref), 2–5 times = 1, more than five times = 2; relation to the offender at first abuse occasion– stranger = 0 (ref), family/relative = 1, friend/acquaintance = 2; offender on alcohol or drugs at first abuse occasion – no = 0, yes = 1; immigrant – no = 0, yes = 1; educational program – academic = 0, vocational = 1; family structure – living with both mom and dad = 0, not living with both mom and dad = 1; parental bonding – high care, low overprotection (optimal bonding) = 0 (ref), high care, high overprotection (affectionate constraint) = 1, low care, low overprotection (absent/weak bonding) = 2, low care, high overprotection (affectionless control) = 3.

a Wald statistic.

the differences between retrospective studies of adolescents and adults could be that there is less recall bias in studies of adolescents as it is easier for them to remember that they disclosed to someone. Another explanation is that the studies of adolescents included peer abuse while several of the studies of adults did not specify if peer abuse was included and may have excluded it. The mean age at the first abuse occasion was relatively high in our study (14.4 years), which probably was due to the fact that peer abuse was included. We know from other studies that high disclosure rates are related to sexual abuse during adolescence (London, Bruck, Ceci, & Shuman, 2005) and that adolescents are more likely to make the decision to disclose compared to preschool children (Sorensen & Snow, 1991).

The disclosure rate for boys was higher in our study than in Edgardh and Ormstad (2000) study. Nevertheless, the pattern in both studies is that the rates for boys are lower than those for girls.

In this study, it was more likely that girls who had been exposed to non-contact abuse disclosed compared to girls exposed to contact abuse with or without penetration. This finding is supported by the lower disclosure rates in studies only including penetrating sexual abuse (Hanson et al., 1999; Smith et al., 2000). On the other hand, Kogan (2004) found that girls who experienced penetrating abuse were more likely to disclose to an adult than to non-disclose.

Disclosures are hidden from adult society

Even if the disclosure rate was high, child sexual abuse was largely hidden from the adult society. By far the most com- mon recipient of disclosure from both girls and boys was a friend of their own age and as many as 42% had only talked to a friend of their own age and nobody else. This was most evident when it came to peer abuse and one explanation could be that young people prefer to talk to a friend about their experiences and do not want to involve their parents. Our find- ings are in line with Kogan (2004), and one interpretation could be the adolescent’s fear of embarrassment, blame, and accusations.

Data from other studies about disclosure during childhood suggest that most children make their initial disclosure to a parent or a parent-figure (Arata, 1998; Jensen et al., 2005; Paine & Hansen, 2002) while studies of older adolescents (Kellogg & Huston, 1995) and adult women (Ruggiero et al., 2004; Smith et al., 2000) about lifetime disclosure show that the victim would most often make the disclosure to a friend. In our study, we did not distinguish between initial disclosure and later disclosure. It can be presumed that our data reflect the increasing importance of friends at the same age (both peers and romantic partners) in late adolescence and the inclusion of peer abuse in the study. Even if they perceived their parents as supportive, those who disclose in late adolescence may prefer to talk to friends about sexual abuse, especially peer abuse.

G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108 1105

Gender differences in disclosure patterns

There were several interesting differences in the disclosure patterns between the answers from girls and boys. Even when boys reported experiences of sexual abuse, many were hesitant to answer follow-up questions about the abuse, resulting in a high number of non-completers. This is a finding per se but, as a consequence, the results for boys have to be regarded with some caution. Boy completers had significantly more often not disclosed to anyone and fewer of them had talked to their mother or a friend but had instead more often talked to someone in the category “another” compared to girls. Boys from vocational educational programs were overrepresented among non-disclosers. For girls, disclosure was not associated with educational program. In Sweden vocational programs are often either male or female dominated while academic programs are more mixed. This means that boys at vocational programs often live in a male school context which may make it more difficult for them to disclose sexual abuse. Our results might indicate that boys make other choices or encounter other difficulties than girls when disclosing experience of sexual abuse. Boys who have been sexually abused by men do often report confusion over their sexual identity, fear of being regarded as homosexual by others and concern for being a potential offender or being regarded by others as a potential offender (Durham, 2003; Teram, Stalker, Hovey, Schachter, & Lasiuk, 2006; Watkins & Bentovim, 1992). Men who had been abused by a woman have reported that they felt that in meeting with health professionals, some of these might have expectations like “this should be every man’s dream” (Teram et al., 2006). Especially peers’ negative reactions, both real and expected reactions, strongly influenced male adolescents’ decision not to tell about the sexual abuse they had experienced (Durham, 2003). Theories about masculinities and males sexualities offer the concept of hegemonic masculinity – the norm that “real men” are heterosexual and powerful (Connell, 2000; Kimmel, Hearn, & Connell, 2005) – and homophobia as a consequence of this (Kimmel, 1994). Sexually abused boys may feel that being a victim of sexual abuse is not compatible with appropriate masculinity (Browne, 1991; Durham, 2003; Holmes, Offen, & Waller, 1997).

The professional system

Few adolescents exposed to sexual abuse reported that they had talked to a professional about the abuse (9% of the girls and 3% of the boys) or that the incident had been reported to social authorities or the police (7% of the girls and 4% of the boys). These findings are well in line with other studies when it comes to professional contacts (Collings, 1995; Edgardh & Ormstad, 2000) and the frequency of incidents reported to the authorities is usually in the range between 6% and 15% (Arata, 1998; Hanson et al., 1999; Helweg-Larsen & Larsen, 2006; Smith et al., 2000; Tang, 2002). In our study, girls who had been exposed to penetrating abuse had more often been in contact with a professional than girls exposed to less severe abuse. This indicates that girls who come to see a professional are more severe cases and in greater need for help. On the other hand, only one out of 52 boys exposed to penetrating abuse had been in contact with a professional, underlining men’s difficulties in talking about their abuse experiences. Less severe abuse was relatively more often reported to social authorities compared to severe abuse, especially if the victim was a boy.

The low rate of disclosures to professionals could partially be explained by older children’s fears that a report to the authorities will be made (mandatory reporting). The support duty is maybe hindered by the report duty but so far there is no evidence that the disclosure rates to professionals would be substantially higher, if there was no mandatory reporting. The numbers are low in all studies that have come to our attention representing different countries and regions such as USA/Southern Alabama (Arata, 1998), USA/South Carolina (Hanson et al., 1999; Smith et al., 2000), South Africa (Collings, 1995), Denmark (Helweg-Larsen & Larsen, 2006), and Hong Kong (Tang, 2002).

It is important to note in this and other retrospective studies the possibility that not all adolescents who actually have been in contact with a professional were aware of this or remembered it correctly when asked in the study. Berliner and Conte (1995) found that, although all children in their study had been seen at least once by a person from the professional system, only 71% recalled this at follow-up on average 3.5 years later. Another source of recall bias could also be that contact abuse is more easily remembered than non-contact abuse when answering questions in a questionnaire, but on the other hand, as Bouvier et al. (1999) point out, the individual experience of the abuse may be very different irrespective of the severity of the abuse.

Predictors of non-disclosure

Even if a number of variables were associated with disclosure versus non-disclosure only few of them remained associated when controlled for in the logistic regression model. For girls, the severity and the frequency of the sexual abuse, the relation to the offender at first abuse occasion and the perception of parental bonding when growing up were identified as predictors of non-disclosure. For boys, the predictors of non-disclosure were educational program, family structure and parental bonding. Thus, the only predictor common for both genders was parental bonding. Compared to other types of parental bonding, young persons who had perceived their parents as caring and not overprotective were most likely to disclose. A plausible interpretation seems to be that growing up in a caring and not overprotective family climate facilitates young people’s talking to other people including both peers and parents about their abusive experience. Our results are, in a way, supported by the findings in studies of children undergoing forensic interviewing, in which abused children who had supportive caretakers were more likely to disclose than those who had not (Elliott & Briere, 1994; Lawson & Chaffin, 1992).

1106 G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108

There are few studies reporting on associations between disclosure/non-disclosure and abuse characteristics or socio- demographic variables. Associations similar to those we have found have been found between lower disclosure rates and contact abuse (Arata, 1998) and higher disclosure rate and the offender being a stranger (Arata, 1998; Hanson et al., 1999; Smith et al., 2000). In our study, socio-demographic variables had some importance only for the boys. The finding that boys living with both parents are less likely to disclose seems counter-intuitive and needs further exploration. As found in previous studies (Hanson et al., 1999; Kellogg & Huston, 1995) the associations between disclosure and socio-demographic variables seem to be weak for both girls and women.

Mental health

Both girls and boys with sexual abuse experiences reported significantly more mental health symptoms than participants without this experience, but the scores were not very high in either group. Concerning sexual abuse in general, girls and boys who had disclosed to somebody reported significantly better health than those who had not but when data were analyzed separately for non-contact, contact and penetrating abuse the findings could not be confirmed. Hanson et al. (1999) found no significant differences in prevalence for PTSD and Major Depression among those who disclosed versus not disclosed. This could be explained by the fact that disclosure for some is a relief but for others adds to the burden of being a victim. It might also reflect that a disclosure per se is not enough in order to get support and help to recover from the negative impact of sexual abuse.

Our results must be interpreted with some caution. First, there was a 23% general drop-out rate from the study and a high drop-out rate among boys answering the follow-up questions concerning disclosure. A participation rate of 77% may be seen as acceptable to good since 10% of the pupils in the upper secondary schools in Sweden are usually absent on any day for a variety of reasons (illness, practical occupational experience). Truancy could also account for 3–4%, according to another Swedish study (Sundell, El-Khouri, & Månsson, 2005), and there is good reason to believe that the reported prevalence of sexual abuse would have been a little bit higher if these pupils had been present, since these student also are at higher risk for being a victim of crimes like theft and rape (Sundell et al., 2005). The number of boy non-completers not answering the disclosure questions is more problematic and the results may have been different as concerns both variety and statistical power had the response rate been higher. Since the analyses concerning non-completers did not show any decisive differences between this group and the others we decided to keep the boys in the analysis since the drop-out could be seen as a result per se, strengthening the view that boys experience greater difficulties in talking about their abuse history. Second, the study design allowed participants to report several kinds of sexual abuse with different severity or several abuse occasions without the possibility of identifying what kind of abuse or occasion they actually talked about with different persons. Another shortcoming of the study design, compared with for example Kogan (2004), was that the timing of the disclosure (immediate or delayed) was not measured. Third, there is the general validity problem that can arise in a retrospective study based only on self-reports (Hardt & Rutter, 2004). Recall bias about both the sexual abuse events and the disclosure process are to be expected but perhaps this is less of a problem with older adolescents than with adults since the self-reported events are closer in time. Fourth, the participants may be reluctant to report sexual abuse in a school setting as they are afraid of being identified. The high rates of reported abuse events contradict this but this could perhaps contribute to the drop-out rate on sensitive questions in the follow-up questions of sexual abuse. Fifth, due to the design of this study, the analyses were limited to between-group differences and it was not possible to analyze within-person-differences related to different situations and contexts. A qualitative approach would have fit these kinds of questions better with the pros and cons connected with this kind of design. Sixth, another topic is the students’ understanding of the wording in the questions. A sentence like “against your will” could for example have different meanings to different individuals depending on their earlier life events, gender, and culture. Hardt and Rutter (2004) addressed the problem that the ways in which the participants were asked about adverse experiences are likely to contribute to the validity of the reports. According to them, interviews allow a clarification of different understandings, but, on the other hand, questionnaires have a possible advantage in anonymity, which may facilitate valid response to questions on sensitive issues. Finally, it may be difficult to compare results from different studies as there are differences concerning the wording of the questions about sexual abuse and disclosure, the study settings, age limits and so on.

Implications for practice

Professionals, especially in the school system, need to be more aware of the finding that few sexually abused children seek help from professionals or other adults and that support offers should be directly addressed not only to the vulnerable young persons themselves but also to peers who wish to help a friend. For example, information about sexual abuse and available support for victims and peers should be included in family and sex education at schools. Education of teachers is also needed since sexuality in general, and sexual abuse in special, are sensitive and difficult topics to talk about. To provide the support system with education and guidelines is probably also crucial when it comes to support for both victims and their peers. Even if not all young people exposed to sexual abuse are in need of therapeutic help it is a matter of concern that so few abused young people, especially boys, are seen by a professional for assessment of their needs. It is important to take steps to achieve this since there is a two to fourfold increased risk for different psychiatric problems later in life (Fergusson & Mullen, 1999).

G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108 1107

Conclusion

The important message from this study is that sexual abuse is largely hidden from the adult society, especially for pro- fessionals and the legal system. Disclosing child sexual abuse is a complex process and when young persons disclose sexual abuse, friends seem to be the most important receivers of the information. The multiple logistic regression analysis shows that children’s difficulties with disclosure are associated with both abuse factors such as severity and frequency, with the relation to the perpetrator and with family background factors such as family structure and living in a caring environment. The implications for the society might be that it must consider relying to a lesser extent on established organizations for information and instead must find ways to give young people better information and guidance about how to support a sex- ually abused peer and how to mediate help when necessary. There are also different disclosure patterns for girls and boys. This indicates that a gender perspective may be helpful when developing support efforts and education to professionals. As a complement to our study, qualitative research is needed for a better understanding of young peoples’ choices—why, when and to whom they disclose or not!

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, L., & Conte, J. R. (1995). The effects of disclosure and intervention on sexually abused children. Child Abuse & Neglect, 19, 371–384. Bouvier, P., Halpérin, D., Rey, H., Jaffé, P. D., Laederach, J., Mounoud, R.-L., & Pawlak, C. (1999). Typology and correlates of sexual abuse in children and youth:

Multivariate analyses in a prevalence study in Geneva. Child Abuse & Neglect, 23, 779–790. Bradley, A. R., & Wood, J. M. (1996). How do children tell? The disclosure process in child sexual abuse. Child Abuse & Neglect, 20, 881–891. Browne, A. (1991). The victim’s experience: Pathways to disclosure. Psychotherapy: Theory/Research/Practice/Training, 28, 150–156. Collings, S. J. (1995). The long-term effects of contact and noncontact forms of child sexual abuse in a sample of university men. Child Abuse & Neglect, 19,

1–6. Connell, R. W. (2000). The men and the boys. Cambridge: Polity. Derogatis, L. R. (1977). SCL-90. Administration, scoring & procedures manual – I for the R(evised) version. Baltimore: John Hopkins University School of Medicine. DeVoe, E. R., & Faller, K. C. (1999). The characteristics of disclosure among children who may have been sexually abused. Child Maltreatment, 4,

217–227. Durham, A. W. (2003). Young men surviving child sexual abuse. Research stories and lessons for therapeutic practice. Chichester: Wiley. Edgardh, K. (2001). Adolescent sexuality and sexual abuse: A Swedish perspective. Unpublished Doctoral Dissertation, Karolinska Institutet, Stockholm. Edgardh, K., & Ormstad, K. (2000). Prevalence and characteristics of sexual abuse in a national sample of Swedish seventeen-year-old boys and girls. Acta

Paediatrica, 88, 310–319. Elliott, D. M., & Briere, J. (1994). Forensic sexual abuse evaluations of older children: Disclosures and symptomatology. Behavioral Sciences & the Law, 12,

261–277. Fergusson, D. M., Lynskey, M. T., & Horwood, L. J. (1996). Childhood sexual abuse and psychiatric disorder in young adulthood. Part 1: The prevalence of

sexual abuse and factors associated with sexual abuse. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 1355–1364. Fergusson, D. M., & Mullen, P. E. (1999). Childhood sexual abuse. An evidence based perspective London: Sage Publications Ltd. Finkelhor, D., Hotaling, G., Lewis, I. A., & Smith, C. (1990). Sexual abuse in a national survey of adult men and women: Prevalence, characteristics, and risk

factors. Child Abuse & Neglect, 14, 19–28. Ganzeboom, H. B. G., de Graaf, P. M., & Treiman, D. J. (1992). A standard international socio-economic index of occupational status. Social Science Research,

21, 1–56. Hammarén, N., & Johansson, T. (2001). Ungdom och sexualitet i gränslandet.Delrapport 1, Ung och kåt? Om ungdomars sexuella berättelser [Youth and sexuality

in the borderland. Report part 1: Young and horny? About young persons sexual narratives]. Skövde: Institutionen för beteendevetenskap, Högskolan i Skövde.

Hammarén, N., & Johansson, T. (2002). Könsordning eller könsoordning? [Gender order or gender disorder?] (No. 2). Göteborg: Centrum för kulturstudier/Forum för studier av samtidskultur, Göteborgs universitet.

Hanson, R. F., Resnick, H. S., Saunders, B. E., Kilpatrick, D. G., & Best, C. (1999). Factors related to the reporting of childhood rape. Child Abuse & Neglect, 23, 559–569.

Hardt, J., & Rutter, M. (2004). Validity of adult retrospective reports of adverse childhood experiences: Review of the evidence. Journal of Child Psychology and Psychiatry, 45, 260–273.

Helweg-Larsen, K., & Larsen, H. B. (2006). The prevalence of unwanted and unlawful sexual experiences reported by Danish adolescents: Results from a national youth survey in 2002. Acta Paediatrica, 95, 1270–1276.

Holmes, G. R., Offen, L., & Waller, G. (1997). See no evil, hear no evil, speak no evil: Why do relatively few male victims of childhood sexual abuse receive help for abuse-related issues in adulthood? Clinical Psychology Review, 17, 69–88.

Jensen, T. K., Gulbrandsen, W., Mossige, S., Reichelt, S., & Tjersland, O. A. (2005). Reporting possible sexual abuse: A qualitative study on children’s perspectives and the context for disclosure. Child Abuse & Neglect, 29, 1395–1414.

Kellogg, N. D., & Huston, R. L. (1995). Unwanted sexual experiences in adolescents. Patterns of disclosure. Clinical Pediatrics, 34, 306–312. Kimmel, M. S. (1994). Masculinity as homophobia: Fear, shame, and silence in the construction of gender identity. In H. Brod & M. Kaufman (Eds.), Theorizing

masculinities (pp. 119–141). London: Sage Publications. Kimmel, M. S., Hearn, J., & Connell, R. W. (2005). Handbook of studies on men and masculinities. Thousand Oaks, CA: Sage Publications. Kogan, S. M. (2004). Disclosing unwanted sexual experiences: Results from a national sample of adolescent women. Child Abuse & Neglect, 28, 147–165. Lawson, L., & Chaffin, M. (1992). False negatives in sexual abuse disclosure interviews. Incidence and influence of caretaker’s belief in abuse in cases of

accidental abuse discovery by diagnosis of STD. Journal of Interpersonal Violence, 7, 532–542. London, K., Bruck, M., Ceci, S. J., & Shuman, D. W. (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. Lynskey, M. T., & Fergusson, D. M. (1997). Factors protecting against the development of adjustment difficulties in young adults exposed to childhood sexual

abuse. Child Abuse & Neglect, 21, 1177–1190. Mossige, S. (2001). Ungdoms holdninger til seksuelle krenkelser og overgrep [Young persons’ attitudes towards sexual violations and abuse] (NOVA Rapport No.

16/01). Oslo: NOVA – Norwegian Social Research. Nagel, D. E., Putnam, F. W., Noll, J. G., & Tricket, P. K. (1997). Disclosure patterns of sexual abuse and psychological functioning at a 1-year follow-up. Child

Abuse & Neglect, 21, 137–147. Paine, M. L., & Hansen, D. J. (2002). Factors influencing children to self-disclose sexual abuse. Clinical Psychology Review, 22, 271–295. Parker, G. (1990). The Parental Bonding Instrument. A decade of research. Social Psychiatry and Psychiatric Epidemiology, 25, 281–282. Parker, G., Tupling, H., & Brown, L. B. (1979). A parental bonding instrument. British Journal of Medical Psychology, 52, 1–10. Roesler, T. A. (1994). Reactions to disclosure of childhood sexual abuse. The effect on adult symptoms. Journal of Nervous & Mental Disease, 182, 618–624.

1108 G. Priebe, C.G. Svedin / Child Abuse & Neglect 32 (2008) 1095–1108

Ruggiero, K. J., Smith, D. W., Hanson, R. F., Resnick, H. S., Saunders, B. E., Kilpatrick, D. G., & Best, C. L. (2004). Is disclosure of childhood rape associated with mental health outcome? Results from the National Women’s Study. Child Maltreatment, 9, 62–77.

Sinclair, B. B., & Gold, S. R. (1997). The psychological impact of withholding disclosure of child sexual abuse. Violence and Victims, 12, 137–145. Sjöberg, R. L., & Lindblad, F. (2002). Limited disclosure of sexual abuse in children whose experiences were documented by videotape. American Journal of

Psychiatry, 159, 312–314. 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, 24, 273–287. Somer, E., & Szwarcberg, S. (2001). Variables in delayed disclosure of childhood sexual abuse. American Journal of Orthopsychiatry, 71, 332–341. Sorensen, T., & Snow, B. (1991). How children tell: The process of disclosure in child sexual abuse. Child Welfare, 70, 3–15. Spaccarelli, S., & Kim, S. (1995). Resilience criteria and factors associated with resilience in sexually abused girls. Child Abuse & Neglect, 19, 1171–1182. Sundell, K., El-Khouri, B., & Månsson, J. (2005). Elever på vift: Vilka är skolkarna? [Students on the loose: Who are the truants?]. Stockholm: Forsknings- och

utvecklingsenheten Socialtjänstförvaltningen Stockholms stad. Svedin, C. G., & Back, C. (2003). Varför berättar de inte? Om att utnyttjas i barnpornografi [Why don’t they tell? About being exploated in child pornography] (1st

ed.). Stockholm: Rädda Barnen. Sweden’s Statistical Database. (2003). Retrieved January 30, 2007, from http://www.ssd.scb.se/databaser/makro/Produkt.asp?produktid=UF0507 Tambs, K. (1994). Undersøkelse av seksuelle overgrep mot barn [Study of child sexual abuse]. Oslo: Statens Institutt for Folkehelse. Tang, C. S. (2002). Childhood experience of sexual abuse among Hong Kong Chinese college students. Child Abuse & Neglect, 26, 23–37. Teram, E., Stalker, C., Hovey, A., Schachter, C., & Lasiuk, G. (2006). Towards malecentric communication: Sensitizing health professionals to the realities of

male childhood sexual abuse survivors. Issues in Mental Health Nursing, 27, 499–517. Watkins, B., & Bentovim, A. (1992). The sexual abuse of male children and adolescents: A review of current research. Journal of Child Psychology and Psychiatry,

33, 197–248.