Skip to main content
NCBI home page
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2018 Aug 14.
Published in final edited form as: Child Abuse Negl. 2018 Feb 20;79:164–172. doi: 10.1016/j.chiabu.2018.01.025

Correlates of disclosure of sexual violence among Kenyan youth

Courtney L Boudreau a,b,*, Howard Kress c, Roger W Rochat a,b, Kathryn M Yount b,d
PMCID: PMC6091645  NIHMSID: NIHMS978561  PMID: 29459242

Abstract

Introduction

Sexual violence (SV) against children is a global health and human rights issue that can have short and long-term consequences for health and wellbeing. Disclosing SV increases the likelihood that children can access health and protective services and receive psychosocial support. Research in high-income countries has found that child SV survivors are more likely to disclose when they are girls/women, experience fewer SV events, and experience SV perpetrated by a stranger. No studies have examined correlates of SV disclosure in Kenya.

Objective

The objective of this research was to assess the correlates of disclosing SV among Kenyan youth ages 13–24 who reported an SV experience before age 18.

Methods

In 2010, the Kenya Ministry of Gender, Children and Social Development, the U.S. Centers for Disease Control and Prevention’s (CDC) Division of Violence Prevention, the UNICEF Kenya Country Office, and the Kenya National Bureau of Statistics (KNBS) conducted a national survey of violence against children. These data were used to conduct weighted logistic regression analyses to determine which factors were correlated with reporting SV disclosure.

Results

Among the 27.8% of girls/women and 14.5% of boys/men who reported SV before age 18, 44.6% of girls/women and 28.2% of boys/men reported to have disclosed the experience. In weighted logistic regression analysis, the odds of disclosure were lower among survivors who were boys/men and among survivors who reported more SV events, and higher when any perpetrator was a family member.

Conclusion

More context-specific research on SV disclosure among young people is needed globally.

Keywords: Kenya, Sexual violence, Violence against children, Disclosure, Child sexual abuse, Intimate partner violence, Child health

1. Introduction

Sexual violence (SV) against children is a global health and human rights issue that can have detrimental short and long-term effects on the survivor’s health and wellbeing (Collin-Vezina, Daigneault, & Hebert, 2013). Disclosure of SV, in some cases, has a positive effect on health, by linking survivors to care and child protection systems, or providing psychosocial support, and decreasing psychological symptoms (Paine & Hansen, 2002; Ruggiero et al., 2004; Ullman & Filipas, 2001). Immediate linkage to services is critical when the survivor is at risk of HIV transmission, pregnancy, or severe physical injury. However, disclosure can increase psychopathology symptoms, particularly when the response to disclosure is negative (Edwards, Dardis, Sylaska, & Gidycz, 2014; Roesler, 1994). Reactions to disclosure depend on factors such as community perceptions of SV and the relationship of the perpetrator to the survivor (Jewkes, Penn-Kekana, & Rose-Junius, 2005; Ullman & Filipas, 2001). Whether or not the child discloses also depends on other factors, including the survivor’s perception of the event and the survivor’s demographic characteristics (Goodman-Brown, Edelstein, Goodman, Jones, & Gordon, 2003). In this analysis, disclosure is defined as telling anyone about an SV experience and may include both formal and informal disclosure. The majority of research on SV tends to examine either intimate partner violence (IPV) or child sexual abuse (CSA) but does not often examine SV during childhood as a whole. All instances of CSA are considered SV during childhood, but IPV is only considered to be such when it is sexual in nature and occurs prior to the age of 18.

Few studies have examined SV against children in Africa, and even fewer have examined this problem in Kenya. This study explored the characteristics that are correlated with retrospective reporting of SV disclosure among Kenyan youth who reported SV. This analysis offers insights about which children who report SV in a survey are more likely to have disclosed SV, enabling public health practitioners to encourage disclosure among key sub-groups and to design interventions that enhance access to adequate medical and legal services for SV survivors.

1.1. Factors influencing SV disclosure

Community norms with respect to SV influence the likelihood that the child SV survivor will disclose, the recipient’s reaction to the disclosure, and the health and legal infrastructure available for response. Only one study has examined Kenyan beliefs about SV, looking specifically at perceptions of CSA (Plummer & Njuguna, 2009). This study employed focus groups with child protection workers from different tribes to understand the risk and protective factors for CSA in Kenya (Plummer & Njuguna, 2009). Protective factors identified were: traditional values, placing a high value on children, and taboos. Risk factors identified were tribal factors, a culture of silence around sexual matters, gender roles, patriarchy, foreign influence, and individual-level factors such as orphan status or presence of a step-father (Plummer & Njuguna, 2009). These perceived risk and protective factors might influence the likelihood that a survivor will disclose. For example, gender norms, such as those that place a high value on virginity may influence a female survivor to not disclose SV, to protect her status as a virgin.

Community norms about gender and violence affect the outcome of SV disclosure in different ways across gender. In East and Central Africa, women reported that disclosure could reduce women’s marriage prospects and increase the likelihood of mistreatment and stigmatization by family and community members (Byrskog, Olsson, Essen, & Allvin, 2014; Kelly et al., 2012). Boys may not disclose SV because of norms around masculinity (Easton, 2014), which expect them to resist sexual advances or to deal with the abuse on their own (Donnelly & Kenyon, 1996). Boys may believe that it is impossible for a girl or woman to be a perpetrator or that sexual abuse cannot happen to boys (Donnelly & Kenyon, 1996). In fact, a common reason for non-disclosure by boys is not perceiving the event as a crime (Lehrer, Lehrer, & Koss, 2013). Boys may fear being viewed as homosexual (Lehrer et al., 2013), especially in parts of Sub-Saharan Africa, where homosexuality is stigmatized and, at times, criminalized (Taegtmeyer et al., 2013).

1.2. Correlates of disclosing SV

Several studies in middle to high income countries in Europe, North America, and Asia have examined the correlates of SV disclosure. Although the correlates of disclosure may vary across contexts (population vs. clinical, different countries), outcomes (CSA, childhood rape), data collection methods (telephone survey, chart abstraction, qualitative interviews), and analytic techniques (different modeling strategies), many studies have identified several common correlates of disclosure.

Girls more so than boys (Hanson et al., 2003; Lam, 2014; Lippert, Cross, Jones, & Walsh, 2009; Ullman & Filipas, 2005), and children living with both parents more than their counterparts (Kogan, 2004; Schonbucher, Maier, Mohler-Kuo, Schnyder, & Landolt, 2012) are more likely to disclose SV. Having both parents in the home may reflect higher social support; whereas, orphans may not have a trusted adult in their lives to whom they can disclose. In some studies, children who are older at the time of victimization are more likely than younger children to disclose SV (Adefolalu, 2014; Lippert et al., 2009; Smith et al., 2000); however, other studies have found the opposite (Goodman-Brown et al., 2003). Older survivors may be more aware of the inappropriateness of the incident; whereas, younger children may not understand the abuse (Smith et al., 2000). However, younger children may more often tell someone about the abuse if they do not realize the stigma that accompanies SV (Smith et al., 2000). The survivor’s age at the time of the survey may also influence the likelihood of disclosure. Studies have found that respondents who are older at the time of survey may be more likely to disclose (Lam, 2014; Lippert et al., 2009). For older respondents, there is likely to be a longer time period between the abuse and the survey. This distance from the event(s) may increase the respondent’s willingness to disclose.

Survivors of intrafamilial perpetration less often disclose their experience than do survivors of perpetration by a non-family member (Goodman-Brown et al., 2003; Hanson et al., 2003; Kogan, 2004; Lange et al., 1999; Schonbucher et al., 2012; Smith et al., 2000). As intrafamilial abuse is associated with negative reactions from the community, children may anticipate such reactions and avoid disclosure. Lower rates of disclosure among intrafamilial SV survivors, however, may be confounded by other associated factors, such as lower survivor age and increased emotional violence (EV) (Lange et al., 1999; Taveira, Frazao, Dias, Matos, & Magalhaes, 2009). Children who experience SV perpetrated by a stranger are more likely to disclose SV in higher-income countries (Priebe & Svedin, 2008; Smith et al., 2000). This finding may be due to community perceptions that strangers usually perpetrate SV, implying that it may not translate to settings with different perceptions of SV (Davidsson, Benjaminsson, Wijma, & Swahnberg, 2009). Additionally, children may more clearly understand that the act is inappropriate when the perpetrator is unfamiliar to them (Smith et al., 2000).

Research on SV in high-income settings often discusses the ‘severity’ of the violence, but clear definitions for severity are not always provided. That said, more ‘serious’ or ‘severe’ cases of SV may more often be disclosed (Finkelhor, 1994; Lam, 2014), especially by girls/women (Priebe & Svedin, 2008). Child survivors may be more likely to perceive more serious cases as inappropriate, or they may be unable to hide being physically harmed by the abuse. Experiencing penetrative SV, as compared to non-penetrative SV, is not associated with disclosure (Hanson et al., 2003; Kogan, 2004; Schonbucher et al., 2012). Findings on the impact of being threatened during SV on disclosure of the event are conflicting (Hanson et al., 2003; Smith et al., 2000).

Frequent abuse may also be deemed more ‘severe,’ but researchers rarely differentiate between repeated incidents by the same perpetrator or re-victimization by another perpetrator. Experiencing a series of assaults has been associated with not disclosing SV (Kogan, 2004; Schonbucher et al., 2012; Smith et al., 2000). However, other research has found that disclosure behavior of those experiencing a single assault does not differ from that of those experiencing a series of assaults (Lam, 2014; Lippert et al., 2009; Priebe & Svedin, 2008). An association between repeated abuse and lower rates of disclosure may exist because children who do not disclose may be more likely to face re-victimization compared to children who do disclose. The correlation between chronic violence and disclosure may occur because children who experience chronic violence may fear repercussions from the perpetrator.

One study showed that experiencing physical abuse was not associated with disclosure behaviors (Priebe & Svedin, 2008). This finding corroborates research that being injured during the assault is not associated with disclosure (Hanson et al., 2003; Smith et al., 2000). No prior studies have examined EV prior to SV as a correlate of disclosure. However, researchers have found a correlation between intrafamilial SV and increased EV and an association between intrafamilial SV and a decreased likelihood of disclosure (Hershkowitz, Lanes, & Lamb, 2007; Taveira et al., 2009). Therefore, EV may be associated with a lower likelihood of disclosure.

1.3. Outcomes of disclosing SV

A survivor’s disclosure of SV may have positive and negative outcomes, depending on the context and circumstances of the disclosure. The majority of research on SV disclosure has been conducted in high-income countries. Disclosure of SV may assist the survivor’s healing process by facilitating access to appropriate medical and psychological services (Easton, 2014). Theoretical models of trauma recovery cite the importance of disclosure to reduce the negative psychological and physical consequences of SV in childhood (Easton, 2014). Disclosure of SV soon after the event has been associated with lower levels of psychological symptoms later in life among women in the United States (Ruggiero et al., 2004; Ullman & Filipas, 2001).

However, disclosure does not guarantee access to services or appropriate support when services are accessed (Paine & Hansen, 2002). In some studies, disclosure has adversely affected the survivor’s psychological health, with higher levels of depression reported among those who disclosed (Lee & Kim, 2011; Sciolla et al., 2011). Additionally, the disclosure recipient may not be empowered to respond: research has shown that adolescents may be more likely to disclose to peers (Priebe & Svedin, 2008). Survivors often do not disclose because they are afraid of further violence to themselves or their families (Adefolalu, 2014; Lehrer et al., 2013).

The effect of disclosure on mental health depends on reactions to disclosure. In the United States, survivors who disclose during childhood and receive a negative reaction (such as disbelief or blame) have higher levels of post-traumatic stress than do those who receive a positive reaction (Edwards et al., 2014; Roesler, 1994). Reactions to disclosure are influenced by SV characteristics as well as community perceptions of SV.

1.4. Objectives of this analysis

The objective of this study is to examine which factors are correlated with disclosure of SV among young people in Kenya. Most research on SV in childhood has occurred in high-income countries. No studies examine the correlates of disclosing SV in Kenya. Examining the correlates of disclosure can inform policy efforts to encourage disclosure of SV. Encouraging safe disclosure could enable survivors to receive appropriate medical treatment (counseling, HIV post-exposure prophylaxis (PEP), sexually transmitted infection (STI) prophylaxis, emergency contraception (EC), etc.), could connect them with the legal system to protect them from the perpetrator and could provide avenues to hold perpetrators accountable. Understanding disclosure could help the courts to recognize that SV survivors may not always disclose, but that this does not imply that their perpetrator is innocent (Paine & Hansen, 2002). Finally, convicting the perpetrator may prevent him or her from abusing other children (Paine & Hansen, 2002).

2. Methods

Data for this analysis come from the 2010 Violence Against Children Study in Kenya. The Kenya Ministry of Gender, Children and Social Development, with the Kenya National Bureau of Statistics, Centers for Disease Control and Prevention (CDC) Atlanta and UNICEF, undertook this cross-sectional, nationally representative survey. Estimates in this analysis were produced from the combined data of 13–24 year old participants and are therefore different than the estimates published in the official VACS report, which were produced separately for participants 13–17 and 18–24.

2.1. Study site

Kenya is an East-African country with a population of 45.5 million people, with approximately 25% of the population living in urban areas (“Human Development Report, 2015: Work for Human Development,” 2015). It is a lower middle income country, with an estimated 48% of the population living in poverty (Human Development Report, 2015). Gender inequality persists, with boys completing an average of 1.4 additional grades of schooling as compared with girls (Human Development Report, 2015). The average life expectancy is 61.6 years, which is higher than Sub-Saharan Africa’s life expectancy of 58.5 years. Kenya’s under-5 mortality rate is 70.7 per 1000 live births, lower than the Sub-Saharan African rate of 86 per 1000 live births (Human Development Report, 2015). The HIV prevalence among adults is 6% (Human Development Report, 2015).

2.2. Sample

The survey employed a three-stage, stratified cluster household design to obtain a nationally representative sample. Household response rates were 90.3% for girls/women and 89.6% for boys/men. Individual response rates were 94.0% for girls/women and 89.8% for boys/men. Completed surveys were available for 1227 girls/women and 1456 boys/men.

In this analysis, SV is defined as having experienced at least of one of the following SV types: sexual touching, attempted sex, physically forced sex, or pressured sex. Each type of SV was defined by the interviewer before asking the participant to respond. For each type of SV, the respondent was asked how old they were the first time the event occurred.

The responses of participants who refused to answer or replied “Don’t know” when asked about experiences of the four types of SV, were coded as missing (refused: 0.7% of girls/women and 0.5% of boys/men; “Don’t know”: 0.5% of girls/women and 1.3% of boys/men). One respondent had missing data for age at first SV experience. Among those coded as having experienced SV before age 18 (n = 517), 94.6% (n = 489) had complete data for the variables of interest in this analysis. Of those with incomplete data, 24 were missing data on disclosure only, two were missing data for physically forced sex only, and another two respondents were missing data for both variables. Among respondents who had missing data for disclosure (n = 26), 22 were truly missing, three responded, “don’t know,” and one refused to answer. Only those with complete data were included in the analysis (n = 489, 66.5% girls/women).

2.3. Data collection

Data were collected in November and December 2010. This survey followed the World Health Organization’s guidelines on ethics and safety in studies on violence against women. The CDC’s Institutional Review Board and the Ethical Review Committee of the Kenya Medical Research Institute approved the study.

A technical working group of local and national experts developed the survey and pilot-tested it in villages outside the sampling frame. The survey was translated from English into 12 languages, back-translated into English, and validated across back-translations. Trained interviewers administered the survey to boys and girls between the ages of 13 and 24. For participants under 18, consent to participate in a survey on “health, education and life experiences” was obtained from the legal guardian. A two-stage assent process was followed to ensure that all respondents were willing to participate in the survey. The survey was administered in a private location, and respondents who requested services were connected with a counselor, and all participants received a list of service organizations and agencies.

Data were entered into Epi Info version 3.5.1; 12.7% of surveys were entered twice to ensure data quality.

2.4. Outcome measure: disclosure of SV

The outcome for this analysis was a dichotomous variable capturing whether participants who reported experiencing SV before the age of 18 told anyone about the incident(s) (“Did you ever tell anybody about any of these incidents- unwanted touching, attempted sex, physically forced sex, or pressured sex?”). Therefore, disclosure in this analysis refers not to reporting SV to the interviewer, but disclosure of reported SV to anyone prior to survey administration.

2.5. Correlates of SV disclosure

2.5.1. Survivor demographics

Gender, age in years at the time of the survey, and family status (whether or not the child was a single or double orphan before experiencing SV) were considered as correlates of SV.

2.5.2. SV characteristics

The following variables were considered as correlates of disclosure: Physically forced sex, age at first experience of SV in years, and total number of experiences of SV. Intoxication during SV was considered but was not included because so few respondents (< 5%) reported being intoxicated during SV. Prior emotional or physical violence (EV, PV) captured whether or not the participant reported experiencing any type of emotional or physical violence before their first experience of SV. This variable was calculated based on the reported ages at first experiences of SV, PV, and EV. Experiences of SV that occured in the same year as PV or EV were not included in this variable because SV itself may lead to PV or EV.

2.5.3. Perpetrator type

Captured whether any perpetrator of SV was a romantic partner, family member, or another known person. Respondents could report multiple occurrences of SV, meaning that one respondent could have experienced SV perpetrated by multiple perpetrator types. The variable for “other known perpetrator” denotes a perpetrator who was known to the survivor but was not a romantic partner or family member.

2.6. Analysis

An analysis of the prevalence of SV among the entire sample was conducted prior to assessing the correlates of SV disclosure. Only respondents who reported experiencing SV were included in further analyses. Univariate analyses of each selected variable and bivariate analyses (χ2) of each covariate with the outcome were conducted. All variables then were entered into a logistic regression model. Variables were assessed for collinearity based on variance decomposition proportions and conditions (SAS Macro, Department of Epidemiology, Rollins School of Public Health at Emory University). Mean-centered variables were created for age in years at the survey and age at first experience of SV to address collinearity between each of these variables with the intercept. Backwards elimination was conducted manually to eliminate non-significant variables (p < 0.05) in analyses of the total sample, boys/men only, and girls/women only. Model selection decisions were data-driven because no literature existed on the correlates of disclosure in the Kenyan context. All analyses accounted for the sample design by using weight, cluster, and strata statements in SAS 9.4 (Cary, N.C.) survey procedures.

3. Results

3.1. Prevalence of reported SV

Among the total sample, 27.8% of girls/women and 14.5% of boys/men reported experiencing SV before age 18 (Table 1). The reported prevalence of SV was higher among girls/women than boys/men for all types of SV. For boys/men and girls/women, sexual touching was the most common type of SV reported.

Table 1.

Prevalence of reported sexual violence before age 18, respondents 13–24 years by gender, Kenya VACS, 2010.a

Girls/Women Boys/Men Total
N = 1227 N = 1456 N = 2683
% (95% CI) % (95% CI) % (95% CI)
Sexual touching 18.9 (15.4, 22.5) 8.9 (6.6, 11.3)b 14.0 (11.6, 16.4)
Attempted sex 12.0 (9.5, 14.7) 6.6 (5.0, 8.3)b 9.4 (7.7, 11.1)
Physically forced sex 5.9 (3.7, 8.0) 1.1 (0.3, 1.8)b 3.5 (2.3, 4.7)
Pressured sex 7.6 (5.5, 9.6) 3.2 (2.0, 4.5)b 5.5 (4.1, 6.8)
Any sexual violence 27.8 (23.1, 32.4) 14.5 (11.7, 17.3)b 21.2 (18.2, 24.3)
Open in a new tab
a

Percentages are weighted to account for sample design.

b

T-tests for differences across gender are significant at the 0.05 level.

3.2. Characteristics of reported SV

Among those who reported experiencing SV before age 18, the average age at first reported experience of SV was 14.4 years for girls/women and 13.4 years for boys/men; with boys/men reporting being younger at age of first SV (Table 2). Almost one quarter (23.9%) of girl/women SV survivors and 18.9% of boys/men survivors reported becoming a single or double orphan before the reported experience of SV.

Table 2.

Characteristics of reported sexual violence before age 18, respondents 13–24 years by gender, Kenya VACS, 2010.a

Girls/Women
n = 304		 Boys/Men
n = 185		 Total
n = 489
Demographic characteristics
Age at survey in years, Mean (SD) 18.6 (0.2) 18.2 (0.2) 18.4 (0.2)
Single or double orphan before SV, % (95% CI) 23.9 (13.3, 34.5) 18.9 (11.3, 26.4) 22.2 (14.7, 29.8)
Characteristics of reported SV exposure
Age at first experience of SV, Mean (SD) 14.4 (0.2) 13.4 (0.3)b 14.08 (0.2)
Total number of experiences of SV, Mean (SD) 10.9 (2.1) 5.9 (1.0) 9.22 (1.4)
Any physically forced sex % (95% CI) 22.4 (15.5, 29.3) 8.0 (3.2, 12.8)b 17.5 (12.6, 22.6)
EV prior to SV % (95% CI) 12.8 (6.9, 18.7) 14.9 (8.5, 21.3) 13.5 (9.2, 17.8)
PV prior to SV % (95% CI) 59.4 (47.8, 71.0) 60.2 (49.9, 70.6) 59.7 (51.4, 68.0)
Any perpetrators were romantic partners % (95% CI) 53.3 (44.2, 62.4) 48.4 (38.1, 58.7) 51.6 (44.7, 58.6)
Any perpetrators were family members % (95% CI) 20.5 (11.0, 30.0) 15.4 (9.4, 21.4) 18.8 (12.2, 25.5)
Any perpetrators were knownc % (95% CI) 44.7 (38.0, 51.4) 40.0 (29.2, 50.7) 43.1 (37.3, 48.9)
Disclosed SV to anyone % (95% CI) 44.6 (37.8, 51.3) 28.2 (20.4, 35.7)b 39.1 (33.6, 44.6)
Open in a new tab
a

Percentages are weighted to account for sample design.

b

T-tests and Rao-Scott chi-square tests for differences across gender are significant at 0.05 level.

c

Denotes existence of any perpetrators who were known to the survivor but who were not family members or romantic partners.

Among those who reported SV, girls/women reported experiencing an average of 10.9 experiences of SV. Boys/men reported an average of 5.9 experiences of SV; the difference between genders was not significant. Girls/women reported a higher prevalence of physically forced sex as compared with boys/men (22.4% of girls/women and 8.0% of boys/men). The majority of reported SV survivors reported having experienced PV before experiencing SV (59.4% of girls/women and 60.2% of boys/men).

About half of girls/women (53.3%) and boys/men (48.4%) reported that at least one of the perpetrators of reported SV was a romantic partner. Fewer SV survivors reported that any perpetrator was a family member (20.5% of girls/women, 15.4% of boys/men). Nearly half of survivors reported that at least one perpetrator was known to them, but was not a family member or a romantic partner (44.7% of girls/women and 40.0% of boys/men). Very few SV survivors reported that any perpetrator was unknown to them (5.9% of girls/women and 5.6% of boys/men). The differences between genders for the three perpetrator types were not significant. Disclosure of SV before the survey differed for girls/women and boys/men, with 44.6% of girls/women and 28.2% of boys/men reporting to have ever disclosed any reported experience of SV to anyone.

3.3. Unadjusted odds ratios of reported disclosure of SV

Unadjusted odds ratios of disclosure for girls/women, boys/men, and the total sample are presented in Table 3. Due to small cell sizes (< 25 observations), odds ratios for boys/men are not presented for four variables. Among girls/women who reported experiencing SV, the total number of experiences of SV was negatively associated with disclosure. This association was also present in the total sample. For every additional reported experience of SV, the odds of disclosure were 2% lower. Also among girls/women, having a perpetrator who was a family member increased the odds of disclosure by a factor of 2.19. This association was no longer present when data for all respondents were combined. Gender was correlated with disclosure: the odds of disclosure among boys/men were 0.48 times the odds of disclosure among girls/women. No unadjusted odds ratios were significant for boys/men.

Table 3.

Unadjusted odds ratios and 95% confidence intervals of reported disclosure of SV, respondents 13–24 years who reported SV exposure before age 18 years, overall and by gender, Kenya VACS, 2010.

Girls/Women
n = 304		 Boys/Men
n = 185		 Total
n = 489
Demographic characteristics
Age at time of survey 1.00 (0.91, 1.09) 1.10 (0.96, 1.26) 1.03 (0.96, 1.11)
Single or double orphan prior to SV 2.42 (0.81, 7.26) c 2.02 (0.83, 4.94)
Male gender N/A N/A 0.48 (0.30, 0.76)a
Characteristics of reported SV exposure
Age at first experience of SV 1.02 (0.88, 1.17) 1.10 (0.96, 1.26) 1.07 (0.98, 1.18)
Total number of experiences of SV 0.98 (0.96, 0.99)a 0.99 (0.96, 1.02) 0.98 (0.97, 0.99)a
Any physically forced sex 1.24 (0.58, 2.64) c 1.38 (0.71, 2.69)
EV prior to SV 1.63 (0.51, 5.22) c 1.47 (0.63, 3.47)
PV prior to SV 0.70 (0.35, 1.39) 0.81 (0.34, 1.90) 0.73 (0.42, 1.28)
Any perpetrators romantic partners 0.75 (0.38, 1.48) 0.94 (0.43, 2.08) 0.83 (0.50, 1.38)
Any perpetrators family members 2.19 (1.06, 4.52)a c 1.76 (0.99, 3.16)
Any perpetrators were knownb 1.30 (0.77, 2.20) 1.36 (0.67, 2.76) 1.35 (0.89, 2.05)
Open in a new tab
a

OR is significant at the 0.05 level.

b

Denotes existence of any perpetrators who were known to the survivor but who were not family members or romantic partners.

c

Odds ratios not presented dues to unstable estimates (< 25 observations/cell).

3.4. Adjusted odds ratios of reported disclosure of SV

Backwards elimination model selection was conducted for girls/women, boys/men, and the total sample. Among boys/men, no variables were associated with disclosure. In the total sample, the final selected model found gender, total number of experiences of SV, and family member perpetrator to be significant correlates of disclosing SV (Table 4). Boys/men were less likely to disclose: the adjusted odds of disclosure for boys/men were 0.45 (0.28, 0.72) times the odds of disclosure for girls/women. The more events of SV the respondent reported experiencing, the less likely s/he was to have told someone about any experience. The odds of disclosure decreased by a factor of 0.98 (0.96, 0.99) for every additional reported experience of SV. If any perpetrator was a family member, the odds of disclosure were 1.92 (1.11, 3.33) higher than the odds of disclosure among survivors who had no perpetrators that were family members. Among girls/women, the total number of experiences of SV and having a family member perpetrator were more strongly associated with disclosure (Table 4).

Table 4.

Adjusted odds ratios of reported disclosure of SV based on the final model (gender, total number of reported experiences of SV, family member perpetrator) among respondents ages 13–24 who reported experiencing SV before age 18, Kenya VACS, 2010.

Adjusted Odds Ratio of Reported Disclosure of SV (95% CI) Adjusted Odds Ratio of Reported Disclosure of SV (95% CI)
Full model n = 489 Female model n = 304
Demographic Characteristics
Male gender 0.45 (0.28, 0.72)a N/A
Characteristics of reported SV exposure
Total number of experiences of SV 0.98 (0.96, 0.99)a 0.97 (0.96, 0.99)a
Any perpetrators were family members 1.92 (1.11, 3.33)a 2.70 (1.31, 5.54)a
Open in a new tab
a

OR is significant at the 0.05 level.

4. Discussion

This study is the first to assess the correlates of disclosure of reported SV before age 18 among young people in Kenya. The high prevalence of childhood SV reported in this nationally representative sample underscores the importance of addressing this issue. Gaining a deeper understanding of the factors associated with reported disclosure of SV could inform policy-makers and public health practitioners to take targeted action to prevent SV and to aid child and adolescent survivors of SV. The World Health Organization (WHO) recommends strengthening response and support services as one of seven key strategies to end violence against children (44). The WHO suggests this strategy will lead to reductions in the incidences of further violence, trauma symptoms and negative reproductive health outcomes.

Among those who reported SV prior to age 18, over half of all girls/women and over two-thirds of boys/men reported that they had never told anyone about their experiences. By disclosing their experience to someone, these children may have been able to access health services, receive counseling, or be removed from the abusive situation. Observed disclosure rates are lower than those observed in prior research, though this study is the first to examine disclosure rates in a lower-income country (Kogan, 2004; Lam, 2014; Lippert et al., 2009). Possible reasons for the lower rates of reported disclosure of SV in Kenya are multiple. Cultural beliefs about SV and gender norms may prevent children from telling anyone about their experiences or understanding them as violent. Stigma towards SV survivors may be higher in Kenya than in the other locations where surveys were conducted. Children may be aware of the attitudes towards SV and anticipate a negative reaction to disclosure. There may be less awareness of SV and what constitutes SV as compared with high-income countries. Finally, children may not disclose SV because of the lack of accessible services for SV survivors. Where services are available, children may not be aware of them, may not have the means to access them, or may not be aware that they should seek help. These explanations may not only prevent children from disclosing SV to someone in their life, but also may prevent the child from reporting exposure to SV during the interview. Survivors who disclosed their experience of SV in the past may be more likely to report SV during the survey. Therefore, the reported prevalence of disclosure (before the survey) may be over-estimated. This possibility could bias results if those SV survivors who do not disclose to the enumerator and did not disclose to anyone in the past have SV characteristics that are different from those who disclosed to the interviewer but did not in the past. Enumerators for the Kenya VACS were trained in survey administration and precautions were taken to ensure that children were surveyed in a private area.

Consistent with research from high-income countries, boys/men who reported exposure to SV were less likely than girls/women to report disclosing (Hanson et al., 2003; Lam, 2014; Lippert et al., 2009). Norms about masculinity may influence boys’ decisions to disclose (Easton, 2014). Boys may feel emasculated by the experience, believe that they should have been able to resist the perpetrator, or feel that they should deal with the experience alone (Donnelly & Kenyon, 1996). They may believe that SV cannot happen to boys, that women cannot perpetrate SV, or that SV is not a crime (Donnelly & Kenyon, 1996; Lehrer et al., 2013). Boys may also fear being perceived as homosexual, as it is stigmatized in certain Sub-Saharan African countries (Lehrer et al., 2013; Taegtmeyer et al., 2013).

Having more experiences of SV was associated with lower odds of disclosure. This correlation was observed among all SV survivors and female survivors, but was not significant among boys/men. The lack of association among boys/men may be a result of the small number of boys/men who reported disclosing SV. Research in high-income countries has also found a negative association between the frequency of SV and disclosure (Kogan, 2004; Schonbucher et al., 2012; Smith et al., 2000). More frequent SV may normalize the abuse, making children feel that it is an inescapable occurrence in their lives. More SV events may be associated with prolonged manipulation to convince the child not to disclose. The child may become more ashamed to disclose SV the more often it has happened. This association may also exist because children who do not disclose may be at higher risk of re-victimization by the same perpetrator or a different perpetrator.

Paradoxically, participants who reported experiencing intrafamilial SV were more likely to have disclosed SV. This result contradicts findings from high-income countries, where survivors of intrafamilial SV were less likely to disclose (Goodman-Brown et al., 2003; Hanson et al., 2003; Kogan, 2004; Lange et al., 1999; Schonbucher et al., 2012; Smith et al., 2000). This relationship was found to be significant among all SV survivors and among female survivors. It was non-significant among boys/men, which may be a result of small sample size or the effect may only be present among girls/women. Survivors of intrafamilial SV may be more likely to disclose because of cultural family values. Perhaps in the Kenyan context, SV among family members is more clearly a violation than SV by another perpetrator. Different family and housing structures may lead to a higher likelihood that someone may discover the SV, essentially forcing the child to disclose. Children who experience intrafamilial SV may be more concerned that further violence will occur, as compared with a perpetrator that they may not see as often. This may prompt the child to disclose to prevent another event from occurring. Participants who have previously disclosed intrafamilial SV may also be more likely to report this to the enumerator. Participants who have never disclosed intrafamilial SV may be less likely to disclose to the enumerator because they are ashamed or may still be living with the perpetrator and fear the ramifications of disclosure to the enumerator. If this were the case, the disclosure rate among survivors of intrafamilial SV may be biased in the opposite direction.

One limitation to this research is the ambiguity of the outcome variable. Although the respondent reports disclosure, we cannot know to whom s/he disclosed or when s/he disclosed. S/he may have told someone without the ability or will to help them or may have disclosed years after the event. We therefore cannot ascertain whether disclosure was beneficial to the child. Delayed disclosure may still be emotionally beneficial for the child, but this depends on the reaction to disclosure (Edwards et al., 2014; Roesler, 1994). Another limitation of the research is that if the child experienced multiple SV events with different perpetrators, it is unclear which event they disclosed. Additionally, as the sample includes all SV prior to 18, the correlates of disclosure may differ depending on whether the incident would be classified as IPV or CSA. Though perpetrator identity was included as a covariate in the analysis, it was not possible to determine which event(s) the survivor disclosed when s/he experienced SV perpetrated by a romantic partner as well as another perpetrator type. Finally, this research is not likely to be generalizable to other contexts.

Despite these limitations, this analysis has several strengths. The VAC survey is nationally representative; implying that the SV characteristics observed in this sample should be similar throughout Kenya. Additionally, the survey had a very high response rate. Another strong point is that the definition of SV was not limited to cases of rape, allowing researchers to gather data on a wide-range of SV experiences. Since it is a population-level survey, it provides a more accurate representation of disclosure rates than would a clinical sample. Participants ranged from 13 to 24 years old, decreasing the amount of recall bias that may be present if older adults were surveyed. Finally, the private nature of survey administration increased the likelihood that participants would report SV.

5. Conclusion

The purpose of examining correlates of disclosure is to understand which SV survivors are likely to disclose. The benefits of disclosure are more likely to be obtained if the recipient of disclosure responds positively, the survivor has access to services, the clinic staff and police have the training and resources to respond, and there are laws in place to convict perpetrators. We must also ensure the immediate safety of SV survivors. The child’s access to services depends on the recipient of disclosure, and whether s/he feels that a response is necessary and whether or not s/he has access to or knowledge of existing services. In addition to the immediate benefits of disclosure, telling someone about the violence may mitigate future mental and physical health outcomes. This, again, is dependent on the disclosure recipient’s response – disclosing to an unsympathetic party may further traumatize the child. The survivor’s willingness to disclose, the recipient’s response, the criminal justice infrastructure, and the healthcare infrastructure are all influenced by community perceptions of SV. Therefore, programs to promote disclosure of childhood SV will need to take a multilevel approach to improve the response to SV. The Kenya VACS analysis found that just 16.6% of 13–17 year old males and 15.8% of 13–17 year old females who had experienced sexual violence in the last year either received or tried to seek professional help. Programs should educate the public, and particularly those that interact with children, about the services available at health centers, which may increase disclosure rates and service seeking. Ensuring that Kenya’s child SV survivors receive adequate services may reduce the future burden on the healthcare system by reducing morbidities associated with SV (Edwards et al., 2014; Roesler, 1994).

In Kenya, the lack of legal framework for SV, combined with society’s perception that rates of SV were increasing, led to the passage of the Sexual Offenses Act in 2006 (Onyango-Ouma, Ndung’u, Baraza, & Birungi, 2006). This bill criminalized the purposeful transmission of HIV, expanded definitions of sex crimes, and provided free health care and counseling to rape victims (Onyango-Ouma et al., 2009). Before the passage of this law, there were no minimum sentences for offenders and the sexual abuse of males was not criminalized (Plummer & Njuguna, 2009). In 2009, the Kenyan Ministry of Health released a second edition of national guidelines on the management of SV that included sections on medical management, psychosocial support, forensic management, humanitarian issues of SV, quality assurance and quality improvement for post-rape care (National Guidelines on Management of Sexual Violence in Kenya, 2009). According to these guidelines, PEP, EC, and STI prophylaxis should be provided when indicated. Healthcare workers are responsible for collecting evidence and filling out a post-rape care report, which is transferred to police (National Guidelines on Management of Sexual Violence in Kenya, 2009). These policies are a step in the right direction, yet more research on SV against children in the Kenyan context should be conducted to understand the complexities of the issue and to design context-specific interventions.

Female gender, fewer experiences of SV, and intrafamilial perpetrator identity are positively associated with disclosure among Kenyan children who reported experiencing SV. Public health practitioners are encouraged to consider these findings when developing interventions to respond to or to prevent SV against children. Awareness campaigns could emphasize that SV is not something that only happens to girls and women. Community organizations that work with young boys can be trained to respond properly to disclosure and to recognize signs of SV. Because about half of SV survivors reported that at least one perpetrator was a romantic partner, interventions to encourage healthy relationships among adolescents may be beneficial. Future research should aim to understand the intersections of different types of SV across life stages, and how these intersections impact disclosure behavior. The overall low disclosure rate highlights the need for increased awareness of SV and available response services. Community norms have been shown to play a role in responses to disclosure and children’s willingness to disclose SV; research should further examine Kenyan attitudes towards SV (Davidsson et al., 2009). This study may inform the Government of Kenya’s efforts to strengthen the health, justice, and education infrastructures to respond to SV against children and to raise awareness about the availability and confidential nature of these services. This analysis highlights the needs of child SV survivors and will help clinicians and practitioners understand which children may be likely to disclose experiences of SV.

Acknowledgments

We are grateful to the research participants for their time and willingness to take part in the study. We are appreciative of the collaborators who worked to conduct the Kenya VAC Survey, including the Kenya National Bureau of Statistics, the Centers for Disease Control and Prevention, and the UNICEF Kenya Country Office for providing the technical support and financial assistance. Finally, we would like to thank to VACS Kenya team who collected the survey data.

This manuscript is based on findings from an MPH thesis produced by the first author in 2015. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

References

  1. Adefolalu AO. Fear of the perpetrator: A major reason why sexual assault victims delayed presenting at hospital. Tropical Medicine and International Health. 2014;19(3):342–347. doi: 10.1111/tmi.12249. [DOI] [PubMed] [Google Scholar]
  2. Byrskog U, Olsson P, Essen B, Allvin MK. Violence and reproductive health preceding flight from war: Accounts from Somali born women in Sweden. BMC Public Health. 2014;14:892. doi: 10.1186/1471-2458-14-892. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Collin-Vezina D, Daigneault I, Hebert M. Lessons learned from child sexual abuse research: Prevalence, outcomes, and preventive strategies. Child and Adolescent Psychiatry and Mental Health. 2013;7(1):22. doi: 10.1186/1753-2000-7-22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Davidsson J, Benjaminsson G, Wijma B, Swahnberg K. Association between experiencing rape, police reporting, and self-reported health among women visiting three gynecology clinics in Sweden. Acta Obstetricia et Gynecologica Scandinavica. 2009;88(9):1000–1005. doi: 10.1080/00016340903147397. [DOI] [PubMed] [Google Scholar]
  5. Donnelly DA, Kenyon S. “Honey, we don’t do men”: Gender stereotypes and the provision of services to sexually assaulted males. Journal Of Interpersonal Violence. 1996;11(3):441–448. doi: 10.1177/088626096011003009. [DOI] [Google Scholar]
  6. Easton SD. Masculine norms, disclosure, and childhood adversities predict long-term mental distress among men with histories of child sexual abuse. Child Abuse and Neglect. 2014;38(2):243–251. doi: 10.1016/j.chiabu.2013.08.020. [DOI] [PubMed] [Google Scholar]
  7. Edwards KM, Dardis CM, Sylaska KM, Gidycz CA. Informal social reactions to college women’s disclosure of intimate partner violence: Associations with psychological and relational variables. Journal of Interpersonal Violence. 2014 doi: 10.1177/0886260514532524. [DOI] [PubMed]
  8. Finkelhor D. Current information on the scope and nature of child sexual abuse. Future of Children. 1994;4(2):31–53. [PubMed] [Google Scholar]
  9. Goodman-Brown TB, Edelstein RS, Goodman GS, Jones DP, Gordon DS. Why children tell: A model of children’s disclosure of sexual abuse. Child Abuse and Neglect. 2003;27(5):525–540. doi: 10.1016/s0145-2134(03)00037-1. [DOI] [PubMed] [Google Scholar]
  10. Hanson RF, Kievit LW, Saunders BE, Smith DW, Kilpatrick DG, Resnick HS, et al. Correlates of adolescent reports of sexual assault: Findings from the National Survey of Adolescents. Child Maltreatment. 2003;8(4):261–272. doi: 10.1177/1077559503257087. [DOI] [PubMed] [Google Scholar]
  11. Hershkowitz I, Lanes O, Lamb ME. Exploring the disclosure of child sexual abuse with alleged victims and their parents. Child Abuse and Neglect. 2007;31(2):111–123. doi: 10.1016/j.chiabu.2006.09.004. [DOI] [PubMed] [Google Scholar]
  12. Human Development Report. Work for human development. United Nations Development Programme; 2015. [Google Scholar]
  13. Jewkes R, Penn-Kekana L, Rose-Junius H. “If they rape me, I can’t blame them”: Reflections on gender in the social context of child rape in South Africa and Namibia. Social Science and Medicine. 2005;61(8):1809–1820. doi: 10.1016/j.socscimed.2005.03.022. [DOI] [PubMed] [Google Scholar]
  14. Kelly J, Kabanga J, Cragin W, Alcayna-Stevens L, Haider S, Vanrooyen MJ. ‘If your husband doesn’t humiliate you, other people won’t’: Gendered attitudes towards sexual violence in eastern Democratic Republic of Congo. Global Public Health. 2012;7(3):285–298. doi: 10.1080/17441692.2011.585344. [DOI] [PubMed] [Google Scholar]
  15. Kogan SM. Disclosing unwanted sexual experiences: Results from a national sample of adolescent women. Child Abuse and Neglect. 2004;28(2):147–165. doi: 10.1016/j.chiabu.2003.09.014. [DOI] [PubMed] [Google Scholar]
  16. Lam KY. Factors associated with adolescents’ disclosure of sexual abuse experiences in Hong Kong. Journal of Child Sexual Abuse. 2014 doi: 10.1080/10538712.2014.950398. [DOI] [PubMed]
  17. Lange A, de Beurs E, Dolan C, Lachnit T, Sjollema S, Hanewald G. Long-term effects of childhood sexual abuse: Objective and subjective characteristics of the abuse and psychopathology in later life. Journal of Nervous And Mental Disease. 1999;187(3):150–158. doi: 10.1097/00005053-199903000-00004. [DOI] [PubMed] [Google Scholar]
  18. Lee Y, Kim S. Childhood maltreatment in South Korea: Retrospective study. Child Abuse and Neglect. 2011;35(12):1037–1044. doi: 10.1016/j.chiabu.2011.09.005. [DOI] [PubMed] [Google Scholar]
  19. Lehrer JA, Lehrer EL, Koss MP. Unwanted sexual experiences in young men: Evidence from a survey of university students in Chile. Archives of Sexual Behavior. 2013;42(2):213–223. doi: 10.1007/s10508-012-0004-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lippert T, Cross TP, Jones L, Walsh W. Telling interviewers about sexual abuse: Predictors of child disclosure at forensic interviews. Child Maltreatment. 2009;14(1):100–113. doi: 10.1177/1077559508318398. [DOI] [PubMed] [Google Scholar]
  21. National Guidelines on Management of Sexual Violence in Kenya. Kenyan Ministry of Health. 2 2009. [Google Scholar]
  22. Onyango-Ouma W, Ndung’u N, Baraza N, Birungi H. The making of the Kenya sexual offenses act, 2006: Behind the scenes. Nairobi: Kwani Trust; 2009. [Google Scholar]
  23. Paine ML, Hansen DJ. Factors influencing children to self-disclose sexual abuse. Clinical Psychology Review. 2002;22(2):271–295. doi: 10.1016/s0272-7358(01)00091-5. [DOI] [PubMed] [Google Scholar]
  24. Plummer CA, Njuguna W. Cultural protective and risk factors: Professional perspectives about child sexual abuse in Kenya. Child Abuse & Neglect. 2009;33(8):524–532. doi: 10.1016/j.chiabu.2009.02.005. [DOI] [PubMed] [Google Scholar]
  25. Priebe G, Svedin CG. Child sexual abuse is largely hidden from the adult society. An epidemiological study of adolescents’ disclosures. Child Abuse and Neglect. 2008;32(12):1095–1108. doi: 10.1016/j.chiabu.2008.04.001. [DOI] [PubMed] [Google Scholar]
  26. Roesler TA. Reactions to disclosure of childhood sexual abuse. The effect on adult symptoms. Journal of Nervous and Mental Disease. 1994;182(11):618–624. doi: 10.1097/00005053-199411000-00004. [DOI] [PubMed] [Google Scholar]
  27. Ruggiero KJ, Smith DW, Hanson RF, Resnick HS, Saunders BE, Kilpatrick DG, et al. Is disclosure of childhood rape associated with mental health outcome? Results from the National Women’s Study. Child Maltreatment. 2004;9(1):62–77. doi: 10.1177/1077559503260309. [DOI] [PubMed] [Google Scholar]
  28. Schonbucher V, Maier T, Mohler-Kuo M, Schnyder U, Landolt MA. Disclosure of child sexual abuse by adolescents: A qualitative in-depth study. Journal Of Interpersonal Violence. 2012;27(17):3486–3513. doi: 10.1177/0886260512445380. [DOI] [PubMed] [Google Scholar]
  29. Sciolla A, Glover DA, Loeb TB, Zhang M, Myers HF, Wyatt GE. Childhood sexual abuse severity and disclosure as predictors of depression among adult African-American and Latina women. Journal of Nervous and Mental Disease. 2011;199(7):471–477. doi: 10.1097/NMD.0b013e31822142ac. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Smith DW, Letourneau EJ, Saunders BE, Kilpatrick DG, Resnick HS, Best CL. Delay in disclosure of childhood rape: Results from a national survey. Child Abuse and Neglect. 2000;24(2):273–287. doi: 10.1016/s0145-2134(99)00130-1. [DOI] [PubMed] [Google Scholar]
  31. Taegtmeyer M, Davies A, Mwangome M, van der Elst EM, Graham SM, Price MA, et al. Challenges in providing counselling to MSM in highly stigmatized contexts: Results of a qualitative study from Kenya. PLoS One. 2013;8(6):e64527. doi: 10.1371/journal.pone.0064527. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Taveira F, Frazao S, Dias R, Matos E, Magalhaes T. Intra and extra-familiar sexual abuse. Acta Médica Portuguesa. 2009;22(6):759–766. [PubMed] [Google Scholar]
  33. Ullman SE, Filipas HH. Predictors of PTSD symptom severity and social reactions in sexual assault victims. Journal Of Traumatic Stress. 2001;14(2):369–389. doi: 10.1023/a:1011125220522. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Ullman SE, Filipas HH. Gender differences in social reactions to abuse disclosures, post-abuse coping, and PTSD of child sexual abuse survivors. Child Abuse and Neglect. 2005;29(7):767–782. doi: 10.1016/j.chiabu.2005.01.005. [DOI] [PubMed] [Google Scholar]

RESOURCES