G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS Child Abuse & Neglect xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Child Abuse & Neglect
Research article
Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls夽 Lauren Rumble a,∗ , Taizivei Mungate b , Handrick Chigiji b , Peter Salama c , Anthony Nolan d , Elayn Sammon e , Leon Muwoni f a United Nations Children’s Fund (UNICEF) Indonesia, Wisma Metropolitan II, 11th Floor, Jl. Jend. SudirmanKav. 31, Jakarta 12920, Indonesia b Zimbabwe National Statistics Agency (ZIMSTAT), 20th Floor, Kaguvi Building, Cnr 4th Central Avenue, Harare, Zimbabwe c United Nations Children’s Fund (UNICEF), Ethiopia, P.O. Box 1169, Africa Hall, Addis Ababa, Ethiopia d United Nations Children’s Fund (UNICEF) Sudan, UNICEF Totto Chan Compound, P.O. Box 45, Juba, South Sudan e UNICEF Mozambique, 1440 Zimbabwe Avenue, Maputo, Mozambique f Child Protection Specialist, UNICEF Zimbabwe, 6 Fairbridge Avenue, Belgravia, Zimbabwe
a r t i c l e
i n f o
Article history: Received 3 November 2014 Received in revised form 16 April 2015 Accepted 27 April 2015 Available online xxx Keywords: Adolescent girls Sexual violence Children
a b s t r a c t Sexual abuse during childhood is a public health and human rights concern throughout the world, including Sub-Saharan Africa. In 2011, Zimbabwe initiated national prevalence data collection on violence against children to inform government policy and programs. We interviewed 567 females and 589 males, aged 18–24 years following standardized and previously tested survey methods from the region. Of females 32.5%, and of males 8.9%, reported experiencing sexual violence before age 18. Most female (62.7%) and male (47.9%) victims of sexual violence experienced more than one incident of sexual violence prior to age 18 years. Three in four females (77.7%) and one in four males (26.7%) of those who experienced sexual violence reported that the first incident was perpetrated by a boyfriend or girlfriend. Few victims received professional help (2.7% of females and 2.4% of males who had reported experiencing sexual violence). Violence against girls is at epidemic levels in Zimbabwe. Most sexual violence against girls occurs within the context of peer relationships. Child victims who seek potentially life-saving support tend not to receive it. This study is evidence of a national public health and child rights emergency in the country and a case for increased, longer-term investment by the government and its development partners in policy reform for enhancing adolescent girls’ empowerment and protection. © 2015 Elsevier Ltd. All rights reserved.
Introduction Reducing violence is fundamental to creating the social conditions for stability and development (Elhawary, Foresti, & Pantuliano, 2011). Despite concern, major gaps in knowledge about the extent of the problem persist (World Health Organization, 2014). Researchers have made a certain amount of progress in some areas. In Zimbabwe, for example, investments in national health surveys in the past decade have resulted in the availability of more robust data on how violence
夽 Funding was provided by UNICEF global thematic funding and the Nduna Foundation. Neither party was involved in the analysis nor played a lead role at any stage of the design, data collection or interpretation. We confirm our free access to the data and no interference from any party during publication of the results of the study or in the writing of the article. ∗ Corresponding author. http://dx.doi.org/10.1016/j.chiabu.2015.04.015 0145-2134/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS
2
L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
affects women. As one instance, the Zimbabwe Demographic Health Survey reported in 2012 that 22% of Zimbabwean women said that their first sexual experience was forced. The same survey noted that 33% of men and 40% of women believed that wife beating is acceptable for some circumstances (ZIMSTAT & ICF International, 2012). These findings are in line with the global multi-country study on sexual violence prevalence led by the World Health Organization (WHO), which estimates that up to 59% of women worldwide experience sexual violence in their lifetimes (García-Moreno, Jansen, Ellsberg, Heise, & Watts, 2005). Much less progress has been made in obtaining data on childhood sexual violence; it tends to be under-reported and undocumented. The first global effort to document the scale of violence against children found evidence that sexual violence was widespread across multiple settings but that systematic and comprehensive data needed to be collected in many countries (Pinheiro, 2006). Since then, few studies from Africa have used consistent prevalence measures or analyzed risk and protective factors (Meinck, Cluver, Boyes, & Mhlongo, 2014). Evidence from low- and middle-income countries is generally lacking (Veenema, Thornton, & Corley, 2014). Where such data exists, it is likely an underestimate because there are substantial barriers to disclosure (Watts & Zimmerman, 2002). Research on childhood sexual violence across the world employs different definitions of sexual violence against children, making it difficult to compare prevalence (Stoltenberg, van Ijzendoorn, Euser, & Bakermans-Kranenburg, 2011). Researchers also face considerable ethical concerns in undertaking such research (Undie, Mullik, & Askew, 2013). The United Nations Children’s Fund (UNICEF) has noted that under-reporting fuels public misconception that violence is a marginal phenomenon (UNICEF, 2014). Although the 2000 United Nations Millennium Declaration refers to freedom from violence as a fundamental value, no Millennium Development Goal was dedicated to addressing sexual or other forms of violence (United Nations General Assembly, 2000). Consequently, public policy discourse has not attended as much to sexual and other forms of violence against children as to other child rights, including the rights to health and education. In 2011, Zimbabwe undertook its first nationally representative survey on physical, emotional, and sexual violence against children. The purpose of the study was to examine the nature and extent of violence affecting children in the country (ZIMSTAT, UNICEF, & CCORE, 2013) to inform policy development. To measure the prevalence of violence against children and associated factors, we applied a methodology successfully implemented in three Sub-Saharan African countries (Together for Girls, 2013). In this article, we report on the study results that pertain especially to sexual violence. Methods Definitions We assessed both the lifetime and 12-month prevalence of sexual abuse and violence, in accordance with the WHO global study definition (Krug, Dahlberg, Mercy, Zwi, & Lozano, 2002). We defined sexual violence for the purposes of the study as including sexual touching, attempted sex, physically forced sex, and pressured sex. Sex or sexual intercourse was described to respondents as “someone penetrating your vagina or anus with their penis, hands, fingers, tongue, or other objects, or penetrating your mouth with their penis.” We classified respondents as victims of sexual violence in childhood if they reported having ever experienced any of the following before the age of 18 years: forced intercourse (unwanted intercourse imposed through physical force); coerced intercourse (unwanted intercourse imposed through non-physical pressure); attempted unwanted intercourse; unwanted sexual touching of the respondent (i.e., touching, kissing, grabbing, or fondling of sexual body parts); and forced touching of the perpetrator’s sexual body parts. Any respondent who answered “yes” to a specific question was asked a follow-up question about his or her age when the incident occurred. We defined children as any child below 18 years as stipulated in the Convention on the Rights of the Child, ratified by Zimbabwe. Study Design and Sampling Procedure We interviewed adolescents and young adults between 13 and 24 years of age for this study. Women and men aged 18–24 years were included to assess the prevalence of violence in those who had lived through the entire age range of interest (under 18 years) at a young enough age so that recall bias would have a negligible effect, as with previous studies of this kind (Reza et al., 2009). At the time of the study, three preceding national violence-against-children surveys in the region had already piloted this age range and the violence measures—Swaziland, Kenya, and Tanzania. We drew a nationally representative sampling frame from the 2002 Zimbabwe Population Census Master Sample to develop a stratified two-stage sample design. We selected 7,797 households over 223 enumeration areas for inclusion in the survey. When visited, 96% of the households were occupied, and 30% of these had eligible respondents who agreed to participate. At the second stage of sampling, we randomly selected 35 households from the listing for interviews in each enumeration area. The household response rate was 92.9% for males and 91.5% for females. If more than one eligible person was identified in a household, the respondent was randomly selected using the Kish method (Kish, 1949). In all, 2,410 respondents were interviewed, of which 1,062 (44%) were female and 1,348 (56%) were male. Amongst the 18–24 year old age range, 567 were females and 589 were male. A split sample approach was used (separate enumeration areas for males and for females) to protect the confidentiality of respondents and to reduce the chance that a perpetrator and survivor of violence in the same community would both be interviewed.
Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
3
Procedures and Ethical Controls This government-led study aimed to inform future investments in national policy and program. The Zimbabwe National Statistics Agency (ZIMSTAT) conducted the study under the coordination of an inter-ministerial committee co-chaired by the ministries responsible for child protection and health. UNICEF, a child rights organization, and the Centers for Disease Control and Prevention (CDC) provided technical support. ZIMSTAT aligned the study’s ethical protocol with Zimbabwe’s Census and Statistics Act of 2007, the Children’s Act, and the Ministry of Labor and Social Services’ National Child Protection and Participation Policy. The CDC Institutional Review Board, Medical Research Council of Zimbabwe, and later the Attorney-General in Zimbabwe approved the protocol. The protocol stipulated training for enumerators on reporting and referral procedures for any child who exhibited signs of distress or who asked for help with current or past violence during the interview. It also included information on locally available services for post-rape care for any child reporting sexual violence in the preceding 72 h, as well as for other urgent cases. The protocol described partnerships in place with child protection agencies, including statutory social services and local civil society organizations. Researchers interviewed respondents in private and provided them with a list of available child-sensitive health (including HIV testing and counseling), welfare, and legal services in their area at the end of the interview. Research supervisors reported cases for referral to designated social workers with the Department of Social Services. A local child welfare organization conducted community-based awareness activities about their free helpline and 24-h drop-in support centers to support child victims who wished to report anonymously about any aspect treated in the study. UNICEF and the Ministry’s Department of Social Services monitored compliance throughout the survey. We noted the ongoing risk of carrying out such a survey in a context of limited service coverage, a situation not limited to Zimbabwe (García-Moreno et al., 2014). Our survey instruments and methodology followed standardized and previously tested survey methods from the region (Reza et al., 2009; UNICEF, CDC, & Muhumbili University of Health and Allied Sciences, 2011; UNICEF, CDC, & the Kenya National Bureau of Statistics, 2012). We sought information from male and female respondents about the circumstances of the first incident of sexual violence during childhood, such as the type of sexual violence, the relationship with the perpetrator, disclosure, possible health-related conditions, and service seeking behaviors. Participants provided details about their age, education, religion, marital status, community setting, and orphan status. We also measured risky sexual behaviors, such as condom use, multiple sexual partnerships, and HIV testing. We aligned, where possible, the survey instruments with other national and inter-country survey instruments, such as the Demographic and Health Surveys, Behavioral Surveillance Surveys (World Health Organization, 2000), the CDC Youth Risk Behavior Surveillance System, the Hopkins Symptoms Checklist, and the WHO Multi-country Study on Women’s Health and Domestic Violence Against Women (García-Moreno et al., 2005). An inter-ministerial steering committee reviewed the methodology, and we modified the instruments to reflect the local context. We translated the survey from English into two local languages, Shona and Ndebele, and then back into English. Survey team members fluent in these languages reviewed and revised the questionnaires. Participants could also choose to take the survey in English. The use of these three languages is consistent with practice for previous national surveys administered across Zimbabwe. Field supervisors employed by ZIMSTAT conducted a pilot survey after training in the methodology, questionnaire interviewing on sensitive topics, and ethical considerations and responses. The pilot was undertaken in two villages that had not been selected as enumeration areas in the sampling frame, using the “split sample” approach described above. We selected both urban and rural sites for the test. We further refined the survey instruments, and the supervisors and enumerators received detailed training for the full fieldwork, including the protocol for referral of child protection cases. Data collection took place during the school holidays between August and October 2011 so that we could include children enrolled in boarding schools. Researchers adopted special measures and caution with data collection owing to the sensitive nature of the information. Enumerators of the same sex as the eligible respondents conducted the interviews. We implemented a three-stage informed consent procedure. Initially, the head of household consented to participate in a household questionnaire based on a general explanation that the survey sought to explore the health and life experiences of adolescents. At this stage, we adhered to ethical standards that recommend excluding references to violence from the first stage of consent (World Health Organization, 2001). Parents then provided further consent, if the eligible respondent was under age 18 years, using the same general explanation provided in the first stage, involving an initial information form. The enumerator then asked to have the rest of the consent process and interview take place in a private space. Once alone with the respondent, the enumerator provided a comprehensive disclosure of the full nature of the survey and sought consent, or assent, for respondents under 18 years. Researchers also sought consent at various points during the interview in advance of especially sensitive questions about violence and sexual behavior. At this stage, just 1% of respondents declined to complete the survey.
Statistical Analyses ZIMSTAT led the data analysis, with technical support from CDC. All data was double-entered, captured using CSPro 4.0 and analyzed using SAS 9.3. ZIMSTAT applied a three-step weighting process—calculating base weights, non-response Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS
4
L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
Table 1 Percentage of respondents aged 18–24 years who reported experiencing any form of sexual violence prior to age 18 by the perpetrator of the first event of sexual violence. Perpetrators of sexual violence
Boyfriend Husband Uncle Neighbor Friend Other relative Authority figure Other person Domestic worker N
Females
Males
Perpetrators of sexual violence
%
LLa
ULb
77.7 9.0 6.1 10.3 4.9 5.2 1.2 2.8 0.0
70.5 4.8 2.4 4.5 1.3 1.7 0.0 0.0 0.0 186
84.9 13.3 9.8 16.2 8.5 8.8 3.4 5.8 0.0
Girlfriend Wife Uncle Neighbor Friend Other relative Authority figure Other person Domestic worker N
%
LLa
ULb
26.7 0.0 1.2 32.5 3.1 14.0 0.9 13.1 17.9
11.3 0.0 0.0 17.9 0.0 4.6 0.0 1.8 8.7 55
42.1 0.0 3.7 47.0 7.8 23.4 2.8 24.4 35.2
Source: ZIMSTAT et al. (2013). Notes: a Lower confidence limit (LL) at 95% CI. b Upper confidence limit (UL) at 95% CI. Table 2 Percentage of respondents aged 18–24 years reporting any form of sexual violence by gender. Types of sexual violence
Sexual touching Attempted sex Physically forced sex Pressured sex N
Females a
Males b
%
LL
UL
%
20.2 15.0 9.0 7.4
16.3 11.6 5.9 5.0 567
24.1 18.3 12.2 9.8
5.6 3.8 0.4 1.4
LLa
ULb
3.3 2.2 0.0 0.5 589
7.9 5.5 1.0 2.3
Source: ZIMSTAT et al. (2013). Notes: a Lower confidence limit (LL) at 95% CI. b Upper confidence limit (UL) at 95% CI.
adjustments, and calibration. We examined associations between sexual violence, individual demographics, and perpetrator variables. For the purpose of this article, we report results for the 18–24 age range only. Results More than half of female respondents (56.5%; 95% CI, here and in all following results [52.4, 60.6]) and 53.2% of male respondents [49.7, 56.7] were aged 18–24 years old. Most female respondents had completed primary school: 94.1% of girls [91.6, 96.7] and 90.5% of boys [87.7, 93.4]. One-third of all girls (32.5%) and almost one in 10 boys (8.9%) reported experiencing sexual violence before the age of 18 years. Approximately one-third of victims reported that the perpetrator of the first incident appeared to be ten or more years older: 29.3% of females [22.2, 36.4] and 35.0% of males [17.7, 52.3]. Someone known to the survivor most commonly perpetrated sexual violence against females: 77.7% reported that a boyfriend had perpetrated the first incident. For males, 26.7% and 32.5% of the perpetrators were girlfriends and neighbors, respectively (Table 1). Notably, very few victims cited an authority figure as the perpetrator of their first incident of sexual violence: 1.2% for females and 0.9% for males. This may require a shift in advocacy and social awareness efforts currently targeting authority figures (defined in the study as “police, military, or teachers”) to community efforts targeting young people and neighbors. Females and males reporting sexual violence experienced various forms of sexual violence before age 18 years. For respondents 18–24 years, 20.2% of females and 5.6% of males reported experiencing sexual touching. Fifteen percent of girls and 3.8% of boys reported at least some type of attempted unwanted sex. Physically forced sex and pressured sex were less commonly reported (Table 2). Most female (62.7% [53.4, 72.0]) and male victims of sexual violence (47.9% [32.3, 63.5]) experienced more than one incident of sexual violence prior to age 18 years. For females reporting sexual violence, 51.8% told someone about the incident, most commonly a relative or friend. Of these, just 2.7% of girls received any form of professional help (Table 3). We observed a comparable trend for male survivors. We did not assess reasons for low health-seeking behavior except to note that the time of the study survivors of sexual violence were obliged to pay for post-rape medical care, including emergency contraception. Trained government social workers numbered less than 150 nationally (Oxford Policy Management & Ministry of Labour and Social Services, 2010). Despite the high risk of contracting HIV in Zimbabwe for adolescent girls, only 25% [16.5, 32.8] of female victims of childhood Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
5
Table 3 Percentage of respondents aged 18–24 years, among those who experienced sexual violence before age 18, who received help from a clinic or nongovernmental organization. 18–24 year olds who experienced any sexual violence prior to age 18
Received professional help (such as from a clinic or NGO) for any incident of sexual violence
Females
Males a
b
n
%
LL
UL
n
%
LLa
ULb
183
2.7
0.4
5.0
55
2.4
0.0
5.5
Source: ZIMSTAT et al. (2013). Notes: a Lower confidence limit (LL) at 95% CI. b Upper confidence limit (UL) at 95% CI.
sexual violence in our study had sought HIV testing and even fewer boys (16.5% [0.0, 34.6]). This result is of some concern because although HIV prevalence rates in Zimbabwe are falling, they are still amongst the highest in the world at 15% (UNICEF, 2015). Amongst young people, only 14% of males and 30% of females reported being tested and receiving treatment (UNICEF, 2015). An association between orphan status and sexual violence emerged in our findings, especially amongst female respondents aged 13–17 years, but further analysis is required. We assessed other associated factors, such as drug and alcohol use in the preceding 12 months, but the sample size was too small for analysis (n < 25). We were also not able to analyze data collected for females and males who experienced any form of sexual violence and who reported that they had multiple sexual partners before 18 years for the same reason.
Discussion and Conclusions This study found that one in three girls in Zimbabwe experience some form of sexual violence before they turn 18. Sexual violence prevalence in Zimbabwe is consistent with high prevalence rates of between 27% and 38% amongst the same age range in Swaziland, Kenya, and Tanzania (Reza et al., 2009; UNICEF et al., 2011, 2012). These estimates are much higher amongst girls than those published in meta-analyses on childhood sexual violence prevalence, which estimate 7.6% for boys, and 18.0% for girls globally (Stoltenberg et al., 2011). They are also higher than some estimates from the continent. A recent literature review of childhood sexual abuse in Africa found rates ranging from 1.6% to 77.7%, depending on the definitions used, as well as the populations sampled (Meinck et al., 2014). Arguably, the use of a standard definition of childhood sexual violence and a common methodology is providing more reliable prevalence data on this common concern in the region. The circumstances of the first instances of sexual violence for girls can be grouped into two predominant patterns. First, most survivors experience sexual violence in the context of a relationship, and the majority of perpetrators are the victim’s male peers. This dynamic may be linked to the prevalence of intimate partner violence against women. The Zimbabwe Demographic Health Survey found that domestic violence is a common experience for women, with 91% of these survivors indicating that a partner perpetrated the violence. These studies suggest that for many Zimbabwean women, their first relationships as teenagers are characterized by violence. Hlavka (2014) argues that almost half of all children experience physical violence from a parent or adult relative in their lifetime, and that these experiences are likely to inculcate a sense of normalcy toward violent behavior. Recent systematic reviews suggest that the perpetration of intimate partner violence is associated with exposure to child maltreatment and abuse (Fulu, Jewkes, Roselli, & García-Moreno, 2013; Heise, 2012). We argue that this cycle of violence requires immediate attention with early interventions targeting attitudes and beliefs about the justifiability of domestic violence within households and communities. As Heise (2012) points out, the usual efforts to equip women with the knowledge and skills to recognize and walk away from violent relationships are often not successful. Prevention efforts may have more impact if they target girls’ and boys’ early romantic relationships, promoting self-empowerment and knowledge about their sexual and reproductive rights, in addition to informing them of where and how to seek professional support and care. Second, at least half of all survivors do not tell anyone, and those who do are unlikely to seek or receive professional help. A better understanding is needed of why help-seeking behaviors are so low amongst adolescent boys and girls in Zimbabwe and in the Sub-Saharan African region. Past studies have revealed a few reasons. In Swaziland, for example, over 60% of females who reported sexual violence indicated that they would have liked help, but they felt that services were not available (Reza et al., 2009). The well-documented health consequences of sexual violence worldwide stress a need to increase rates of early reporting to professional support services (García-Moreno & Watts, 2011). Additional studies have shown links between sexual violence and HIV infection (Jewkes, Dunkle, Nduna, & Shai, 2010; World Health Organization & UNAIDS, 2013). Our results confirm findings from other smaller studies in Zimbabwe (Birdthistle et al., 2011; Chatora, Mapingure, & Hatzold, 2014) that the majority of adolescent and adult sexual violence victims do not seek HIV testing or post-exposure prophylaxis (PEP), despite the high prevalence of the disease in the country. HIV policy and programming needs accelerated actions to reach out to and support both boys and girls at risk of and exposed to sexual violence. Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS
6
L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
Targeted programs for prevention amongst adolescents require investment. Good models already exist in Zimbabwe, such as Zvandiri (meaning “As I Am”), which includes dedicated efforts for adolescents coping with unwanted pregnancies, rape, and stigma, which are linked to broader counseling, treatment, and referral programming for HIV treatment and care. The Ministry of Health and Child Care has adopted this program in 20 communities across Zimbabwe, where boys and girls are reporting increased adherence to anti-retroviral therapies, increasingly positive relationships with sexual partners, and a reduction in HIV risk behaviors (Sharer & Fullem, 2012). Government and child protection civil society partners in the country have invested increasingly in ways to prevent and respond to sexual violence, and children have indicated that such measures have improved their well-being, subsequent to the issuance of the survey findings (Sammon et al., 2014). In line with the growing international evidence base linking economic and social empowerment for adolescent girls (Marcus & Page, 2014), programs for the economic empowerment of girls and young women are now recommended in the new Girls’ Empowerment Framework for the country (Together for Girls, 2014). Interventions should address other forms of violence against children, including physical violence, which may occur concurrently. Another study from Zimbabwe in 2013 reported that 42.7% of females aged 13–20 years living in extremely poor households had experienced physical violence in the previous 12 months (American Institutes for Research, 2013). Our survey has several strengths. First, it provides a baseline for monitoring success in sexual violence prevention programming for children and adolescents by measuring health seeking behaviors amongst child victims. It also provides data on perpetrator types and associated factors. Second, it has extremely high response rates for both females and males, despite the subject matter and the sensitive political context in which the study took place. Third, its results are internationally comparable. Zimbabwe’s efforts to conduct this survey are contributing to a more accurate picture of the extent of sexual violence against children in the region and beyond. The survey is also subject to several limitations. Reports of childhood sexual violence were retrospective and still subject to recall bias. Additionally, we collected information only about the circumstances surrounding the first incident of sexual violence, which does not represent possible subsequent incidents of sexual violence. Another limiting factor is that the study did not differentiate between sexual violence prevalence in rural vs. urban settings. Nor did our study interrogate the impact of study conditions on responses—for example, the use of face-to-face interviews rather than self-administered questionnaires or the effect of the sex of the interviewer on participant disclosure. For further study, we recommend more analysis by the government and its research partners on risk and protective factors, such as relationship to the caregiver, orphan status, and household size to inform prevention programming in Zimbabwe and elsewhere. Evidence about effective programs for addressing childhood sexual abuse in middle-to-low income settings is growing, but still generally lacking (Sommarin, Kilbane, Mercy, Moloney-Kitts, & Ligiero, 2014). In-depth study is also needed on associated contextual factors such as the recent political and economic strife in the country. Whilst we are confident that the findings of this study are nationally representative, we recommend an independent analysis of the findings compared to non-governmental surveys and other studies in the country to ensure the objectivity of the results. We acknowledge an ongoing ethical concern about the still emerging case management system in the country hampering comprehensive care and support for all victims when studies of this kind are undertaken. These limitations notwithstanding, the survey has a number of positive implications. The findings indicate that many girls do grow up healthy and safe in Zimbabwe. For every girl who experiences sexual violence, two of her peers do not. This represents a critical entry point for further research to promote prevention and positive practices amongst adolescents, parents, and other caregivers. The survey also has played an important national advocacy role, emphasizing the vulnerability of both boys and girls, and especially of adolescent girls, in the country. That this discussion on children’s protection is taking place in public, in the media, and in the national development agenda discourse means that the issue of violence against children is no longer a marginalized phenomenon, but is rather deemed worthy of interest and preventative investment. Acknowledgements We thank the study participants who bravely shared their experience of violence and the local civil society organizations supporting their care, including Childline Zimbabwe and AFRICAID. We thank the Ministry of Public Service, Labor and Social Welfare for their leadership of the referral protocol and the Ministry of Women Affairs, Gender and Community Development for publicly launching the study findings and Girls Empowerment Framework. We thank the US Centers for Disease Control and Prevention for providing ongoing technical support throughout the study and to the review of this paper. References American Institutes for Research. (2013). Baseline report for Zimbabwe’s Harmonised Social Cash Transfer program. Washington, DC: Author. Birdthistle, I. J., Floyd, S., Mwanasa, S., Nyagadza, A., Gwiza, E., & Glynn, J. R. (2011). Child sexual abuse and links to HIV and orphanhood in urban Zimbabwe. Journal of Epidemiology and Community Health, 65, 1075–1082. http://dx.doi.org/10.1136/jech.2009.094359 Chatora, K., Mapingure, M., & Hatzold, K. (2014, July). Barriers and motivators to accessing HIV post-exposure prophylaxis services among adult female survivors of sexual violence in Zimbabwe: Mixed methods results. Poster session presented at the 20th International AIDS Conference, Melbourne, Australia. Retrieved from http://pag.aids2014.org/Abstracts.aspx?AID=6514 Elhawary, S., Foresti, M., & Pantuliano, S. (2011). Development, security and transitions in fragile states. Meeting Series Report London, England: ODI. García-Moreno, C., & Watts, C. (2011). Violence against women: An urgent public health priority. Bulletin of the World Health Organization, 89, 2. http://dx.doi.org/10.2471/BLT.10.085217
Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
7
García-Moreno, C., Jansen, H. A. F. M., Ellsberg, M., Heise, L., & Watts, C. (2005). WHO Multi-country study on women’s health and domestic violence against women. Geneva, Switzerland: World Health Organization. García-Moreno, C., Hegarty, K., d’Oliveira, A. F. L., Koziol-MacLain, J., Colombini, M., & Feder, G. (2014). The health-systems response to violence against women. Lancet, http://dx.doi.org/10.1016/S0140-6736(14)61837-7 Fulu, E., Jewkes, R., Roselli, T., & García-Moreno, C. (2013). Prevalence of and factors associated with male perpetration of intimate partner violence: Findings from the UN multi-country cross-sectional study on men and violence in Asia and the Pacific. Lancet Global Health, 1, e187–e207. http://dx.doi.org/10.1016/S2214-109X (13)70074-3 Heise, L. (2012). What works to prevent partner violence? An evidence overview. Retrieved from http://researchonline.lshtm.ac.uk/21062/ Hlavka, H. R. (2014). Normalizing sexual violence: Young women account for harassment and abuse. Gender and Society, 28, 337–358. http://dx.doi.org/10.1177/0891243214526468 Hopkins Symptom Checklist. Retrieved from http://hprt-cambridge.org/screening/hopkins-symptom-checklist/ Jewkes, R. K., Dunkle, K., Nduna, M., & Shai, N. (2010). Intimate partner violence, relationship power inequity, and incidence of HIV infection in young women in South Africa: A cohort study. Lancet, 376, 41–48. http://dx.doi.org/10.1016/S0140-6736(10)60548-X Kish, L. (1949). A procedure for objective respondent selection within the household. Journal American Statistics Association, 44, 380–387. http://dx.doi.org/10.1080/01621459.1949.10483314 Krug, E. G., Dahlberg, L. L., Mercy, J. A., Zwi, A. B., & Lozano, R. (2002). World report on violence and health. Geneva, Switzerland: World Health Organization. Marcus, R., & Page, E. (2014). Changing discriminatory norms affecting adolescent girls through communication activities: A review of evidence. Sussex, England: Overseas Development Institute. Meinck, F., Cluver, L., Boyes, M., & Mhlongo, E. (2014). Risk and protective factors for physical and sexual abuse of children and adolescents in Africa: A review and implications for practice. Trauma, Violence, & Abuse, 16, 81–107. http://dx.doi.org/10.1177/1524838014523336 Oxford Policy Management & Ministry of Labour and Social Services. (2010). A capacity audit of the Department of Social Services. Harare, Zimbabwe: United Nations Children’s Fund. Pinheiro, P. (2006). World report on violence against children. Geneva, Switzerland: United Nations. Reza, A., Breiding, M. J., Gulaid, J., Mercy, J. A., Blanton, C., Mthethwa, Z., Anderson, M., Bamrah, S., Dahlberg, L. L., & Anderson, M. (2009). Sexual violence and its health consequences for female children in Swaziland: A cluster survey study. Lancet, 373, 1966–1972. http://dx.doi.org/10.1016/S0140-6736(09)60247-6 Sharer, M., & Fullem, A. (2012). Transitioning of care and other services for adolescents living with HIV in Sub-Saharan Africa. Arlington, VA: USAID’s AIDS Support and Technical Assistance Resources, AIDSTAR-One, Task Order 1. Retrieved from http://www.aidstar-one.com/sites/default/files/AIDSTAR-One TechnicalBrief ALHIV Transition.pdf Sammon, E., Godwin, M., Rumble, L., Nolan, A., Matsika, A. B., & Mayanga, N. (2014, August). Make the promise true: A monitoring and evaluation framework for measuring quality in child protection service delivery in Zimbabwe. Child Indicators Research, http://dx.doi.org/10.1007/s12187-014-9267-1 Sommarin, C., Kilbane, T., Mercy, J., Moloney-Kitts, M., & Ligiero, D. (2014). Preventing sexual violence and HIV in children. Journal of Acquired Immune Deficiency Syndrome, 66, S217–S223. http://dx.doi.org/10.1097/QAI.0000000000000183 Stoltenberg, M., van Ijzendoorn, M. H., Euser, E. M., & Bakermans-Kranenburg, M. J. (2011). A global perspective on child sexual abuse: Meta-analysis of prevalence around the world. Child Maltreatment, 16, 79–101. http://dx.doi.org/10.1177/1077559511403920 Together for Girls. (2014). Zimbabwe to combat gender-based violence with its first national girls and 2014 young women’s empowerment framework. Retrieved from http://www.togetherforgirls.org/safe/?p=464 Together for Girls. (2013). Linking violence against children surveys to coordinated and effective action. Factsheet. Retrieved from http://www.togetherforgirls.org/wp-content/uploads/Together-for-Girls-Process-Paper Linking-VACS-to-Coordinated-and-Effecti. . ..pdf Undie, C., Mullick, S., & Askew, I. (2013). The missing ‘C’: Sexual violence against children in sub-Saharan Africa. Retrieved from http://www.unicef-irc.org/research-watch/Violence-against-children–a-silent-threat/987/ UNICEF. (2015). The state of the world’s children report. New York, NY: UNICEF. UNICEF. (2014). Hidden in plain sight: A statistical analysis of violence against children. New York, NY: UNICEF. UNICEF, CDC, & Muhumbili University of Health and Allied Sciences. (2011). Violence against Children in Tanzania: Findings from a national survey 2009. Dar es Salaam, Tanzania: UNICEF. UNICEF, CDC, & the Kenya National Bureau of Statistics. (2012). Violence against children in Kenya: Findings from a 2010 national survey. Nairobi, Kenya: UNICEF. United Nations General Assembly. (2000). Resolution adopted by the General Assembly A/RES/55/2: United Nations millennium declaration. New York, NY: United Nations. Retrieved from http://www.un.org/millennium/declaration/ares552e.pdf Veenema, T., Thornton, C., & Corley, A. (2014). The public health crisis of child sexual abuse in low and middle-income countries: An integrative review of the literature. International Journal of Nursing Studies, 52, 864–881. http://dx.doi.org/10.1016/j.ijnurstu.2014.10.017 Watts, C., & Zimmerman, C. (2002). Violence against women: Global scope and magnitude. Lancet, 359, 1232–1237. http://dx.doi.org/10.1016/S0140-6736(02)08221-1 World Health Organization (WHO), United Nations Office on Drugs and Crime (UNODC), & United Nations Development Programme (UNDP). (2014). Global status report on violence prevention 2014. Geneva, Switzerland: WHO. Retrieved from http://www.who.int/violence injury prevention/violence/status report/2014/en/ World Health Organization. (2000). Behavioural surveillance surveys: Guidelines for repeated behavioural surveys in populations at risk of HIV. Geneva, Switzerland: WHO. Retrieved from http://www.who.int/hiv/strategic/pubbss/en/ World Health Organization. (2001). Putting women first: Ethical and safety recommendations for research on domestic violence against women. Geneva, Switzerland: WHO. Retrieved from http://whqlibdoc.who.int/hq/2001/WHO FCH GWH 01.1.pdf?ua=1 World Health Organization, & UNAIDS. (2013). 16 ideas for addressing violence against women in the context of the HIV epidemic: A programming tool. Geneva, Switzerland: WHO. ZIMSTAT, & ICF International. (2012). Zimbabwe demographic and health survey 2010-11 (ZDHS). Calverton, MD: IMSTAT and ICF International Inc. ZIMSTAT, UNICEF, & CCORE. (2013). National baseline survey on life experiences of adolescents 2011. Harare, Zimbabwe: ZIMSTAT.
Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
ARTICLE IN PRESS Child Abuse & Neglect xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Child Abuse & Neglect
Research article
Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls夽 Lauren Rumble a,∗ , Taizivei Mungate b , Handrick Chigiji b , Peter Salama c , Anthony Nolan d , Elayn Sammon e , Leon Muwoni f a United Nations Children’s Fund (UNICEF) Indonesia, Wisma Metropolitan II, 11th Floor, Jl. Jend. SudirmanKav. 31, Jakarta 12920, Indonesia b Zimbabwe National Statistics Agency (ZIMSTAT), 20th Floor, Kaguvi Building, Cnr 4th Central Avenue, Harare, Zimbabwe c United Nations Children’s Fund (UNICEF), Ethiopia, P.O. Box 1169, Africa Hall, Addis Ababa, Ethiopia d United Nations Children’s Fund (UNICEF) Sudan, UNICEF Totto Chan Compound, P.O. Box 45, Juba, South Sudan e UNICEF Mozambique, 1440 Zimbabwe Avenue, Maputo, Mozambique f Child Protection Specialist, UNICEF Zimbabwe, 6 Fairbridge Avenue, Belgravia, Zimbabwe
a r t i c l e
i n f o
Article history: Received 3 November 2014 Received in revised form 16 April 2015 Accepted 27 April 2015 Available online xxx Keywords: Adolescent girls Sexual violence Children
a b s t r a c t Sexual abuse during childhood is a public health and human rights concern throughout the world, including Sub-Saharan Africa. In 2011, Zimbabwe initiated national prevalence data collection on violence against children to inform government policy and programs. We interviewed 567 females and 589 males, aged 18–24 years following standardized and previously tested survey methods from the region. Of females 32.5%, and of males 8.9%, reported experiencing sexual violence before age 18. Most female (62.7%) and male (47.9%) victims of sexual violence experienced more than one incident of sexual violence prior to age 18 years. Three in four females (77.7%) and one in four males (26.7%) of those who experienced sexual violence reported that the first incident was perpetrated by a boyfriend or girlfriend. Few victims received professional help (2.7% of females and 2.4% of males who had reported experiencing sexual violence). Violence against girls is at epidemic levels in Zimbabwe. Most sexual violence against girls occurs within the context of peer relationships. Child victims who seek potentially life-saving support tend not to receive it. This study is evidence of a national public health and child rights emergency in the country and a case for increased, longer-term investment by the government and its development partners in policy reform for enhancing adolescent girls’ empowerment and protection. © 2015 Elsevier Ltd. All rights reserved.
Introduction Reducing violence is fundamental to creating the social conditions for stability and development (Elhawary, Foresti, & Pantuliano, 2011). Despite concern, major gaps in knowledge about the extent of the problem persist (World Health Organization, 2014). Researchers have made a certain amount of progress in some areas. In Zimbabwe, for example, investments in national health surveys in the past decade have resulted in the availability of more robust data on how violence
夽 Funding was provided by UNICEF global thematic funding and the Nduna Foundation. Neither party was involved in the analysis nor played a lead role at any stage of the design, data collection or interpretation. We confirm our free access to the data and no interference from any party during publication of the results of the study or in the writing of the article. ∗ Corresponding author. http://dx.doi.org/10.1016/j.chiabu.2015.04.015 0145-2134/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS
2
L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
affects women. As one instance, the Zimbabwe Demographic Health Survey reported in 2012 that 22% of Zimbabwean women said that their first sexual experience was forced. The same survey noted that 33% of men and 40% of women believed that wife beating is acceptable for some circumstances (ZIMSTAT & ICF International, 2012). These findings are in line with the global multi-country study on sexual violence prevalence led by the World Health Organization (WHO), which estimates that up to 59% of women worldwide experience sexual violence in their lifetimes (García-Moreno, Jansen, Ellsberg, Heise, & Watts, 2005). Much less progress has been made in obtaining data on childhood sexual violence; it tends to be under-reported and undocumented. The first global effort to document the scale of violence against children found evidence that sexual violence was widespread across multiple settings but that systematic and comprehensive data needed to be collected in many countries (Pinheiro, 2006). Since then, few studies from Africa have used consistent prevalence measures or analyzed risk and protective factors (Meinck, Cluver, Boyes, & Mhlongo, 2014). Evidence from low- and middle-income countries is generally lacking (Veenema, Thornton, & Corley, 2014). Where such data exists, it is likely an underestimate because there are substantial barriers to disclosure (Watts & Zimmerman, 2002). Research on childhood sexual violence across the world employs different definitions of sexual violence against children, making it difficult to compare prevalence (Stoltenberg, van Ijzendoorn, Euser, & Bakermans-Kranenburg, 2011). Researchers also face considerable ethical concerns in undertaking such research (Undie, Mullik, & Askew, 2013). The United Nations Children’s Fund (UNICEF) has noted that under-reporting fuels public misconception that violence is a marginal phenomenon (UNICEF, 2014). Although the 2000 United Nations Millennium Declaration refers to freedom from violence as a fundamental value, no Millennium Development Goal was dedicated to addressing sexual or other forms of violence (United Nations General Assembly, 2000). Consequently, public policy discourse has not attended as much to sexual and other forms of violence against children as to other child rights, including the rights to health and education. In 2011, Zimbabwe undertook its first nationally representative survey on physical, emotional, and sexual violence against children. The purpose of the study was to examine the nature and extent of violence affecting children in the country (ZIMSTAT, UNICEF, & CCORE, 2013) to inform policy development. To measure the prevalence of violence against children and associated factors, we applied a methodology successfully implemented in three Sub-Saharan African countries (Together for Girls, 2013). In this article, we report on the study results that pertain especially to sexual violence. Methods Definitions We assessed both the lifetime and 12-month prevalence of sexual abuse and violence, in accordance with the WHO global study definition (Krug, Dahlberg, Mercy, Zwi, & Lozano, 2002). We defined sexual violence for the purposes of the study as including sexual touching, attempted sex, physically forced sex, and pressured sex. Sex or sexual intercourse was described to respondents as “someone penetrating your vagina or anus with their penis, hands, fingers, tongue, or other objects, or penetrating your mouth with their penis.” We classified respondents as victims of sexual violence in childhood if they reported having ever experienced any of the following before the age of 18 years: forced intercourse (unwanted intercourse imposed through physical force); coerced intercourse (unwanted intercourse imposed through non-physical pressure); attempted unwanted intercourse; unwanted sexual touching of the respondent (i.e., touching, kissing, grabbing, or fondling of sexual body parts); and forced touching of the perpetrator’s sexual body parts. Any respondent who answered “yes” to a specific question was asked a follow-up question about his or her age when the incident occurred. We defined children as any child below 18 years as stipulated in the Convention on the Rights of the Child, ratified by Zimbabwe. Study Design and Sampling Procedure We interviewed adolescents and young adults between 13 and 24 years of age for this study. Women and men aged 18–24 years were included to assess the prevalence of violence in those who had lived through the entire age range of interest (under 18 years) at a young enough age so that recall bias would have a negligible effect, as with previous studies of this kind (Reza et al., 2009). At the time of the study, three preceding national violence-against-children surveys in the region had already piloted this age range and the violence measures—Swaziland, Kenya, and Tanzania. We drew a nationally representative sampling frame from the 2002 Zimbabwe Population Census Master Sample to develop a stratified two-stage sample design. We selected 7,797 households over 223 enumeration areas for inclusion in the survey. When visited, 96% of the households were occupied, and 30% of these had eligible respondents who agreed to participate. At the second stage of sampling, we randomly selected 35 households from the listing for interviews in each enumeration area. The household response rate was 92.9% for males and 91.5% for females. If more than one eligible person was identified in a household, the respondent was randomly selected using the Kish method (Kish, 1949). In all, 2,410 respondents were interviewed, of which 1,062 (44%) were female and 1,348 (56%) were male. Amongst the 18–24 year old age range, 567 were females and 589 were male. A split sample approach was used (separate enumeration areas for males and for females) to protect the confidentiality of respondents and to reduce the chance that a perpetrator and survivor of violence in the same community would both be interviewed.
Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
3
Procedures and Ethical Controls This government-led study aimed to inform future investments in national policy and program. The Zimbabwe National Statistics Agency (ZIMSTAT) conducted the study under the coordination of an inter-ministerial committee co-chaired by the ministries responsible for child protection and health. UNICEF, a child rights organization, and the Centers for Disease Control and Prevention (CDC) provided technical support. ZIMSTAT aligned the study’s ethical protocol with Zimbabwe’s Census and Statistics Act of 2007, the Children’s Act, and the Ministry of Labor and Social Services’ National Child Protection and Participation Policy. The CDC Institutional Review Board, Medical Research Council of Zimbabwe, and later the Attorney-General in Zimbabwe approved the protocol. The protocol stipulated training for enumerators on reporting and referral procedures for any child who exhibited signs of distress or who asked for help with current or past violence during the interview. It also included information on locally available services for post-rape care for any child reporting sexual violence in the preceding 72 h, as well as for other urgent cases. The protocol described partnerships in place with child protection agencies, including statutory social services and local civil society organizations. Researchers interviewed respondents in private and provided them with a list of available child-sensitive health (including HIV testing and counseling), welfare, and legal services in their area at the end of the interview. Research supervisors reported cases for referral to designated social workers with the Department of Social Services. A local child welfare organization conducted community-based awareness activities about their free helpline and 24-h drop-in support centers to support child victims who wished to report anonymously about any aspect treated in the study. UNICEF and the Ministry’s Department of Social Services monitored compliance throughout the survey. We noted the ongoing risk of carrying out such a survey in a context of limited service coverage, a situation not limited to Zimbabwe (García-Moreno et al., 2014). Our survey instruments and methodology followed standardized and previously tested survey methods from the region (Reza et al., 2009; UNICEF, CDC, & Muhumbili University of Health and Allied Sciences, 2011; UNICEF, CDC, & the Kenya National Bureau of Statistics, 2012). We sought information from male and female respondents about the circumstances of the first incident of sexual violence during childhood, such as the type of sexual violence, the relationship with the perpetrator, disclosure, possible health-related conditions, and service seeking behaviors. Participants provided details about their age, education, religion, marital status, community setting, and orphan status. We also measured risky sexual behaviors, such as condom use, multiple sexual partnerships, and HIV testing. We aligned, where possible, the survey instruments with other national and inter-country survey instruments, such as the Demographic and Health Surveys, Behavioral Surveillance Surveys (World Health Organization, 2000), the CDC Youth Risk Behavior Surveillance System, the Hopkins Symptoms Checklist, and the WHO Multi-country Study on Women’s Health and Domestic Violence Against Women (García-Moreno et al., 2005). An inter-ministerial steering committee reviewed the methodology, and we modified the instruments to reflect the local context. We translated the survey from English into two local languages, Shona and Ndebele, and then back into English. Survey team members fluent in these languages reviewed and revised the questionnaires. Participants could also choose to take the survey in English. The use of these three languages is consistent with practice for previous national surveys administered across Zimbabwe. Field supervisors employed by ZIMSTAT conducted a pilot survey after training in the methodology, questionnaire interviewing on sensitive topics, and ethical considerations and responses. The pilot was undertaken in two villages that had not been selected as enumeration areas in the sampling frame, using the “split sample” approach described above. We selected both urban and rural sites for the test. We further refined the survey instruments, and the supervisors and enumerators received detailed training for the full fieldwork, including the protocol for referral of child protection cases. Data collection took place during the school holidays between August and October 2011 so that we could include children enrolled in boarding schools. Researchers adopted special measures and caution with data collection owing to the sensitive nature of the information. Enumerators of the same sex as the eligible respondents conducted the interviews. We implemented a three-stage informed consent procedure. Initially, the head of household consented to participate in a household questionnaire based on a general explanation that the survey sought to explore the health and life experiences of adolescents. At this stage, we adhered to ethical standards that recommend excluding references to violence from the first stage of consent (World Health Organization, 2001). Parents then provided further consent, if the eligible respondent was under age 18 years, using the same general explanation provided in the first stage, involving an initial information form. The enumerator then asked to have the rest of the consent process and interview take place in a private space. Once alone with the respondent, the enumerator provided a comprehensive disclosure of the full nature of the survey and sought consent, or assent, for respondents under 18 years. Researchers also sought consent at various points during the interview in advance of especially sensitive questions about violence and sexual behavior. At this stage, just 1% of respondents declined to complete the survey.
Statistical Analyses ZIMSTAT led the data analysis, with technical support from CDC. All data was double-entered, captured using CSPro 4.0 and analyzed using SAS 9.3. ZIMSTAT applied a three-step weighting process—calculating base weights, non-response Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS
4
L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
Table 1 Percentage of respondents aged 18–24 years who reported experiencing any form of sexual violence prior to age 18 by the perpetrator of the first event of sexual violence. Perpetrators of sexual violence
Boyfriend Husband Uncle Neighbor Friend Other relative Authority figure Other person Domestic worker N
Females
Males
Perpetrators of sexual violence
%
LLa
ULb
77.7 9.0 6.1 10.3 4.9 5.2 1.2 2.8 0.0
70.5 4.8 2.4 4.5 1.3 1.7 0.0 0.0 0.0 186
84.9 13.3 9.8 16.2 8.5 8.8 3.4 5.8 0.0
Girlfriend Wife Uncle Neighbor Friend Other relative Authority figure Other person Domestic worker N
%
LLa
ULb
26.7 0.0 1.2 32.5 3.1 14.0 0.9 13.1 17.9
11.3 0.0 0.0 17.9 0.0 4.6 0.0 1.8 8.7 55
42.1 0.0 3.7 47.0 7.8 23.4 2.8 24.4 35.2
Source: ZIMSTAT et al. (2013). Notes: a Lower confidence limit (LL) at 95% CI. b Upper confidence limit (UL) at 95% CI. Table 2 Percentage of respondents aged 18–24 years reporting any form of sexual violence by gender. Types of sexual violence
Sexual touching Attempted sex Physically forced sex Pressured sex N
Females a
Males b
%
LL
UL
%
20.2 15.0 9.0 7.4
16.3 11.6 5.9 5.0 567
24.1 18.3 12.2 9.8
5.6 3.8 0.4 1.4
LLa
ULb
3.3 2.2 0.0 0.5 589
7.9 5.5 1.0 2.3
Source: ZIMSTAT et al. (2013). Notes: a Lower confidence limit (LL) at 95% CI. b Upper confidence limit (UL) at 95% CI.
adjustments, and calibration. We examined associations between sexual violence, individual demographics, and perpetrator variables. For the purpose of this article, we report results for the 18–24 age range only. Results More than half of female respondents (56.5%; 95% CI, here and in all following results [52.4, 60.6]) and 53.2% of male respondents [49.7, 56.7] were aged 18–24 years old. Most female respondents had completed primary school: 94.1% of girls [91.6, 96.7] and 90.5% of boys [87.7, 93.4]. One-third of all girls (32.5%) and almost one in 10 boys (8.9%) reported experiencing sexual violence before the age of 18 years. Approximately one-third of victims reported that the perpetrator of the first incident appeared to be ten or more years older: 29.3% of females [22.2, 36.4] and 35.0% of males [17.7, 52.3]. Someone known to the survivor most commonly perpetrated sexual violence against females: 77.7% reported that a boyfriend had perpetrated the first incident. For males, 26.7% and 32.5% of the perpetrators were girlfriends and neighbors, respectively (Table 1). Notably, very few victims cited an authority figure as the perpetrator of their first incident of sexual violence: 1.2% for females and 0.9% for males. This may require a shift in advocacy and social awareness efforts currently targeting authority figures (defined in the study as “police, military, or teachers”) to community efforts targeting young people and neighbors. Females and males reporting sexual violence experienced various forms of sexual violence before age 18 years. For respondents 18–24 years, 20.2% of females and 5.6% of males reported experiencing sexual touching. Fifteen percent of girls and 3.8% of boys reported at least some type of attempted unwanted sex. Physically forced sex and pressured sex were less commonly reported (Table 2). Most female (62.7% [53.4, 72.0]) and male victims of sexual violence (47.9% [32.3, 63.5]) experienced more than one incident of sexual violence prior to age 18 years. For females reporting sexual violence, 51.8% told someone about the incident, most commonly a relative or friend. Of these, just 2.7% of girls received any form of professional help (Table 3). We observed a comparable trend for male survivors. We did not assess reasons for low health-seeking behavior except to note that the time of the study survivors of sexual violence were obliged to pay for post-rape medical care, including emergency contraception. Trained government social workers numbered less than 150 nationally (Oxford Policy Management & Ministry of Labour and Social Services, 2010). Despite the high risk of contracting HIV in Zimbabwe for adolescent girls, only 25% [16.5, 32.8] of female victims of childhood Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
5
Table 3 Percentage of respondents aged 18–24 years, among those who experienced sexual violence before age 18, who received help from a clinic or nongovernmental organization. 18–24 year olds who experienced any sexual violence prior to age 18
Received professional help (such as from a clinic or NGO) for any incident of sexual violence
Females
Males a
b
n
%
LL
UL
n
%
LLa
ULb
183
2.7
0.4
5.0
55
2.4
0.0
5.5
Source: ZIMSTAT et al. (2013). Notes: a Lower confidence limit (LL) at 95% CI. b Upper confidence limit (UL) at 95% CI.
sexual violence in our study had sought HIV testing and even fewer boys (16.5% [0.0, 34.6]). This result is of some concern because although HIV prevalence rates in Zimbabwe are falling, they are still amongst the highest in the world at 15% (UNICEF, 2015). Amongst young people, only 14% of males and 30% of females reported being tested and receiving treatment (UNICEF, 2015). An association between orphan status and sexual violence emerged in our findings, especially amongst female respondents aged 13–17 years, but further analysis is required. We assessed other associated factors, such as drug and alcohol use in the preceding 12 months, but the sample size was too small for analysis (n < 25). We were also not able to analyze data collected for females and males who experienced any form of sexual violence and who reported that they had multiple sexual partners before 18 years for the same reason.
Discussion and Conclusions This study found that one in three girls in Zimbabwe experience some form of sexual violence before they turn 18. Sexual violence prevalence in Zimbabwe is consistent with high prevalence rates of between 27% and 38% amongst the same age range in Swaziland, Kenya, and Tanzania (Reza et al., 2009; UNICEF et al., 2011, 2012). These estimates are much higher amongst girls than those published in meta-analyses on childhood sexual violence prevalence, which estimate 7.6% for boys, and 18.0% for girls globally (Stoltenberg et al., 2011). They are also higher than some estimates from the continent. A recent literature review of childhood sexual abuse in Africa found rates ranging from 1.6% to 77.7%, depending on the definitions used, as well as the populations sampled (Meinck et al., 2014). Arguably, the use of a standard definition of childhood sexual violence and a common methodology is providing more reliable prevalence data on this common concern in the region. The circumstances of the first instances of sexual violence for girls can be grouped into two predominant patterns. First, most survivors experience sexual violence in the context of a relationship, and the majority of perpetrators are the victim’s male peers. This dynamic may be linked to the prevalence of intimate partner violence against women. The Zimbabwe Demographic Health Survey found that domestic violence is a common experience for women, with 91% of these survivors indicating that a partner perpetrated the violence. These studies suggest that for many Zimbabwean women, their first relationships as teenagers are characterized by violence. Hlavka (2014) argues that almost half of all children experience physical violence from a parent or adult relative in their lifetime, and that these experiences are likely to inculcate a sense of normalcy toward violent behavior. Recent systematic reviews suggest that the perpetration of intimate partner violence is associated with exposure to child maltreatment and abuse (Fulu, Jewkes, Roselli, & García-Moreno, 2013; Heise, 2012). We argue that this cycle of violence requires immediate attention with early interventions targeting attitudes and beliefs about the justifiability of domestic violence within households and communities. As Heise (2012) points out, the usual efforts to equip women with the knowledge and skills to recognize and walk away from violent relationships are often not successful. Prevention efforts may have more impact if they target girls’ and boys’ early romantic relationships, promoting self-empowerment and knowledge about their sexual and reproductive rights, in addition to informing them of where and how to seek professional support and care. Second, at least half of all survivors do not tell anyone, and those who do are unlikely to seek or receive professional help. A better understanding is needed of why help-seeking behaviors are so low amongst adolescent boys and girls in Zimbabwe and in the Sub-Saharan African region. Past studies have revealed a few reasons. In Swaziland, for example, over 60% of females who reported sexual violence indicated that they would have liked help, but they felt that services were not available (Reza et al., 2009). The well-documented health consequences of sexual violence worldwide stress a need to increase rates of early reporting to professional support services (García-Moreno & Watts, 2011). Additional studies have shown links between sexual violence and HIV infection (Jewkes, Dunkle, Nduna, & Shai, 2010; World Health Organization & UNAIDS, 2013). Our results confirm findings from other smaller studies in Zimbabwe (Birdthistle et al., 2011; Chatora, Mapingure, & Hatzold, 2014) that the majority of adolescent and adult sexual violence victims do not seek HIV testing or post-exposure prophylaxis (PEP), despite the high prevalence of the disease in the country. HIV policy and programming needs accelerated actions to reach out to and support both boys and girls at risk of and exposed to sexual violence. Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS
6
L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
Targeted programs for prevention amongst adolescents require investment. Good models already exist in Zimbabwe, such as Zvandiri (meaning “As I Am”), which includes dedicated efforts for adolescents coping with unwanted pregnancies, rape, and stigma, which are linked to broader counseling, treatment, and referral programming for HIV treatment and care. The Ministry of Health and Child Care has adopted this program in 20 communities across Zimbabwe, where boys and girls are reporting increased adherence to anti-retroviral therapies, increasingly positive relationships with sexual partners, and a reduction in HIV risk behaviors (Sharer & Fullem, 2012). Government and child protection civil society partners in the country have invested increasingly in ways to prevent and respond to sexual violence, and children have indicated that such measures have improved their well-being, subsequent to the issuance of the survey findings (Sammon et al., 2014). In line with the growing international evidence base linking economic and social empowerment for adolescent girls (Marcus & Page, 2014), programs for the economic empowerment of girls and young women are now recommended in the new Girls’ Empowerment Framework for the country (Together for Girls, 2014). Interventions should address other forms of violence against children, including physical violence, which may occur concurrently. Another study from Zimbabwe in 2013 reported that 42.7% of females aged 13–20 years living in extremely poor households had experienced physical violence in the previous 12 months (American Institutes for Research, 2013). Our survey has several strengths. First, it provides a baseline for monitoring success in sexual violence prevention programming for children and adolescents by measuring health seeking behaviors amongst child victims. It also provides data on perpetrator types and associated factors. Second, it has extremely high response rates for both females and males, despite the subject matter and the sensitive political context in which the study took place. Third, its results are internationally comparable. Zimbabwe’s efforts to conduct this survey are contributing to a more accurate picture of the extent of sexual violence against children in the region and beyond. The survey is also subject to several limitations. Reports of childhood sexual violence were retrospective and still subject to recall bias. Additionally, we collected information only about the circumstances surrounding the first incident of sexual violence, which does not represent possible subsequent incidents of sexual violence. Another limiting factor is that the study did not differentiate between sexual violence prevalence in rural vs. urban settings. Nor did our study interrogate the impact of study conditions on responses—for example, the use of face-to-face interviews rather than self-administered questionnaires or the effect of the sex of the interviewer on participant disclosure. For further study, we recommend more analysis by the government and its research partners on risk and protective factors, such as relationship to the caregiver, orphan status, and household size to inform prevention programming in Zimbabwe and elsewhere. Evidence about effective programs for addressing childhood sexual abuse in middle-to-low income settings is growing, but still generally lacking (Sommarin, Kilbane, Mercy, Moloney-Kitts, & Ligiero, 2014). In-depth study is also needed on associated contextual factors such as the recent political and economic strife in the country. Whilst we are confident that the findings of this study are nationally representative, we recommend an independent analysis of the findings compared to non-governmental surveys and other studies in the country to ensure the objectivity of the results. We acknowledge an ongoing ethical concern about the still emerging case management system in the country hampering comprehensive care and support for all victims when studies of this kind are undertaken. These limitations notwithstanding, the survey has a number of positive implications. The findings indicate that many girls do grow up healthy and safe in Zimbabwe. For every girl who experiences sexual violence, two of her peers do not. This represents a critical entry point for further research to promote prevention and positive practices amongst adolescents, parents, and other caregivers. The survey also has played an important national advocacy role, emphasizing the vulnerability of both boys and girls, and especially of adolescent girls, in the country. That this discussion on children’s protection is taking place in public, in the media, and in the national development agenda discourse means that the issue of violence against children is no longer a marginalized phenomenon, but is rather deemed worthy of interest and preventative investment. Acknowledgements We thank the study participants who bravely shared their experience of violence and the local civil society organizations supporting their care, including Childline Zimbabwe and AFRICAID. We thank the Ministry of Public Service, Labor and Social Welfare for their leadership of the referral protocol and the Ministry of Women Affairs, Gender and Community Development for publicly launching the study findings and Girls Empowerment Framework. We thank the US Centers for Disease Control and Prevention for providing ongoing technical support throughout the study and to the review of this paper. References American Institutes for Research. (2013). Baseline report for Zimbabwe’s Harmonised Social Cash Transfer program. Washington, DC: Author. Birdthistle, I. J., Floyd, S., Mwanasa, S., Nyagadza, A., Gwiza, E., & Glynn, J. R. (2011). Child sexual abuse and links to HIV and orphanhood in urban Zimbabwe. Journal of Epidemiology and Community Health, 65, 1075–1082. http://dx.doi.org/10.1136/jech.2009.094359 Chatora, K., Mapingure, M., & Hatzold, K. (2014, July). Barriers and motivators to accessing HIV post-exposure prophylaxis services among adult female survivors of sexual violence in Zimbabwe: Mixed methods results. Poster session presented at the 20th International AIDS Conference, Melbourne, Australia. Retrieved from http://pag.aids2014.org/Abstracts.aspx?AID=6514 Elhawary, S., Foresti, M., & Pantuliano, S. (2011). Development, security and transitions in fragile states. Meeting Series Report London, England: ODI. García-Moreno, C., & Watts, C. (2011). Violence against women: An urgent public health priority. Bulletin of the World Health Organization, 89, 2. http://dx.doi.org/10.2471/BLT.10.085217
Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
G Model CHIABU-2978; No. of Pages 7
ARTICLE IN PRESS L. Rumble et al. / Child Abuse & Neglect xxx (2015) xxx–xxx
7
García-Moreno, C., Jansen, H. A. F. M., Ellsberg, M., Heise, L., & Watts, C. (2005). WHO Multi-country study on women’s health and domestic violence against women. Geneva, Switzerland: World Health Organization. García-Moreno, C., Hegarty, K., d’Oliveira, A. F. L., Koziol-MacLain, J., Colombini, M., & Feder, G. (2014). The health-systems response to violence against women. Lancet, http://dx.doi.org/10.1016/S0140-6736(14)61837-7 Fulu, E., Jewkes, R., Roselli, T., & García-Moreno, C. (2013). Prevalence of and factors associated with male perpetration of intimate partner violence: Findings from the UN multi-country cross-sectional study on men and violence in Asia and the Pacific. Lancet Global Health, 1, e187–e207. http://dx.doi.org/10.1016/S2214-109X (13)70074-3 Heise, L. (2012). What works to prevent partner violence? An evidence overview. Retrieved from http://researchonline.lshtm.ac.uk/21062/ Hlavka, H. R. (2014). Normalizing sexual violence: Young women account for harassment and abuse. Gender and Society, 28, 337–358. http://dx.doi.org/10.1177/0891243214526468 Hopkins Symptom Checklist. Retrieved from http://hprt-cambridge.org/screening/hopkins-symptom-checklist/ Jewkes, R. K., Dunkle, K., Nduna, M., & Shai, N. (2010). Intimate partner violence, relationship power inequity, and incidence of HIV infection in young women in South Africa: A cohort study. Lancet, 376, 41–48. http://dx.doi.org/10.1016/S0140-6736(10)60548-X Kish, L. (1949). A procedure for objective respondent selection within the household. Journal American Statistics Association, 44, 380–387. http://dx.doi.org/10.1080/01621459.1949.10483314 Krug, E. G., Dahlberg, L. L., Mercy, J. A., Zwi, A. B., & Lozano, R. (2002). World report on violence and health. Geneva, Switzerland: World Health Organization. Marcus, R., & Page, E. (2014). Changing discriminatory norms affecting adolescent girls through communication activities: A review of evidence. Sussex, England: Overseas Development Institute. Meinck, F., Cluver, L., Boyes, M., & Mhlongo, E. (2014). Risk and protective factors for physical and sexual abuse of children and adolescents in Africa: A review and implications for practice. Trauma, Violence, & Abuse, 16, 81–107. http://dx.doi.org/10.1177/1524838014523336 Oxford Policy Management & Ministry of Labour and Social Services. (2010). A capacity audit of the Department of Social Services. Harare, Zimbabwe: United Nations Children’s Fund. Pinheiro, P. (2006). World report on violence against children. Geneva, Switzerland: United Nations. Reza, A., Breiding, M. J., Gulaid, J., Mercy, J. A., Blanton, C., Mthethwa, Z., Anderson, M., Bamrah, S., Dahlberg, L. L., & Anderson, M. (2009). Sexual violence and its health consequences for female children in Swaziland: A cluster survey study. Lancet, 373, 1966–1972. http://dx.doi.org/10.1016/S0140-6736(09)60247-6 Sharer, M., & Fullem, A. (2012). Transitioning of care and other services for adolescents living with HIV in Sub-Saharan Africa. Arlington, VA: USAID’s AIDS Support and Technical Assistance Resources, AIDSTAR-One, Task Order 1. Retrieved from http://www.aidstar-one.com/sites/default/files/AIDSTAR-One TechnicalBrief ALHIV Transition.pdf Sammon, E., Godwin, M., Rumble, L., Nolan, A., Matsika, A. B., & Mayanga, N. (2014, August). Make the promise true: A monitoring and evaluation framework for measuring quality in child protection service delivery in Zimbabwe. Child Indicators Research, http://dx.doi.org/10.1007/s12187-014-9267-1 Sommarin, C., Kilbane, T., Mercy, J., Moloney-Kitts, M., & Ligiero, D. (2014). Preventing sexual violence and HIV in children. Journal of Acquired Immune Deficiency Syndrome, 66, S217–S223. http://dx.doi.org/10.1097/QAI.0000000000000183 Stoltenberg, M., van Ijzendoorn, M. H., Euser, E. M., & Bakermans-Kranenburg, M. J. (2011). A global perspective on child sexual abuse: Meta-analysis of prevalence around the world. Child Maltreatment, 16, 79–101. http://dx.doi.org/10.1177/1077559511403920 Together for Girls. (2014). Zimbabwe to combat gender-based violence with its first national girls and 2014 young women’s empowerment framework. Retrieved from http://www.togetherforgirls.org/safe/?p=464 Together for Girls. (2013). Linking violence against children surveys to coordinated and effective action. Factsheet. Retrieved from http://www.togetherforgirls.org/wp-content/uploads/Together-for-Girls-Process-Paper Linking-VACS-to-Coordinated-and-Effecti. . ..pdf Undie, C., Mullick, S., & Askew, I. (2013). The missing ‘C’: Sexual violence against children in sub-Saharan Africa. Retrieved from http://www.unicef-irc.org/research-watch/Violence-against-children–a-silent-threat/987/ UNICEF. (2015). The state of the world’s children report. New York, NY: UNICEF. UNICEF. (2014). Hidden in plain sight: A statistical analysis of violence against children. New York, NY: UNICEF. UNICEF, CDC, & Muhumbili University of Health and Allied Sciences. (2011). Violence against Children in Tanzania: Findings from a national survey 2009. Dar es Salaam, Tanzania: UNICEF. UNICEF, CDC, & the Kenya National Bureau of Statistics. (2012). Violence against children in Kenya: Findings from a 2010 national survey. Nairobi, Kenya: UNICEF. United Nations General Assembly. (2000). Resolution adopted by the General Assembly A/RES/55/2: United Nations millennium declaration. New York, NY: United Nations. Retrieved from http://www.un.org/millennium/declaration/ares552e.pdf Veenema, T., Thornton, C., & Corley, A. (2014). The public health crisis of child sexual abuse in low and middle-income countries: An integrative review of the literature. International Journal of Nursing Studies, 52, 864–881. http://dx.doi.org/10.1016/j.ijnurstu.2014.10.017 Watts, C., & Zimmerman, C. (2002). Violence against women: Global scope and magnitude. Lancet, 359, 1232–1237. http://dx.doi.org/10.1016/S0140-6736(02)08221-1 World Health Organization (WHO), United Nations Office on Drugs and Crime (UNODC), & United Nations Development Programme (UNDP). (2014). Global status report on violence prevention 2014. Geneva, Switzerland: WHO. Retrieved from http://www.who.int/violence injury prevention/violence/status report/2014/en/ World Health Organization. (2000). Behavioural surveillance surveys: Guidelines for repeated behavioural surveys in populations at risk of HIV. Geneva, Switzerland: WHO. Retrieved from http://www.who.int/hiv/strategic/pubbss/en/ World Health Organization. (2001). Putting women first: Ethical and safety recommendations for research on domestic violence against women. Geneva, Switzerland: WHO. Retrieved from http://whqlibdoc.who.int/hq/2001/WHO FCH GWH 01.1.pdf?ua=1 World Health Organization, & UNAIDS. (2013). 16 ideas for addressing violence against women in the context of the HIV epidemic: A programming tool. Geneva, Switzerland: WHO. ZIMSTAT, & ICF International. (2012). Zimbabwe demographic and health survey 2010-11 (ZDHS). Calverton, MD: IMSTAT and ICF International Inc. ZIMSTAT, UNICEF, & CCORE. (2013). National baseline survey on life experiences of adolescents 2011. Harare, Zimbabwe: ZIMSTAT.
Please cite this article in press as: Rumble, L., et al. Childhood sexual violence in Zimbabwe: Evidence for the epidemic against girls. Child Abuse & Neglect (2015), http://dx.doi.org/10.1016/j.chiabu.2015.04.015
