Abstract
Sexual violence against children is a significant global public health problem, yet limited studies exist from low-resource settings. In Haiti we conducted the country’s first, nationally representative survey focused on childhood violence to help inform the development of a national action plan for violence against children. The Haiti Violence Against Children Survey was a household-level, multistage, cluster survey among youth age 13–24. In this analysis we sought to determine whether sexual violence sentinel events (unwanted sexual touching or unwanted attempted sex) were predictive of later unwanted, completed, penetrative sex in Haiti. We also sought to explore characteristics of sentinel events and help-seeking behavior among Haitian children. Multivariable logistic regression was used to test associations between sentinel events and later unwanted, completed, penetrative sex. Overall, 1,457 females reported on experiences of sexual violence occurring in childhood (before age 18). A sentinel event occurred in 40.4% of females who experienced subsequent unwanted completed sex. Females experiencing a sentinel event were approximately two and a half times more likely to experience later unwanted completed sex (adjusted odds ratio = 2.40, p = .004) compared to individuals who did not experience a sentinel event. The mean lag time from first sentinel event to first unwanted completed sex was 2.3 years. Only half (54.6%) of children experiencing a sentinel event told someone about their experience of sexual violence. Among children, sentinel events occur frequently before later acts of completed unwanted sex and may represent a useful point of intervention. Reporting of sexual violence by children in Haiti is low and can be improved to better act on sentinel events.
Keywords: Child abuse, Sexual violence, Rape, Haiti, Sentinel event
Introduction
Global Sexual Violence Against Girls
Sexual violence against girls, defined as females younger than age 18, is a substantial global public health problem. A growing number of single and multi-country studies have identified markedly elevated rates of unwanted sexual intercourse experienced by children (Reza et al., 2009; Stoltenborgh, van Ijzendoorn, Euser, & Bakermans-Kranenburg, 2011). For example, a World Health Organization school-based study among over 20,000 students from five African countries revealed that 21.5% of female children younger than age 14 had been physically forced to have sex (Brown et al., 2009). In another survey of students in 10 African nations, 19.6% of female students aged 11–16 had experienced unwanted sexual intercourse (Andersson et al., 2012). Investigations into sexual violence among children in other regions of the world have also shown elevated rates of victimization; a recent meta-analysis estimated a prevalence of childhood sexual abuse among females in Asia and South America of 11.3% and 13.4%, respectively (Stoltenborgh et al., 2011). In spite of a growing awareness of the burden of sexual violence against children in many nations, local data to guide preventive strategies is lacking in many countries.
Sexual Violence in Haiti
In Haiti, previous research studies have suggested an elevated burden of sexual violence; although, in general, the research has only peripherally focused on childhood violence. In one survey in Haiti’s capital city of Port-au-Prince, an estimated 35,000 sexual assaults among females occurred over a 22 month period; over half of these (19,000) occurred in children less than 18 years of age, indicating a disproportionate burden of sexual violence among youth (Kolbe & Hutson, 2006). Research from Haiti’s Demographic and Health Surveys (DHS) has also helped to elucidate the burden of intimate partner violence and provided some insight into the experiences of children as well (Gage, 2005; Kishor, 2012; Kishor & Johnson, 2006). In the most recent Haiti DHS survey in 2012, 29% of ever-married women reported experiencing some form of violence (emotional, physical or sexual) by their current or most recent partner (Ministry of Public Health and Population, Haitian Childhood Institute, and ICF International, 2013). In the 2012 Haiti DHS, intimate partner violence was highest among women ages 15–19 (43%), suggesting an elevated burden of such violence among youth. National level DHS data from Haiti has also detected an association between female partner violence and measures of the frequency of men’s physical abuse of children at the community level (Gage, 2005). Nonetheless, previous studies in Haiti have been significantly limited in that they have gathered only limited information on violence in childhood or have focused on smaller geographic regions (Gage, 2005; Kolbe & Hutson, 2006; Kolbe et al., 2010). A nationally representative, detailed investigation of violence against children in Haiti was needed to guide preventive strategies.
Haitian Violence Against Children Survey
To help guide a national action plan on violence against children, the Haitian government invited a broad international collaboration to conduct a national-level survey called the Violence Against Children Survey (VACS). The partnership consisted of a Haitian government led Multi-Sectoral Task Force, the Haitian Interuniversity Institute for Research and Development (INURED), the United Nations Children’s Fund (UNICEF), the Together for Girls Partnership, the President’s Emergency Plan for AIDS Relief (PEPFAR), and the U.S. Centers for Disease Control and Prevention (CDC). A major aim of this partnership was to identify potential avenues for prevention of sexual violence among children given the multiple, long-lasting adverse health effects of sexual violence and related adverse experiences, including sexually transmitted infections, depression, substance abuse, and other sequalae (Felitti et al., 1998; Gómez, Speizer, & Beauvais, 2009; Jewkes, Dunkle, Nduna, Jama, & Puren, 2010; Kishor, 2012).
Preventing childhood sexual violence depends first on accurately identifying those at high risk of abuse. This analysis investigates sentinel events in a child’s life—defined as episodes of unwanted sexual touching or attempted (but not completed) sex—and their ability to predict later completed, unwanted, penetrative sex. We hypothesized that sentinel events would be common prior to completed, unwanted sex and could be a point of intervention. Characteristics, timing, and help-seeking behavior surrounding sentinel events are also explored to help guide efforts to prevent and respond to violence in Haiti.
Methods
Study Design, Sampling, and Participants
From April to June, 2012, INURED with local partners and technical assistance from CDC conducted the Violence Against Children Survey, a national, household survey among Haitian youth age 13–24 years. A multistage, cluster sampling strategy was employed in consultation with host-nation statistical and technical agencies.
Because the study took place after the 2010 Haitian earthquake, which resulted in large-scale population displacement, the study sampling strategy needed to account for individuals displaced from their homes and living in internally displaced person (IDP) camps. IDP camps are temporary settlements which house individuals forced to relocate because of disasters or other events; IDP camps can vary widely but are often tented settlements supported by governmental, international, or non-governmental aid agencies. In the Haiti VACS survey, the sample was stratified initially into IDP camps and non-camp areas. The non-camp area was further stratified based on 11 geographic units known as ‘domaines,’ which represent Haiti’s major provinces as well as the capital city. For the non-camp sample, 177 enumeration areas (EA) were then selected via probability proportional to size (PPS) sampling. For the IDP camp sample, 11 camps were chosen as EAs using PPS sampling. Of these, 79 non-camp EAs and 5 camp EAs were randomly selected for female surveys. A cluster of 35 households was randomly selected from each EA. Subsequently, one female respondent age 13–24 years per household was then selected via the Kish method (Kish, 1949). Further details of sampling methods are available in the Haiti country report (US Centers for Disease Control and Prevention, 2014).
Procedures and Definitions
Participants were asked about experiences with four major kinds of sexual violence—unwanted sexual touching, unwanted attempted sex, pressured penetrative sex, and forced penetrative sex. This study focuses only on events of sexual violence occurring in childhood, before the age of 18. Unwanted sexual touching was defined as being fondled, pinched, grabbed, or touched against one’s will on the genitalia, anus, groin, breast, inner thigh, or buttocks. This could occur either directly or through clothing. ‘Sex’ was defined as someone penetrating the victim’s vagina or anus with their penis, hands, fingers, mouth, or an object, or penetrating the victim’s mouth with their penis. Unwanted attempted sex was defined as someone attempting to try to make the victim have sex without permission but not succeeding. Pressured sex constituted being coerced in a non-physical manner to have unwanted sex. Conversely, forced sex involved being physically forced to have sex. If a participant had experienced either pressured or forced sex, they were coded as having had completed unwanted sex.
All measures of sexual violence were created for the Violence Against Children Survey though the survey development process was informed by classifications and wording from a diverse array of surveys including the Longitudinal Studies of Child Abuse and Neglect, the National Longitudinal Study of Adolescent Health, the Child Sexual Assault Survey, the National Intimate Partner and Sexual Violence System, the Global School-based Health Survey, and others. VACS survey questions and wording have been independently tested and evaluated using cognitive testing methods by the National Center for Health Statistics. Furthermore, in Haiti, a local committee of governmental, non-governmental, and academic partners oversaw the survey; committee members reviewed survey translations and phrasing to ensure validity in the Haitian context and appropriate translation to Creole, the local language in Haiti.
In this study, sentinel events involving non-completed sex were explored for their ability to predict later completed sex. Sentinel events were defined as either an episode of unwanted touching or attempted sex that occurred prior to the first occurrence of unwanted completed sex. Participants reported sexual experiences by age in one year increments. Consequently, a sentinel event had to occur in a year preceding an episode of completed sex. Sentinel events that occurred during the same year as an episode of completed sex were coded as missing and not included given the inability to determine which event occurred first. This provided the most conservative test of the ability of sentinel events to predict completed unwanted sex. In addition to reporting on measures of sexual violence and sentinel events, participants were asked about the characteristics and circumstances of each episode of sexual violence. Survey respondents also were asked whether they told someone about the violence, whom they told, and any reasons for not telling someone.
Surveys were administered through face-to-face interviews that were conducted by survey workers from host-country technical agencies. CDC technical experts provided training and support for survey implementation. Only female interviewers interviewed female respondents. Participants provided informed verbal consent. Ethical approval was provided by the CDC’s Institutional Review Board (IRB) as well as the local Haitian IRB committees of the Haitian Ministry of Public Health and Population and INURED. Enumerators linked respondents desiring care to a local social service agency and all respondents were provided a list of service referrals.
Statistical Analysis
For the main analysis, the proportion of individuals who had a sentinel event was calculated, both for individuals who later experienced completed unwanted sex as well as for individuals who did not. All percentages take into account the complex survey design and are nationally representative estimates. The association between experiencing a sentinel event and later experiencing unwanted completed sex was also explored through logistic regression. Statistical significance was regarded as p < .05. Additionally, to provide context for possible prevention strategies, among those individuals who later experienced completed unwanted sex the circumstances of the first sentinel event were compared to the circumstances of the first episode of unwanted sex. Lastly, to assess barriers to targeting those individuals with sentinel events, we calculated estimates on whether individuals who experienced sexual violence told someone about their abuse as well as reasons for not disclosing. All analyses were performed in SAS (version 9.3).
Results
After consent, 1,457 females completed the VACS questionnaire for an overall response rate of 85.6%. A sentinel event occurred in 21.3% of individuals who did not report subsequent unwanted completed sex and in 40.4% who did experience subsequent unwanted completed sex (Table 1). Among those individuals who experienced later unwanted sex, 27.7% experienced at least one episode of antecedent unwanted sexual touching. Among this same population, 26.3% experienced at least one episode of unwanted attempt sex.
Table 1.
Presence and characteristics of sentinel events among youth age 13–24 by status of subsequently having experienced unwanted completed sex in childhood (N = 1375).a
| Have not experienced unwanted completed sex (N = 1286)
|
Experienced unwanted completed sex (N = 89)
|
|||
|---|---|---|---|---|
| n | % (95% CI) | n | % (95% CI) | |
| Experienced any sentinel event (sexual touching or attempted sex) | 272 | 21.3% (18.3, 24.2) | 44 | 40.4% (24.9, 55.9) |
| Experienced sexual touching | 209 | 16.5% (13.4, 19.6) | 30 | 27.7% (15.3, 40.2) |
| Experienced attempted sex | 150 | 11.7% (9.3, 14.1) | 26 | 26.3% (14.1, 38.5) |
| Have not experienced unwanted completed sex (N = 1286)
|
Experienced unwanted completed sex (N = 89)
|
|||
| n | Mean (95% CI) | n | Mean (95% CI) | |
| Mean age at first sentinel event, years | – | 14.2 (13.9, 14.5) | – | 13.2 (12.4, 14.0) |
| Mean age at first episode of unwanted sex, yearsb | – | n/a | – | 15.5 (14.7, 16.2) |
| Mean time from first sentinel event to first unwanted sex, years | – | n/a | – | 2.3 (1.6, 2.9) |
Note. CI = confidence interval; percents represent weighted percentages.
Sentinel event defined as an episode of sexual touching or attempted sex occurring before an episode of unwanted completed sexual intercourse, all events in childhood before the age of 18. Sample includes those individuals who have complete information on sentinel events and unwanted sex. Those individuals experiencing a potential sentinel event in the same year as unwanted completed sex were coded as missing and not analyzed given the inability to determine which was the initial event.
Among those children having experienced a sentinel event.
Among those females who experienced unwanted completed sex (which occurred at a mean age of 15.5 years), the first sentinel event occurred at a mean age of 13.2 years, corresponding to an average lag time of 2.3 years from first sentinel event to first unwanted completed sex. Of the 44 initial sentinel events occurring among those individuals experiencing later unwanted completed sex, 18 occurred approximately one year prior. Ten episodes occurred 2 years prior, 7 episodes 3 years prior, 5 episodes 4 years prior, and only 4 episodes occurred 5 or more years prior.
Logistic regression analyses revealed that experiencing a sentinel event was strongly associated with later experiencing unwanted completed sex (Table 2). Even after adjustment for demographic and social characteristics, individuals experiencing any sentinel event were approximately two and a half times more likely to later experience completed sex (OR = 2.40, 95% CI [1.33, 4.33], p = .004) compared to individuals who did not experience a sentinel event. When investigating unwanted touching and attempted sex as sentinel events independently, both trended in the direction of an association with later completed sex. However, the association between attempted sex and later completed sex (OR = 2.21, 95% CI [1.32, 3.69], p = .003) was significant while the association between unwanted touching and completed sex was not (OR = 1.51, 95% CI [0.84, 2.73], p = .172).
Table 2.
Association of experiencing a sentinel event with later unwanted completed sex in childhood (N = 1375).
| Unadjusted
|
Adjusteda
|
|||
|---|---|---|---|---|
| OR (95% CI) | p | OR (95% CI) | p | |
| Experienced any sentinel event (sexual touching or attempted sex) | 2.51 (1.34, 4.69) | .004 | 2.40 (1.33, 4.33) | .004 |
| Experienced sexual touching as a sentinel event | 1.59 (0.87, 2.90) | .133 | 1.51 (0.84, 2.73) | .172 |
| Experienced attempted sex as a sentinel event | 2.30 (1.35, 3.93) | .002 | 2.21 (1.32, 3.69) | .003 |
Note. CI = confidence interval; OR = odds ratio.
Adjusted models control for age, primary school completion, having worked as a domestic servant, orphanhood, and presence of supportive friendships.
Among those victims of unwanted sex having a sentinel event, circumstances of the first sentinel event were compared to circumstances of the first event of unwanted completed sex (Table 3). Perpetrators of sentinel events were most often a friend/acquaintance/neighbor (56.6%) whereas perpetrators of completed sex were most often romantic partners (46.4%). For both sentinel events and completed sex episodes, over 50% of events were estimated to occur at either the victim’s or perpetrator’s home. Similarly, perpetrators were almost exclusively all older than victims, both among sentinel events and completed sex episodes.
Table 3.
Circumstances of first reported sentinel events compared to first event of unwanted completed sex among children experiencing any sentinel event (N = 44).
| First sentinel event
|
First completed unwanted sex
|
|||
|---|---|---|---|---|
| n | % (95% CI) | n | % (95% CI) | |
| More than one perpetrator involved | 12 | 25.4 (9.9, 40.8)a | 1 | 1.7 (0.0, 5.2)a |
| Perpetrator | ||||
| Romantic partner | 6 | 20.1 (5.4, 34.9)a | 17 | 46.4 (29.3, 63.6) |
| Family member/relative | 4 | 6.0 (0.0, 14.4)a | 2 | 5.2 (0.0, 13.6)a |
| Friend/acquaintance, neighbor | 26 | 56.6 (38.6, 74.7) | 17 | 29.1 (14.1, 44.2) |
| Stranger/other | 8 | 17.2 (0.0, 35.3)a | 8 | 19.2 (0.4, 38.1)a |
| Location | ||||
| Victim’s home | 17 | 42.1 (22.0, 62.2) | 15 | 29.1 (10.0, 48.3)a |
| Perpetrator’s home | 11 | 17.1 (5.3, 28.9)a | 12 | 22.8 (6.2, 39.5)a |
| Other residence/business | 3 | 5.6 (0.0, 14.0)a | 9 | 35.3 (12.3, 58.3)a |
| Outdoors | 13 | 35.2 (14.9, 55.5) | 8 | 12.7 (1.7, 23.7)a |
| Age of perpetrator | ||||
| Younger | 0 | 0 | 0 | 0 |
| Same | 0 | 0 | 3 | 6.7 (0.0, 15.9)a |
| Older (by less than 5 years) | 9 | 20.5 (6.1, 35.0)a | 9 | 21.1 (6.1, 36.1)a |
| Older (by 5–10 years) | 20 | 49.0 (27.6, 70.4) | 20 | 55.4 (33.7, 77.1) |
| Older (by more than 10 years) | 11 | 30.5 (7.7, 53.3)a | 7 | 16.8 (0.0, 36.2)a |
Note. CI = confidence interval; percents represent weighted percentages.
Relative standard error >30%, estimates based on small numbers should be interpreted with caution.
Of the entire population of individuals experiencing a sentinel event, over half told at least one person about any episode of sexual violence they experienced (54.6%), with 25.9% telling a family member/relative and 26.2% telling a friend (Table 4). Only a small number of children who experienced a sentinel event reported telling a romantic partner, a professional such as a health care worker, counselor, teacher, or police officer, or any other person type. The most common reasons for not telling someone about an experience were not thinking the episode was a problem (31.1%), being embarrassed (30.4%), and being afraid of getting in trouble (19.1%).
Table 4.
Help seeking among children who have experienced a sentinel event.
| Total population of children experiencing a sentinel event (N = 316)a
|
Children who experienced a sentinel event and later unwanted completed sex (N = 44)a
|
Children who experienced a sentinel event without later, unwanted completed sex (N = 272)a
|
|||||
|---|---|---|---|---|---|---|---|
| n | % (95% CI) | n | % (95% CI) | n | % (95% CI) | ||
| Who Victim Told About Abuse | Anyone (total) | 165 | 54.6 (45.8, 63.3) | 20 | 47.6 (25.6, 69.6) | 145 | 55.5 (46.9, 64.0) |
| Family member/relative | 77 | 25.9 (19.7, 32.1) | 8 | 18.9 (0.3, 37.5)b | 69 | 26.8 (20.7, 33.0) | |
| Romantic partner | 5 | 2.1 (0.0, 4.5)b | 0 | 0 | 5 | 2.3 (0.0, 5.1)b | |
| Friend | 86 | 26.2 (19.2, 33.3) | 13 | 29.7 (9.3, 50.2)b | 73 | 25.8 (18.4, 33.2) | |
| Professional | 2 | 0.4 (0.0, 1.0)b | 1 | 0.5 (0.0, 1.5)b | 1 | 0.4 (0.0, 1.1)b | |
| Other | 7 | 3.7 (0.8, 6.5)b | 2 | 12.0 (0.0, 30.2)b | 5 | 2.6 (0.3, 4.8)b | |
| Reasons for not telling anyone about an experience* | Afraid of getting into trouble | 42 | 19.1 (12.5, 25.7) | 13 | 36.2 (16.4, 55.9) | 29 | 16.1 (8.6, 23.6) |
| Embarrassed | 50 | 30.4 (21.4, 39.4) | 8 | 25.0 (8.2, 41.8)b | 42 | 31.3 (21.0, 41.6) | |
| Dependents on perpetrator | 1 | 0.6 (0.0, 1.9)b | 0 | 0 | 1 | 0.8 (0.0, 2.3)b | |
| Threatened by perpetrator | 7 | 2.3 (0.2, 4.5)b | 2 | 4.9 (0.0, 11.9)b | 5 | 1.9 (0.0, 4.1)b | |
| Didn’t think it was a problem | 49 | 31.1 (20.9, 41.3) | 5 | 17.6 (0.0, 38.4)b | 44 | 33.5 (21.6, 45.4) | |
| Felt it was my fault | 8 | 2.4 (0.4, 4.4)b | 1 | 1.6 (0.0, 4.9)b | 7 | 2.6 (0.3, 4.9)b | |
| Didn’t want abuser to get in trouble | 2 | 2.0 (0.0, 5.2)b | 0 | 0 | 2 | 2.3 (0.0, 6.1)b | |
| Afraid of being abandoned | 12 | 13.1 (3.2, 23.1)b | 3 | 13.0 (0.0, 30.8)b | 9 | 13.2 (2.0, 24.3)b | |
| Had no one to confide in | 5 | 2.7 (0.3, 5.0)b | 0 | 0 | 5 | 3.1 (0.4, 5.9)b | |
| Other | 4 | 1.9 (0.0, 3.8)b | 1 | 4.4 (0.0, 12.9)b | 3 | 1.4 (0.0, 3.1)b | |
Note. CI = confidence interval; percents represent weighted percentages.
For section on reported reasons for not telling anyone about an experience, the total N for each column is 185, 32, and 153, respectively.
Relative standard error >30%, estimates based on small numbers should be interpreted with caution.
Respondent could provide more than one response to questions regarding who victim told about sexual abuse as well as reasons for not telling about abuse.
Discussion
To the best of our knowledge this is the first, nationally representative survey focused on childhood violence in Haiti. This investigation provides locally relevant information to help inform violence prevention activities as Haiti develops its national action plan on violence against children. Sexual violence among children is challenging to prevent and there remains a paucity of studies to guide prevention efforts in developing nations. One, initial critical step in violence prevention is to accurately identify high-risk individuals so that prevention programs can be effectively deployed. Our findings from Haiti suggest that sentinel events are highly predictive of later unwanted completed sex and may be a useful way to target high-risk females. Indeed, over 40% of girls who experienced unwanted completed sex had also experienced a sentinel event. On average, the first sentinel event occurred approximately two years prior to the first episode of unwanted completed sex, a feasible time frame for effective intervention to occur. Although other risk factors for sexual violence have been detected in developing nations, such as certain measures of socioeconomic status, family structure, or gender norms, some of these factors—such as poverty or household composition—are difficult to target because they apply to a wide range of the population and are not sufficiently discriminating (Hayati, Hogberg, Hakimi, Ellsberg, & Emmelin, 2011; King et al., 2004; Madu & Peltzer, 2000). In contrast, sentinel events represent discrete experiences that could be easier to screen for and serve as high-yield markers for subsequent abuse. In fact, recent research from the United States in the field of childhood physical abuse has identified that sentinel events in the form of mild injuries detected by clinicians are highly predictive of later, diagnosable, physical abuse among children (Sheets et al., 2013).
The mechanism by which sentinel events among Haitian females predict later unwanted completed sex is not fully understood. For example, sentinel events may indicate the presence of a high-risk perpetrator in the victim’s life. Additionally, the occurrence of sentinel events may identify vulnerabilities or characteristics of a victim or an environment that predispose to exposure to high-risk or violent situations. Our analysis of characteristics of the first sentinel event and the first episode of unwanted completed sex provide novel information for an understanding of the context of such violence in Haiti. Sentinel events were more often committed by multiple perpetrators compared to completed sex. Reasons for this are unclear though some sentinel events, particularly sexual touching, may be viewed by perpetrators as more appropriate to perform as part of a group. We also noted that sentinel events were most often perpetrated by friends or neighbors, whereas perpetrators of completed sex were most often romantic partners. This may reflect the changing relationships present in a victim’s life as they age. However, it also suggests that vulnerabilities identified in one context may predict vulnerabilities in another, such as power differentials in relationships (Jewkes, Dunkle, Nduna, & Shai, 2010).
In addition to quantifying the ability of sentinel events to identify high-risk children in Haiti, it is necessary to understand child reporting and help-seeking after a sentinel event. Consequently, in this study we also assessed the degree to which youth reported their abuse to family, friends, and professionals. Only approximately half of victims of a sentinel event reported the violence to at least one person. Victims who told someone about sexual violence overwhelmingly told either a family member or friend. Victims rarely informed other segments of the populations, including professionals. This may occur because professional services are not readily available, children lack knowledge of service availability, or stigma exists to seeking professional services. More research is needed to understand service preferences, availability, and quality in Haiti. Nonetheless, the most frequently reported reasons for not telling someone about the abuse were not thinking the episode was a problem, being embarrassed, and being afraid of getting in trouble. Thus, attempts to increase reporting must consider local perceptions surrounding sexual violence and assist in the recognition of inappropriate sexual contact. Addressing norms may allow victims to feel confident in reporting experiences of sexual violence and may also aid in reducing perpetration—bystander training programs, small group programs that address gender norms, and community mobilization campaigns have shown promising evidence for reducing violence in international settings and can be considered in Haiti (Abramsky et al., 2014; Coker et al., 2014; Hillis et al., 2015; Jewkes et al., 2008).
Some notable limitations of this investigation should be mentioned. First, victims reported abuse by age in one-year increments. Thus, when a participant indicated both a potential sentinel event in the same year as an episode of unwanted completed sex it was impossible to determine which event occurred first. Future surveys which address this temporality can further elucidate patterns of sentinel events occurring in very close proximity to unwanted sex events. Further, it should be noted that information from victims was collected via self-report and is subject to limitations from recall bias and potential hesitancy to disclose sensitive information. Sociocultural norms regarding children, sexual violence, and sexuality also may affect perceptions of what is considered to be sexual violence. Also, the study was conducted after the Haitian earthquake, where large scale population displacement increased vulnerabilities (Balsari, Lemery, Williams, & Nelson, 2010). The precise effect of this event on trends of sexual violence in Haiti is not known but may have increased victimization risk. Lastly, this research study focuses on prevention strategies aimed at victimization; understanding and addressing perpetration is also essential.
Despite these limitations, this study provides novel information for the country of Haiti in formulating preventive strategies for childhood sexual violence. This analysis identifies sentinel events as a potentially important strategy for targeting high-risk children, describes the characteristics and patterns of sentinel events to better understand the local context of such violence, and explores barriers for disclosure of sentinel events. Linking children to prevention or response programs by identifying high risk populations and improving disclosure could aid in the reduction of violence against children in Haiti.
Footnotes
This study was funded by The President’s Emergency Plan For AIDS Relief (PEPFAR).
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.
Conflict of interest
We declare no competing interests.
References
- Abramsky T, Devries K, Kiss L, Nakuti J, Kyegombe N, Starmann E, Cundill B, Francisco L, Kaye D, Musuya T, Michau L, Watts C. Findings from the SASA! Study: A cluster randomized controlled trial to assess the impact of a community mobilization intervention to prevent violence against women and reduce HIV risk in Kampala, Uganda. BMC Medicine. 2014;12:122. doi: 10.1186/s12916-014-0122-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Andersson N, Paredes-Solis S, Milne D, Omer K, Marokoane N, Laetsang D, Cockcroft A. Prevalence and risk factors for forced or coerced sex among school-going youth: National cross-sectional studies in 10 southern African countries in 2003 and 2007. BMJ Open. 2012;2(2):e000754. doi: 10.1136/bmjopen-2011-000754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Balsari S, Lemery J, Williams TP, Nelson BD. Protecting the children of Haiti. New England Journal of Medicine. 2010;362(9):e25. doi: 10.1056/NEJMp1001820. [DOI] [PubMed] [Google Scholar]
- Brown DW, Riley L, Butchart A, Meddings DR, Kann L, Harvey AP. Exposure to physical and sexual violence and adverse health behaviours in African children: Results from the Global School-based Student Health Survey. Bulletin of the World Health Organization. 2009;87(6):447–455. doi: 10.2471/BLT.07.047423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Coker AL, Fisher BS, Bush HM, Swan SC, Williams CM, Clear ER, DeGue S. Evaluation of the green dot bystander intervention to reduce interpersonal violence among college students across three campuses. Violence Against Women. 2014 doi: 10.1177/1077801214545284. Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, Koss MP, Marks JS. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. American Journal of Preventive Medicine. 1998;14(4):245–258. doi: 10.1016/s0749-3797(98)00017-8. [DOI] [PubMed] [Google Scholar]
- Gage AJ. Women’s experience of intimate partner violence in Haiti. Social Science & Medicine. 2005;61(2):343–364. doi: 10.1016/j.socscimed.2004.11.078. [DOI] [PubMed] [Google Scholar]
- Gómez AM, Speizer IS, Beauvais H. Sexual violence and reproductive health among youth in Port-au-Prince, Haiti. Journal of Adolescent Health. 2009;44(5):508–510. doi: 10.1016/j.jadohealth.2008.09.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hayati EN, Hogberg U, Hakimi M, Ellsberg MC, Emmelin M. Behind the silence of harmony: Risk factors for physical and sexual violence among women in rural Indonesia. BMC Womens Health. 2011;11:52. doi: 10.1186/1472-6874-11-52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hillis S, Mercy J, Saul J, Gleckel J, Abad N, Kress H. THRIVES: A global technical package to prevent violence against children. Atlanta, GA: Centers for Disease Control and Prevention; 2015. [Google Scholar]
- Jewkes RK, Dunkle K, Nduna M, Jama PN, Puren A. Associations between childhood adversity and depression, substance abuse and HIV and HSV2 incident infections in rural South African youth. Child Abuse and Neglect. 2010;34(11):833–841. doi: 10.1016/j.chiabu.2010.05.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jewkes RK, Dunkle K, Nduna M, Shai N. Intimate partner violence, relationship power inequity, and incidence of HIV infection in young women in South Africa: A cohort study. Lancet. 2010;376(9734):41–48. doi: 10.1016/S0140-6736(10)60548-X. [DOI] [PubMed] [Google Scholar]
- Jewkes R, Nduna M, Levin J, Jama N, Dunkle K, Puren A, Duvvury N. Impact of stepping stones on incidence of HIV and HSV-2 and sexual behaviour in rural South Africa: Cluster randomised controlled trial. BMJ. 2008;337:a506. doi: 10.1136/bmj.a506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- King G, Flisher AJ, Noubary F, Reece R, Marais A, Lombard C. Substance abuse and behavioral correlates of sexual assault among South African adolescents. Child Abuse and Neglect. 2004;28(6):683–696. doi: 10.1016/j.chiabu.2003.12.003. [DOI] [PubMed] [Google Scholar]
- Kish L. A procedure for objective respondent selection within the household. Journal of the American Statistical Association. 1949;44(247):380–387. [Google Scholar]
- Kishor S. Married women’s risk of stis in developing countries the role of intimate partner violence and partner’s infection status. Violence Against Women. 2012;18(7):829–853. doi: 10.1177/1077801212455358. [DOI] [PubMed] [Google Scholar]
- Kishor S, Johnson K. Reproductive health and domestic violence: Are the poorest women uniquely disadvantaged? Demography. 2006;43(2):293–307. doi: 10.1353/dem.2006.0014. [DOI] [PubMed] [Google Scholar]
- Kolbe AR, Hutson RA. Human rights abuse and other criminal violations in Port-au-Prince, Haiti: A random survey of households. Lancet. 2006;368(9538):864–873. doi: 10.1016/S0140-6736(06)69211-8. [DOI] [PubMed] [Google Scholar]
- Kolbe AR, Hutson RA, Shannon H, Trzcinski E, Miles B, Levitz N, Puccio M, James L, Noel JR, Muggah R. Mortality, crime and access to basic needs before and after the Haiti earthquake: A random survey of Port-au-Prince households. Medicine, Conflict and Survival. 2010;26(4):281–297. doi: 10.1080/13623699.2010.535279. [DOI] [PubMed] [Google Scholar]
- Madu SN, Peltzer K. Risk factors and child sexual abuse among secondary school students in the Northern Province (South Africa) Child Abuse and Neglect. 2000;24(2):259–268. doi: 10.1016/s0145-2134(99)00128-3. [DOI] [PubMed] [Google Scholar]
- Ministry of Public Health and Population, Haitian Childhood Institute, & ICF International. 2012 Haïti mortality, morbidity, and service utilization survey: Key findings. Calverton, MD: 2013. [Google Scholar]
- Reza A, Breiding MJ, Gulaid J, Mercy JA, Blanton C, Mthethwa Z, Bamrah S, Dahlberg LL, Anderson M. Sexual violence and its health consequences for female children in Swaziland: A cluster survey study. Lancet. 2009;373(9679):1966–1972. doi: 10.1016/S0140-6736(09)60247-6. [DOI] [PubMed] [Google Scholar]
- Sheets LK, Leach ME, Koszewski IJ, Lessmeier AM, Nugent M, Simpson P. Sentinel injuries in infants evaluated for child physical abuse. Pediatrics. 2013;131(4):701–707. doi: 10.1542/peds.2012-2780. [DOI] [PubMed] [Google Scholar]
- Stoltenborgh M, van Ijzendoorn MH, Euser EM, Bakermans-Kranenburg MJ. A global perspective on child sexual abuse: Meta-analysis of prevalence around the world. Child Maltreatment. 2011;16(2):79–101. doi: 10.1177/1077559511403920. [DOI] [PubMed] [Google Scholar]
- U.S. Centers for Disease Control and Prevention, Interuniversity Institute for Research and Development, & Comité de Coordination. Violence against children in Haiti: Findings from a national survey. Haiti: Port-au-Prince; 2014. [Google Scholar]