Abstract
Childhood exposure to violence can lead to physical, mental, and emotional harm, whether a child is a direct victim or a witness to violent events. The aim of this study is to examine the relationship between witnessing intimate partner violence (IPV) and victimization among children. A cross-sectional, national study was conducted in secondary high schools in the five main provinces of Saudi Arabia (SA) using International Society for Prevention of Child Abuse and Neglect (ISPCAN) ISPCAN Child Abuse and Neglect Screening Tool–Child: Home version. Boys and girls, public and private schools were selected to participate. Students (N = 16,939) aged 15 to 18 years completed the survey instrument which included demographics, different types of abuse (physical, psychological, and sexual), neglect, and witnessing IPV. Mean age of the participants was 16.8 ± 0.9 years, and 51% were boys. Eighty-one percent lived with both parents, 6% with single parent, and 2% with step-parent. Fifty-two percent of the participants witnessed IPV. Those who witnessed IPV were more likely to be abused compared with those who did not (p < .01). Among those who witnessed physical IPV, girls had a significantly greater likelihood of experiencing psychological abuse (odds ratio [OR] = 3.7, confidence interval [CI] = [1.9, 6.8]), physical abuse (OR = 1.3, CI = [1.0, 1.6]), and neglect (OR = 1.6, CI = [1.4, 1.9]) but less likelihood of experiencing sexual abuse (OR = 0.6, CI = [0.5, 0.7]) than boys. Among those who witnessed psychological IPV, girls had a significantly greater likelihood of experiencing neglect (OR = 1.3, CI = [1.2, 1.5]) but less likelihood of experiencing sexual abuse (OR = 0.5, CI = [0.5, 0.6]) than boys. Boys who witnessed physical IPV and psychological IPV had a significantly greater likelihood of experiencing sexual abuse compared with girls. Witnessing IPV increases the chances of child and adolescent victimization. Multidisciplinary approaches involving social workers, law enforcement personnel, and domestic violence and child protection workers could effectively respond to this problem.
Intimate partner violence (IPV) is a widespread public health problem around the world and may involve a range of behaviors including verbal threats, physical aggression, and degrading sex (Jouriles, McDonald, Norwood, & Ezell, 2001; Koss et al., 1994). It is still a major problem not only to the involved adults but also to the children who are witnessing violence either directly involving (e.g., trying to intervene or calling the police) or experiencing the aftermath (e.g., seeing bruises or observing maternal depression) as they may be traumatized by the incidents and may also develop behavioral issues (Ganley & Schechter, 1996; Holden, Geffner, & Jouriles, 1998; Spilsbury et al., 2007). Children or adolescents might be affected by more than one type of violence (i.e., poly-victimization), especially in chronic and severe conditions where these situations are related (Lourenço et al., 2013).
Although it has been recognized that different types of violence may co-occur, research addressing the overlap between IPV and child maltreatment (CM) is a relatively recent phenomenon (Bedi & Goddard, 2007). McKibben, DeVos, and Newberger (1989) reported that 40% to 60% of mothers of abused children were abused by their partners, compared with 13% of mothers of unabused children. In a literature review examining 31 empirical studies, Appel and Holden (1998) reported that the rate of IPV and co-occurrence of CM in representative community sample ranged from 6% to 100%. They attributed the inconsistency to a number of methodological issues and reported a median of 40% of children who were exposed to IPV also experienced physical abuse. Edleson (1999) identified 25 studies examining families with IPV, and the co-occurrence of CM ranged from 6.5% to 97%. Most studies identified the mother as the adult victim of the IPV. These studies varied significantly on the type of CM although the majority reported some type of physical abuse.
In the Middle East, research shows that CM is common and underreported (Al-Mahroos & Al-Amer, 2011). Different patterns of aggression and violence in the Arab family significantly correlated with such variables as father’s unemployment, parents’ level of education, large family size, and living with single/step-parent (Almuneef, Al-Ghamdi, & Saleheen, 2016). Haj-Yahia and Ben-Arieh (2000) surveyed 1,640 Arab secondary high school students in Israel, of which 17% and 18% witnessed their fathers threatening to hit something and attacking their mothers, respectively, at least once during the past 12 months. At the same time, 39%, 40%, and 42% of the participants reported that their father, mother, and siblings, respectively, yelled at them and/or done something to insult them at least once. In another study in Israel, Haj-Yahia and Dawud-Noursi (1998) reported on 832 Arab adolescents between the ages of 16 and 18 years, where 56% witnessed their fathers and 32% witnessed their mothers argued heartedly with each other at least once during the past 12 months. The percentages of the participants whose fathers, mothers, and siblings threatened to hit or throw something to them at least once during the past 12 months were 28%, 29%, and 35%, respectively. These results highlight the importance of exploring IPV in the Arab family from a cultural perspective.
In SA, extensive research has been conducted on prevalence of CM, impact of adverse childhood experiences (ACEs) on chronic diseases, and health-risk behaviors among adults (Al-Eissa et al., 2016; Almuneef, Hollinshead, et al., 2016). Al-Eissa et al. (2016) conducted a population-based study with 16,939 secondary high school students and found that psychological abuse (65%) was most commonly disclosed followed by exposure to domestic and community violence (64%), childhood neglect (53%), physical abuse (50%), and sexual abuse (10%). In another study, Almuneef et al. (2016) reported that participants with four or more ACEs were more likely to have chronic diseases and health-risk behaviors. Although there are very few studies that have been conducted on the topic of IPV, none had examined the relationship between witnessing IPV and CM. Such research can play an important role to determine if the impacts of witnessing IPV on children in the Western countries are also present among Saudi children. In the present study, we hypothesized the following:
Method
Participants and Procedures
A cross-sectional study was conducted in secondary high schools in the main five out of the 13 provinces of SA (Riyadh Province: Central, Makkah Province: Western, Eastern Province: Eastern, Tabuk Province: Northern, and Jizan Province: Southern) during 2012 to study the epidemiology of CM. More than 75% of the population in SA resides in these provinces. Compared with available census data for 2013 (General Authority for Statistics, Kingdom of Saudi Arabia, 2015), our sample is a representative sample in terms of age and gender. Boys and girls, private and public schools were selected to participate. Enrollment criteria for this study included male and female students aged 15 to 18 years attending secondary high schools.
Sampling was done through multistage-stratified sampling technique. A list of secondary, public, and private high schools for boys and girls were accessed through the Ministry of Education. Based on student populations in each of the five provinces of the country, a weighted student sample size from each of the provinces was identified. Schools were randomly selected from class A (urban) and class B (rural) cities. Each of the major cities was further divided according to geographical boundaries to ensure that demographic differences in the cities were addressed. All students from each of the randomly selected schools were invited to participate in the study.
The study went through a multistage consent process. The institutional review board (IRB) of the King Abdullah International Medical Research Center (KAIMRC) in SA approved the study. Similarly, the Ministry of Education granted its approval for the study and requested each school principal’s approval. In addition, parental consent and participating students’ assent were obtained before data collection. All participants meeting the enrollment criteria were informed about the study. Information letters and parental consent forms were delivered by schools principals to the parents. The letter included an introduction to the study describing its objectives, methods, expected outcomes, and future impact of the results. A copy of the questionnaire was given only to the parent who requested it. The parental consent and refusal rate was 74% and 26%, respectively, without significant gender difference. There were no direct benefits or incentives provided to the participants. Considering that answering sensitive questions or recalling traumatic memories might cause psychological distress, participants were allowed to skip questions or withdraw from the study. There were no withdrawals or adverse events reported.
Research assistants received standardized training on recruitment skills, interviewing guidelines, and ethical and safety issues. After obtaining parental consent and participant’s assent, participants were provided self-administered International Society for Prevention of Child Abuse and Neglect (ISPCAN) Child Abuse and Neglect Screening Tool–Child: Home version (ICAST-CH, n.d.) questionnaires. The questionnaires were completed at the schools to avoid any parental/guardian influence.
In this study, data were composed of participants’ answers to quantitative surveys. Data were collected only for research purposes and were not used for other purposes. A number of steps were taken to ensure protection of privacy and confidentiality. All project staff received ongoing supervision in areas related to ethical conduct, privacy, and confidentiality protection, and how to deal with child’s disclosure of violence or any anticipated adverse event. The survey did not contain the name of the participant but instead was labeled with a reconstructable personal alphanumeric identifier. All participants’ information entered into the database were assigned a unique ID. All data were stored in a password protected database on computers in locked rooms at the National Family Safety Program (NFSP). The hard copy of this information was stored in a locked cabinet that does not contain other data. During data collection period, if students indicated their desire to disclose abusive events verbally to the researchers, they were encouraged to disclose these to the teachers and counselors they could trust. To help the students who anonymously shared their experience in writing on the questionnaire, Saudi Child Helpline (SCHL) pamphlets were distributed to all students and encouraged them to avail their services.
Measures
The ICAST-CH was developed by the experts of the ISPCAN, with support from UNICEF and the Oak Foundation. The reliability for ICAST-CH was assessed using Cronbach’s alpha. With the exception of the exposure to violence scale which has a fair alpha (α = .69), all other scales have alpha coefficients between .72 and .86 (good to very good; Zolotor et al., 2009). This questionnaire was translated into Arabic and back-translated (into English) for comparison. The SA version of the questionnaire was made available electronically for researchers since 2007 (www.ISPCAN.org/?page=ICAST). The tool was tested on a group of adolescents to ensure clear understanding and clarity of the answer choices. The translation of the tool was modified as a result of this focus group. A pilot cross-sectional study (n = 2,043) was conducted in Al-Kharj city in 2011-2012 utilizing ICAST-CH to estimate the incidence of CM and gender differences in each of five CM categories. It also attempted to assess the usability and acceptability of the Arabic version of the study tool for a future national surveillance study.
To assess 12 months prevalence rate, five domains that represent ICAST-CH in SA were used: neglect (six items), psychological abuse (eight items), physical abuse (10 items), sexual abuse (six items), and exposure to violence (six items). Demographics of the tool include age, gender, type, and level of school. Living arrangements include live with nuclear or extended family. The questionnaire consists of multiple choice responses—“many times,” “sometimes,” “never” in the past year, and “more than a year ago.” To have positive response, “many times” and “sometimes” were combined to determine 12 months prevalence. The responses to “more than a year ago” were excluded to avoid any recall bias.
Analysis
The first step of the analysis was descriptive analysis. Participants were described regarding their selected sociodemographic status (age, gender, and living arrangement). Odds ratio and corresponding 95% confidence interval (CI) were calculated for each of the potential explanatory variables in relation to the outcome. A significance level of .05 was used for all statistical tests. All data were analyzed using SPSS version 20.0. (SPSS Inc., 2013).
Results
A total of 16,939 students participated in the study. Subgroup analysis was conducted with those (n = 16,010) who answered “exposure to violence” questions. Mean age of the participants was 16.8 ± 0.9 years, 51% were boys, and 36% were from Riyadh province. Eighty-three percent of the participants lived with nuclear family, 81% with both parents, 6% with single parent, and 2% with step-parent. Fifty-two percent witnessed IPV (Table 1).
Demographic Characteristics of the Participants (N = 16,010).
Note. Percentages may not add to 100 due to missing data. IPV = intimate partner violence.
The relationship between witnessing IPV and types of CM is shown in Figure 1. In general, psychological abuse was the most common form of abuse accounting for 79% which was followed by physical abuse (63%), neglect (53%), and sexual abuse (13%). Participants who reported witnessing IPV were more likely to be psychologically abused (97% vs. 62%, p < .01), physically abused (83% vs. 45%, p < .01), neglected (73% vs. 32%, p < .01), and sexually abused (21% vs. 6%, p < .01).

Relationship between types of CM and witnessing IPV (N = 16,010).*
A more detailed examination of the relationship between witnessing IPV and types of CM across caregivers and living arrangements of the participants is shown in Table 2. Results of chi-square analyses showed statistically significant differences in witnessing IPV relative to types of CM across all types of caregivers and living arrangements (p < .01). In terms of living with step-parent, participants who reported witnessing IPV were more likely to experience psychological abuse (99% vs. 59%, p < .01), physical abuse (91% vs. 43%, p < .01), neglect (83% vs. 36%, p < .01), and sexual abuse (37% vs. 8%, p < .01). Similar patterns were found for other types of caregivers and living arrangements regarding relationship between types of CM and witnessing IPV.
Relationship Between Types of CM and Witnessing IPV Across Caregivers and Living Arrangements.
Note. CM = child maltreatment; IPV = intimate partner violence.
p < .01.
Table 3 shows risks for CM by gender of the participants who witnessed physical IPV. Among participants who reported witnessing physical IPV, girls had a significantly greater likelihood of experiencing psychological abuse (odds ratio [OR] = 3.7; CI = [1.9, 6.8]), physical abuse (OR = 1.3; CI = [1.0, 1.6]), and neglect (OR = 1.6; CI = [1.4, 1.9]) but less likelihood of experiencing sexual abuse (OR = 0.6; CI = [0.5, 0.7]) than boys after adjusting for age and living arrangement. Among those who witnessed psychological IPV, girls had a significantly greater likelihood of experiencing neglect (OR = 1.3; CI = [1.2, 1.5]) but less likelihood of experiencing sexual abuse (OR = 0.5; CI = [0.5, 0.6]) than boys. Boys who witnessed physical IPV and psychological IPV had a significantly greater likelihood of experiencing sexual abuse compared with girls (Table 4).
Risks for CM by Gender of the Participants Who Witnessed Physical IPV.
Note. CM = child maltreatment; IPV = intimate partner violence; CI = confidence interval.
Adjusted for age, living arrangement.
p < .05.
Risks for CM by Gender of the Participants Who Witnessed Psychological IPV.
Note. CM = child maltreatment; IPV = intimate partner violence; CI = confidence interval.
Adjusted for age, living arrangement.
p < .05.
Discussion
Witnessing IPV increases the chances of child and adolescent victimization. Meta-analyses by Kitzmann, Gaylord, Holt, and Kenny (2003) and Wolfe, Crooks, Lee, McIntyre-Smith, and Jaffe (2003) have reported that children witnessing IPV exhibited significantly worse problems including increased anxieties (Christopoulos et al., 1987), poor school performance (Carlson, 1984), high level of aggression (Jaffe, Wolfe, Wilson, & Zak, 1986), and low self-esteem (Hughes & Barad, 1983). Researchers have identified certain familial risk factors that put children and families at risk for victimization or perpetrations. The co-occurrence of witnessing IPV and physical child abuse was related to fathers’ use of alcohol and drug, and arrest for criminal offenses involving something other than IPV (Hartley, 2002).
To our knowledge, this research is the first published report in SA among children who witnessed IPV and was being victimized. Our study has three major findings. First, there is a high percentage of adolescents who reported witnessed IPV. Second, those who witnessed IPV were more likely to report of being abused compared with those who did not. Third, gender differences were found in terms of relationship between witnessing IPV and forms of CM.
Our study results provide a clear sense of widespread of CM among Saudi children. This study from a representative community sample measured witnessing IPV (52%) and CM was linked to it. Other cross-sectional studies in SA reported prevalence of IPV ranged from 20% to 39% (Afifi, Al-Muhaideb, Hadish, Ismail, and Al-Qeamy, 2011; Barnawi, 2015; Fageeh, 2014). Lack of consistency in the prevalence rate was found because of different methodological issues, for example, sample (community vs. clinical), study site (urban vs. rural), participants (adults vs. children/adolescents), and participants’ knowledge and perception about IPV. Regional population-based studies also reported high percentage of participants who witnessed IPV during their childhood. A study in Jordan revealed that 55% of the women witnessed IPV during their childhood and almost all (98%) reported experiencing at least one type of abuse (Al-Badayneh, 2012). Increasing women’s empowerment and designing effective IPV prevention program may protect them from violent behavior.
Report from National Survey of Children’s Exposure to Violence (NatSCEV) showed that witnessing IPV was closely associated with different forms of abuse (Hamby, Finkelhor, Turner, & Ormrod, 2010). Moffitt and Caspi (2003) reported that the risk for CM at home in which parents physically fought was 3 to 9 times higher than for other children in the study. As information regarding adverse effects of children’s exposure to IPV has been gathered, children’s advocates in both the IPV and child protection fields have searched for ways to work together to keep children and their mothers safe. Although there are different mandates and missions make the collaboration challenging, personnel in both child protection services (CPS) and IPV programs put great effort to find common ground in the common mission of adult and child safety. Such collaboration has allowed CPS and IPV personnel to develop approaches that provide support and safety necessary for shared clients particularly children witnessing IPV (National Resource Center on Domestic Violence, 2002).
Our study finding was similar to a study where 51.7% of girls who witnessed parental abuse also reported child physical abuse compared with 42.9% of the boys (Silvern et al., 1995). In SA, generally, adolescent girls more often report psychological abuse compared with boys who report sexual abuse, and they appear to be at greater risk. As adolescents grow older, they are more likely to experience maltreatment as well. This may be related to the notion that as children mature, they become more aware that certain adult caregiver behaviors like maltreatment are occurring. Alternatively, with advancing adolescence, increasing autonomy and independence may be reflected by greater degrees of conflict between the responding adolescents and their caregivers (Bartle-Haring, Finkelhor, Turner, & Ormrod, 2015) that can become expressed more violently. In addition, boys become more often physically capable compared with girls of defending themselves against a caregiver and thus less likely than girls to experience physical abuse (Bartle-Haring et al., 2015). Several explanations are possible including the possibility that sexual abuse may be regarded as too sensitive among adolescents and, thus, is underreported especially for girls. Alternatively, girls’ sexual victimization are more stigmatizing to family compared with boys due to cultural issues including family honor and respect, the shame surrounding loss of virginity and sexual purity that could affect victims chances of marriage in future. These cultural issues might have resulted in lesser consents to participate in the study from sexually victimized girls or their guardians. A similar observation was noted in a recent epidemiological study in Kuwait (Al-Fayez, Ohaeri, & Gado, 2012).
The study has some limitations. First, potentially sensitive information regarding socially sanctioned behavior may affect responses. Second, as a self-reported questionnaire was used in this study, the reliability of the information was limited by the participants’ ability to recall violent events and to answer trustfully. Third was obtaining parental consents to allow the students to participate in this study. Parents may not consent to their children’s participation if they are fearful about what they will report.
In SA, the last decade witnessed a major progress in the field of child protection with increased awareness, capacity building, and issuing laws and legislations. The law for prevention of violence and abuse and child rights law were approved in August 2013 and November 2014, respectively. It is expected that active implementation of these laws will improve service and outcare of case management. Knowing the family profile of families involved in violence may help better understand this epidemic and provide better services. In 2013, SA initiated a child helpline and a major campaign was conducted in all public schools to inform students to report all problems the children face. This may alleviate the pressure on children by reporting of abuse and getting guidance on how to deal with it. In 2015, violence against children ranked the third among childhood problems reached the line. Around 8% of violence-related calls was describing IPV (Child Helpline Report, 2015). As in this study girls were predominantly the victim of witnessing IPV, more attention should be given to the effect of witnessing IPV among girls. In addition, sexual abuse prevention programs should be targeted among boys who witnessed physical and psychological IPV. Advocacy and The Learning Club (a 16-week intervention for abused women and children; Sullivan, Bybee, & Allen, 2002), Project SUPPORT (for children exposed to IPV who have been diagnosed with aggressive behavior problems; Jouriles, McDonald, Spiller, et al., 2001), and The Kids Club (to enhance children’s recovery from the potentially traumatic effects of exposure to IPV; Graham-Bermann, 2001) are well recognized international intervention programs that can be recommended for children exposed to IPV and should be customized and implemented in SA. Increased awareness of children’s witness to IPV as a form of CM, multidisciplinary approaches involving social workers, law enforcement personnel, domestic violence and child protection workers could be other strategies that could effectively respond to this problem.
Footnotes
Acknowledgements
The authors would like to thank and acknowledge Ms. Sereen Almadani as the main coordinator of the project and for doing the data collection, Ms. Nourah Alkufeidy who contributed in the manuscript preparation, and Ms. Nahrain Quiambao for the administrative support and assistance for the preparation of the article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to thank King Abdullah International Medical Research Center (KAIMRC) for their financial support of this study with grant number: RR11/008.
