Family violence exposure and associated risk factors for child PTSD in a Mexican sample.

G Model CHIABU-2777; No. of Pages 12

ARTICLE IN PRESS Child Abuse & Neglect xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Child Abuse & Neglect

Family violence exposure and associated risk factors for child PTSD in a Mexican sample夽,夽夽 Kara S. Erolin a,∗ , Elizabeth Wieling a , R. Elizabeth Aguilar Parra b a b

Department of Family Social Science, University of Minnesota, 290 McNeal Hall, 1985 Buford Circle, St. Paul, MN 55108, USA Centro de Investigacíon Familiar, A.C., Monterrey, NL, Mexico

a r t i c l e

i n f o

Article history: Received 13 September 2013 Received in revised form 20 April 2014 Accepted 22 April 2014 Available online xxx

Keywords: Mexico Posttraumatic stress disorder Family violence Risk factors Children

a b s t r a c t This study was undertaken in an effort to help illuminate the deleterious effects of traumatic stress on children and families in Mexico. Rates of exposure to traumatic events, family and community violence, and posttraumatic stress disorder (PTSD) were investigated in 87 school-age children and their mothers. Binary logistic regression analysis was performed to examine potential family and ecological risk factors for the presence of child PTSD. A total of 51 children (58.6%) reported an event that met the DSM-IV A criteria, and 36 children (41.4%; 20 boys and 16 girls) met criteria for full PTSD. Traumatic exposure in this sample was considerable, particularly intense, and chronic as a result of interpersonal violence in the home and community. Results support the need for preventive systemic interventions targeting the individual level, parent–child dyadic level, and the larger cultural and community context. Published by Elsevier Ltd.

Introduction Researchers have recognized the importance of a multidisciplinary and multisystemic approach to assessing and treating traumatized children and families, with coordinated services at the individual parent and child level, in addition to the parent–child dyadic level (Appleyard & Osofsky, 2003; Cicchetti & Toth, 1995). A large body of knowledge exists regarding the epidemiology of psychological trauma and posttraumatic stress disorder (PTSD) in industrialized countries. However, our understanding of traumatic stress in economically developing countries is limited. For example, De Girolamo and McFarlane reported in 1996 that only 6% of studies on the prevalence of PTSD had been conducted in developing countries. The paucity of research in developing countries is a concern because these populations appear to be at increased risk for PTSD (Keane, Marshall, & Taft, 2006). For example, reported PTSD rates in the general adult population in Sarajevo (18.6%; Rosner, Powell, & Butollo, 2003) and Afghanistan (20.4%; Scholte et al., 2004) were considerably higher than those in the United States (6.8%) overall (10.4% in women, 5% in men; Kessler et al., 2005) and in other industrialized countries such as Australia (1.5%; Creamer, Burgess, & McFarlane, 2001) and Iceland (0.6% in women, no men met criteria; Lindal & Stefansson, 1993). Furthermore, de Jong et al. (2001) conducted an epidemiological survey in adult survivors of armed conflict, refugees,

夽 This work was supported in part by Centro de Investigacíon Familiar, A.C. Monterrey, N.L. MX; and the Department of Child Protective Services, Monterrey, NL, Mexico. 夽夽 We would like to acknowledge and thank therapists, students and staff at the Centro de Investigacíon Familiar, AC (CIFAC), and the psychologists, social workers and administration at the Desarrollo Integral De La Familia (DIF) whose hard work and dedication were instrumental to this research project. ∗ Corresponding author address: Department of Family Social Science, University of Minnesota, 290 McNeal Hall, 1985 Buford Circle, St. Paul, MN 55108, USA. http://dx.doi.org/10.1016/j.chiabu.2014.04.011 0145-2134/Published by Elsevier Ltd.

Please cite this article in press as: Erolin, K. S., et al. Family violence exposure and associated risk factors for child PTSD in a Mexican sample. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.04.011


ARTICLE IN PRESS

2

K.S. Erolin et al. / Child Abuse & Neglect xxx (2014) xxx–xxx

and displaced persons in four post-conflict, low-income areas and found elevated PTSD prevalence in these areas (Algeria, 37%; Cambodia, 28%; Gaza, 18%; Ethiopia, 16%). Studies of traumatic stress in children show high rates of trauma exposure in non-Western developing countries. For example, lifetime rates of trauma exposure in South African children and adolescents ranged from 40% to 100% in several community samples of South African children and adolescents (Suliman, Kaminer, & Seedat, 2005), and a rate of 77% lifetime exposure to war related trauma and family violence was found in a sample of children aged 7–15 years in Afghanistan (Catani et al., 2009). To our knowledge, no epidemiological study has reported rates of exposure to traumatic events and PTSD or associations between PTSD and family or ecological risk factors in school age children in Mexico. However, such information is critical for policy makers and mental health providers in Mexico tasked with developing a local infrastructure to meet the needs of children and families affected by PTSD, particularly because narcotic-related violence has resulted in Mexico’s current status as one of the most dangerous countries in the world. Gaining an understanding of PTSD in Mexico is also relevant to U.S. policymakers and mental health providers because of the large influx of documented and undocumented immigrants from Mexico into the United States. If we are to provide valid assessment and effective clinical services on this side of the border, it is necessary to know how exposure rates and traumatic stressors prior to immigration contribute to Mexican immigrants’ mental health. Risk Factors for Child PTSD Although we know little about trauma exposure in children in Mexico, epidemiological research on childhood traumatic stress in other countries supports a positive dose–effect relationship between the experience of multiple traumas and adverse outcomes, including the development of PTSD (e.g., Catani, Jacob, Schauer, Kohila, & Neuner, 2008). The risk of PTSD in children is greater after exposure to interpersonal violence in the home and the community than after non-interpersonal traumatic events (McCloskey & Walker, 2000). Lehmann (2000) evaluated 28 studies published from 1980 through 1999 to determine risk factors influencing PTSD symptomatology in children aged 8 months to 9 years. Findings indicated that 85% of children with PTSD had histories of exposure to multiple traumas (physical and sexual abuse, family and community violence, war). Female gender has frequently been linked to increased risk for PTSD (e.g., Cauffman, Feldman, Waterman, & Steiner, 1998; Udwin, Boyle, Yule, Bolton, & O’Ryan, 2000), particularly in children who have been sexually abused (Davis & Siegel, 2000). This has been attributed to a tendency of females to internalize psychological distress, leading to higher rates of mood and anxiety symptoms, contrasting with a tendency of males to exhibit externalizing symptoms (Pine & Cohen, 2002). Additionally, female children may experience more intense reactions to a traumatic event than males. Although female gender is a relatively consistent predictor of PTSD diagnosis in children, the effects are small (Cox, Kenardy, & Hendrikz, 2008; Trickey, Siddaway, Meiser-Stedman, Serpell, & Field, 2012). Parental psychological distress and poor family functioning have been shown to be predictors of child psychopathology (Cox et al., 2008; Trickey et al., 2012). Associations between parental experience of trauma and secondary traumatization of children have been found in studies of war veterans (Dekel & Goldblatt, 2008) and in samples of parents and children who have both been exposed to a range of traumas (Scheeringa & Zeanah, 2001). Lambert, Holzer, & Hasbun, 2014 performed the first meta-analysis examining the association between parents’ PTSD symptoms and children’s psychopathology in 42 studies (34 peer-reviewed journal articles, eight dissertations). The researchers compared studies in which only the parent experienced trauma (n = 18) with studies in which both the parent and the child had a history of trauma exposure (n = 24). A moderate effect size was found for the association between parental PTSD and child distress/behavior problems (r = .35, p < .001), regardless of who experienced the traumatic event(s). This association was significantly stronger for parent–child dyads with a history of interpersonal trauma compared with other event types (r = .46, p < .001). In cases where only the child was exposed to trauma, parental support after the trauma has been found to be one of the most critical factors related to child outcome (Cox et al., 2008; Saile, Neuner, & Catani, 2014; Sriskandarajah, Neuner, & Catani, 2014; Trickey et al., 2012). Parent and family functioning may worsen the impact of a traumatic event on children by serving as proximal reminders and secondary stressors of the trauma. Parents’ reactive behavior after a traumatic event may be more critical to child adjustment than direct exposure to the event (Cox et al., 2008). Parental modeling of maladaptive and avoidant coping strategies may negatively affect a child’s adaptive functioning and may result in less effective parenting skills and decreased ability to provide monitoring and support for the child’s needs. Economic hardship appears to be a significant ecological variable associated with negative responses to trauma and the development of PTSD. The effects of trauma in developing areas of the world are of great concern because these populations may be particularly susceptible to adverse outcomes due to poverty and lack of resources. Norris et al. (2003) identified several risk factors associated with poor countries, including crowded and substandard housing; physically demanding and dangerous work; lack of access to medical and professional care; and enhanced power differentials between rich and poor, adults and children, and men and women. Trauma and PTSD in Mexico Poverty is pervasive throughout Mexico, negatively influencing peoples’ lives in multiple ways. Mexican women and children in particular live in impoverished conditions, which may affect their abilities to cope with traumatic stressors. Please cite this article in press as: Erolin, K. S., et al. Family violence exposure and associated risk factors for child PTSD in a Mexican sample. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.04.011


ARTICLE IN PRESS K.S. Erolin et al. / Child Abuse & Neglect xxx (2014) xxx–xxx

3

Mexican females may be at greater risk for PTSD symptomatology as a result of living in a culture that fosters traditional gender roles such that experiences with discrimination and oppression within society diminishes women’s capacity to adaptively cope with trauma (Baker et al., 2005; Norris et al., 2003). There is a growing body of research about traumatic stress in Mexico, although it is still in its infancy compared to research in the United States. Norris et al. (2003) conducted the first known large-scale epidemiological study in Mexico using a multistage probability sampling design of 2,509 adults (1,602 women, 907 men) representative of four cities: Oaxaca, Guadalajara, Hermosillo, and Mérida. Findings were consistent with previous U.S. and international studies: 76% had lifetime exposure to at least one traumatic event, 70% were exposed to four or more events, and prevalence was higher for men than women (83% vs. 71%). PTSD prevalence (11.2%) was higher than previously reported rates in industrialized countries, but lower than what has been documented in war affected populations. Similar to previous studies women were twice as likely to meet criteria for PTSD as men (15% vs. 7%). Rates were highest in Oaxaca (17%), the poorest city from the total sample population, providing support for the association between impoverished economic conditions and increased PTSD prevalence. A follow-up study was conducted examining the characteristics of traumatic experiences specific to interpersonal violence (sexual assault, sexual molestation, physical assault, threatened with a weapon) and their association with PTSD diagnosis in Mexico (Baker et al., 2005). The lifetime exposure to violence was 34% in this sample, which was appreciably less than in the original study (Norris et al., 2003). Although prevalence of PTSD (11.5%) was similar to that in the previous study, the gender difference was considerably greater, and more than five times as many women met criteria for PTSD than men. Díaz-Olavarrieta, Ellertson, Paz, Ponce de Leon, and Alarcon-Segovia (2002) investigated the incidence and history of battering among women seeking outpatient medical care at a tertiary care teaching hospital in Mexico City and looked for potential risk factors and associations with presenting symptoms. In their sample of 1,780 female patients aged 15 and older exposure to physical and sexual abuse within the past year was reported by 9% of the sample. This percentage was consistent with U.S. estimates (7–30%; Wilt & Olson, 1996), but lower than other representative samples of Mexican women. National data on the prevalence of trauma and PTSD in Mexican youth are scarce, and the few studies that exist concern adolescents. Orozco, Borges, Benjet, Medina-Mora, and López-Carrillo (2007) conducted the first known representative study of adolescents in Mexico City using secondary data from the Mexican Adolescent Mental Health Survey in 3,005 youth (1,440 male; 1,565 female) aged 12–17 years and residing within the metropolitan area. Lifetime exposure to any trauma was 69%, similar to estimates in the adult population. In contrast, only 1.8% of the adolescents (2.4% females; 1.2% males) met criteria for PTSD. Consistent with findings from studies in the U.S. and other parts of the world, rape/sexual assault was the most common traumatic event resulting in PTSD. Pineda-Lucatero, Trujilllo-Hernández, Millán-Guerrero, and Vásquez (2009) determined the prevalence and associated risk factors for child sexual abuse in a sample of 1,067 Mexican junior high school students from both urban and rural backgrounds. Results from the study showed a prevalence rate of 18.7% for the total sample, which is similar to reported rates in other developing countries. This finding may likely be higher given the tendency of underreporting of sexual abuse. Abuse was more frequent in girls (58%) than in boys (42%) although the difference was not statistically significant. The majority (90%) of the experiences occurred between ages 5 and 10 years, and 75% of the cases involved direct physical contact. These results highlight the need for investigation and systematic assessment of factors related to traumatic exposure pre-adolescence in high-risk contexts. The Current Study We therefore conducted a pilot study to begin documenting rates of exposure to traumatic events, prevalence of PTSD, and associations between PTSD and family or ecological risk factors in school age children in the state of Nuevo León, Mexico. Although small in scope, this study was a first attempt by a group of international researchers, a Mexican NGO, and local government to address the need for systematic information regarding these relationships. The specific objectives were to examine (a) individual and family characteristics of mother–child dyads in the presence or absence of child PTSD; and (b) potential family and ecological risk factors for the presence of child PTSD. This study was conducted within the framework of a larger study comprising quantitative measurements, qualitative interviews, and parent–child dyadic observations of structured interaction tasks. The overall study represents a collaborative effort between the University of Minnesota; the Centro de Investigación Familiar, AC (CIFAC), a family therapy training and research center in Monterrey, Mexico; and the Department of Child Protective Services (DCPS) for the state of Nuevo León. Data presented in this manuscript are not intended to stand alone for drawing conclusions regarding the development of systemic assessments and treatments of traumatized populations, and results will not be complete until all data are analyzed. Local Cultural Context of the Current Study The study was conducted in Monterrey (population 3.838 million), the largest city in the Mexican state of Nuevo León, a few hours south of Texas. Monterrey is considered to be the industrial heart of Mexico and is amongst Mexico’s wealthiest cities. It is home to many international corporations, such as Sony, Toshiba, and Ericsson; however, the economic disparity between the rich and the poor is immense. Some of the wealthiest industrialists around the world live in gated communities while countless families are barely able to survive. Many families do not have electricity or running water and many children Please cite this article in press as: Erolin, K. S., et al. Family violence exposure and associated risk factors for child PTSD in a Mexican sample. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.04.011


ARTICLE IN PRESS

4


do not attend school, leaving them with limited options for employment (del Bosque, 2011). Monterrey has historically been known as one of Mexico’s safest cities, but has lately become a dangerous place to live. Since the public launching of President Felipe Calderon’s campaign against drug cartels in 2006 and the subsequent inability of the government to reduce the level of violence, the number of opportunistic crimes (e.g., car theft, robberies) and drug-related crimes has escalated. Nuevo León saw an exponential increase in homicides of 192% from 2009 to 2011 (United States Department of State Bureau of Diplomatic Security, 2012). Method Investigators and Ethical Conduct The co-principal investigators of the overall project were bilingual; one located in the United States and the other in Mexico. Several research team members at CIFAC were also bilingual and a professional bilingual translator/consultant was hired to help manage the assessment-related portions of the project. English is the first language for two and Spanish for one of the co-authors. The pilot study was approved by review boards of the participating institutions in the U.S. and Mexico and adhered to U.S. National Institute of Health guidelines for conducting international research. Sampling and Recruitment Participants for the larger study came from two separate sources. First, 200 mother–child dyads were recruited from a previous epidemiological study conducted in 2007 investigating the prevalence of child maltreatment in 1,000 fifth graders in the state of Nuevo León. Mothers from this study were contacted by phone and asked if they would bring their child to CIFAC for a trauma-focused follow-up interview. A total of 21 children had received a diagnosis of PTSD based on the UCLA PTSD Index (Pynoos, Rodriguez, Steinberg, Stuber, & Frederick, 1998), of which five mother–child dyads were recruited into the pilot study. A random selection table was used to select 30 mother–child dyads from a list of the 179 children who did not meet criteria for PTSD, and once contacted all 30 families agreed to participate in the study. The second source consisted of families reported to the Nuevo León DCPS for child maltreatment who were given the option to undergo the usual 2–3 h unstructured assessment or to participate in the pilot study at CIFAC. A total of 57 families were recruited from this source, including 32 children who met the UCLA PTSD criteria for a PTSD diagnosis and 25 who did not. Children aged 7–14 years were included if the mother qualified and agreed to go through the assessment interview. Mothers were first screened at CIFAC or at DCPS and were excluded from the study if they scored within the psychotic range on the Brief Symptom Inventory (BSI), of which none met exclusionary criteria. The BSI was checked for cultural equivalence with this Mexican sample before being broadly applied. Several safeguards were used to avoid coercion or obligation for participation in the study. For example, one staff person at CIFAC was trained in recruitment procedures including a script of what to say and how to protect all records for confidentiality. Adult males were excluded from the study in order to make comparisons to the existing literature, which indicates heightened vulnerability of women and children to interpersonal violence by male perpetrators (Lehmann, 2000). Interview Procedures Sixteen trained master’s level family therapists and bachelor’s level psychologists and social workers conducted the interviews at CIFAC in pairs. Training consisted of information about research in general and specific to the pilot study; protecting participants’ rights; completing consent/assent procedures, questionnaires, and assessments; how to respectfully ask questions; sensitivity to psychological distress and safety actions during and after the interview; and post-interview paperwork and filing procedures. Three training sessions were held at CIFAC lasting 3–4 full days each. Because of the vulnerability of the study population, a detailed training manual, demonstration role-play videos developed by research team members, and large-group videotaped role playing with the interviewers in training were used to ensure high quality and ethical practice with a vulnerable population. The assessment protocol was tested with two mother–child dyads, and minor modifications were made before implementation with study participants. The first interview for each team was videotaped in its entirety and reviewed by the co-principal investigators for procedural fidelity before approval was given to proceed with more interviews. The interviews were uploaded onto a secure website at CIFAC and downloaded into a data storage and processing program (MediaMill) at the University of Minnesota. The co-principal investigators used an annotated software program (VideoAnt) to provide written comments and feedback at specific places within the video segment that needed improvement. These comments were later accessed by the interviewers through a web-link. All data were encrypted and met safety standards according to institutional review board guidelines. Interviews lasted approximately 2.5 h. The self-report measures were administered by a clinician because many of the mothers were illiterate and potential developmental delays may have existed in the children. Families were informed of the purpose of the study and potential risks and benefits of participating. Each mother signed a consent form giving permission for herself and her child to participate in the study. Children signed a separate assent form. Families were given a gift card (worth US$20) for participating in the study and were provided with a clinical resource list.




5

Characteristics of Participants Four families were excluded from the study because of missing data and a fifth family was excluded because the child was older than 14 years. The total sample comprised 87 mother–child dyads (50 boys, 37 girls). The mean age of the mothers was 35.1 (SD, 6.6; range: 24–50) years. About half (50.6%) were aged 35 and older. The mean age of the children was 10.2 (SD = 2.1) years and 56.3% were aged 10 or older. The average household size was 6.1 children and adults (range: 1–21 people). The mean monthly household income was Mex$5,075 (US$492, based on 6/1/2008 exchange rate of US$0.097 per Mex$1.00), with a maximum income of Mex$15,000 ($1,455). Nearly half (47.1%) of the families lived with a friend or relative, 39.1% resided in a home that they owned (house, apartment, duplex), 12.6% rented, and 1.1% had a different living arrangement. Most mothers (80.5%) were married/cohabitating with their partner, 12.6% were divorced or separated, and 6.9% were single or widowed. The highest educational attainment was middle school (9 years) or less for 66.7% of the mothers, bachelor’s degree or some college for10.3%, and a diploma or some high school for 6.9%. No mothers had graduate level education, and 16.1% had some other educational background. Close to half of the mothers (46%) were self-employed or worked for a company outside the home, 39.1% were full-time homemakers, 9.2% were full-time students, retired, or permanently disabled, and 5.6% were temporarily laid off or unemployed. Sociodemographic variables did not differ between children who met diagnostic criteria for PTSD (n = 36) and those without PTSD (n = 51), or between genders. Compared to children without PTSD (M = 10.61, SD = 2.08), children with PTSD (M = 9.69, SD = 2.03) were significantly younger by an average of 1 year, t(85) = 2.04, p = .045. Measures Quantitative measures were evaluated for cross-cultural equivalency across five dimensions: content, semantics, technical, criterion, and conceptual (Flaherty et al., 1988). The goal was to achieve equivalence across all dimensions; however, this was an ideal standard, because the dimensions were not mutually exclusive. Three measures (UCLA PTSD Index, Posttraumatic Diagnostic Scale, and Composite Abuse Scale) were standardized and demonstrated satisfactory reliability and validity in U.S. and non-U.S. populations. Spanish versions were used and instrument items were evaluated for local dialectal differences; minor changes were made to more accurately reflect the meaning of each item. Trauma and PTSD. The UCLA PTSD Index, a 20-item instrument designed to assess exposure to trauma and PTSD symptomatology in youth 7–18 years old (Pynoos et al., 1998), was administered to all children at study entry. A scoring sheet allows for coding of exposure to a traumatic event in Part I, and criteria A1 and A2 (objective and subjective aspects of exposure) in Part II. The 17 scores in Part III directly correspond to the DSM-IV PTSD symptom criteria to calculate a total PTSD severity score and severity scores for each of the subcategories B (reexperiencing), C (avoidance), and D (arousal). The items are rated on a 5-point Likert scale ranging from 0 (none of the time) to 4 (most of the time) to assess for the frequency of symptoms during the past month. An overall severity score ≥38 is considered to have the greatest sensitivity and specificity for identifying PTSD (Steinberg, Brymer, Decker, & Pynoos, 2004). The instrument has sound psychometric properties and has been translated for broad application to a variety of settings and cultures around the world. Internal consistency with children in this sample was similar to that of previous studies; Cronbach’s alpha coefficient was 0.93 for the full measure, 0.84 for criterion B, 0.86 for C, and 0.83 for D. The Posttraumatic Stress Diagnostic Scale (PDS) (Foa, 1993) was used to assess the mothers’ exposure to trauma and PTSD symptomatology. Parts I and II of the 49-item questionnaire address DSM-IV criterion A. Part III consists of 17 items corresponding to symptom criteria B, C, and D, which are scored for overall PTSD severity and severity for each subcategory. Respondents indicate how often a particular symptom has bothered them within the past month on a 4-point Likert scale: 0 (not at all), 1 (once a week or less), 2 (2–4 times a week), and 3 (almost always). Part IV includes nine yes/no items addressing functional impairment (criterion F). Criteria for diagnosis of PTSD consist of the fulfillment of criterion A; a rating of “1” or higher on at least one B item, three C items, and two D items; symptom duration of at least one month; and impairment in at least one F-item. Symptom severity cutoff scores are: ≤10 (mild), ≥11 and ≤20 (moderate), ≥21 and ≤35 (moderate to severe), and ≥36 (severe). Previous studies with diverse adult populations in the United States and war-affected populations in other countries have demonstrated high internal consistency for the measure (>0.90) and subscale reliabilities for criteria B, C, and D ranging from 0.71 to 0.89 (Ertl et al., 2011; Foa, Cashman, Jaycox, & Perry, 1997). Cronbach’s alpha coefficient within our sample of mothers was 0.93 for the measure, 0.85 for criterion B, 0.83 for C, and 0.85 for D. Family Violence. Children’s exposure to family violence was assessed using the domestic violence subsection from the Family Violence Checklist (Catani et al., 2008). It consists of 31 items across five types of family violence: physical, emotional, sexual, neglect, and witnessing violence. Responses are dichotomous (yes/no) to whether an event has ever occurred (lifetime exposure) and whether the event occurred in the last month (ongoing exposure). Catani et al. developed the instrument for use with children aged 9–15 years in Sri Lanka to assess additional traumatic events apart from war and tsunami-related stress. They selected items from two standardized measures (Early Trauma Inventory, Childhood Trauma Questionnaire) in collaboration with local counselors and experts to determine the most relevant items for Sri Lankan culture. In the current study, local clinicians and research team members evaluated instrument items for appropriate use with a Mexican population. It was determined that all of the items were relevant to the local cultural and trauma context. Minor modifications were made to clarify the meaning of items. The English version of the measure was translated into Spanish and back translated by two bilingual research team members. The psychometric properties of the instrument have not been Please cite this article in press as: Erolin, K. S., et al. Family violence exposure and associated risk factors for child PTSD in a Mexican sample. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.04.011


ARTICLE IN PRESS

6


tested, but it has been used in two studies in Sri Lanka (Catani et al., 2008; Elbert, Schauer, Huschka, Hirth, & Neuner, 2009) and one study in Afghanistan (Catani et al., 2009). An exploratory factor analysis was not conducted in the current study, given the small N for which the subject to item ratio of 3:1 was significantly smaller than the recommended 20:1 ratio yielding the most replicable results (Costello & Osborne, 2005). The 30-item version of the Composite Abuse Scale (CAS) (Hegarty, 2007) was used to assess mother’s exposure to intimate partner violence. Items were selected from four standardized domestic violence measures: the Revised Conflicts Tactics Scale, Psychological Maltreatment Women Index, Index Spouse Abuse, and Measure of Wife Abuse. The instrument addresses four dimensions of intimate partner violence (severe combined abuse, emotional, physical, harassment) on a 6-point Likert scale to assess the frequency of specific events during the past 12 months: 0 (never), 1 (only once), 2 (several times), 3 (monthly), 4 (weekly), and 5 (daily). Frequency scores are summated for each subscale and a total score is calculated by summing all scores across subscales. It has been used in clinical settings with individuals and groups in Australia and has been translated into Vietnamese, Dutch, and Bengali. Hegarty, Bush, and Sheehan (2005) reported the following reliability coefficients: 0.91 severe combined abuse, 0.93 emotional, 0.94 physical, and 0.87 harassment. Cronbach’s alpha on the subscales with this sample of mothers were lower compared to the aforementioned study, but still in the acceptable to good range: 0.73 severe combined abuse, 0.89 emotional, 0.90 physical, and 0.87 harassment. Statistical Procedures The first study aim was to explore individual and family characteristics that distinguish between mother–child dyads with and those without child PTSD. Descriptive statistics were used to identify general characteristics of these groups, taking gender of the child into account. The significance of differences between groups was tested by using Pearson’s chi-square tests for categorical variables and independent-sample t tests for continuous variables. Fischer’s exact test was used to perform chi-square tests in cases where one or more of the cells had an expected frequency of five or less, providing a more accurate estimate of the expected values for small sample sizes (McDonald, 2009). The second aim of the study was to examine potential family and ecological risk factors for the presence of child PTSD. This was accomplished by performing binary logistic regression analyses to examine the relationship between the dichotomous outcome variable presence or absence of child PTSD diagnosis and several independent variables hypothesized to predict diagnosis. The assumptions of logistic regression were first checked and met. Candidate predictors considered for inclusion in the regression model were items from the administered scales, which were chosen on the basis of literature regarding childhood traumatic stress. A common rule of thumb for determining the appropriate number of independent variables to include in a regression model is a minimum of 10 events per variable or at least 10 outcomes for each binary category (Altman & Royston, 2000). Backward stepwise logistic regression was used with selection based on the likelihood ratio to determine candidate predictors significantly contributing to the model (p < .05). One covariate was successively removed at each step. This approach was appropriate given the exploratory nature of the study to build a final model that was parsimonious and reliable, and is commonly used in clinical research (Stoltzfus, 2011). Results A total of 92 mother–child dyads were included in the study. Of these, four dyads were excluded because of missing data, and one was excluded for other reasons, resulting in a total study population consisting of 87 mother–child dyads. Exposure to Trauma and Family Violence Analysis of the UCLA PTSD Index data for the children revealed that 58.6% (51 children: 33 boys, 18 girls) reported an event that met the DSM-IV A criteria. The mean number of lifetime events was 1.29 (SD = 1.41), and 37% of the children were exposed to multiple events. The three most frequent event types were seeing someone being beaten up, shot at, or killed in town (19.5%); seeing a family member being hit, punched, or kicked very hard at home (17.2%); and seeing a dead body in town, not including funerals (17.2%). More boys experienced most event types, except being beaten up, shot at, or threatened to be hurt badly in town, and having an adult touch private sexual body parts without permission. Significantly more boys than girls were affected by hearing about a violent death or serious injury of a loved one (18% vs. 2.7%), 2 (1, N = 87) = 4.89, p = .039 and by being in a very serious accident (12% vs. 0%), 2 (1, N = 87) = 4.77, p = .036. The criteria for full PTSD were met for 41.4% of the total sample (36 children: 20 boys and 16 girls), and 10.3% met the partial PTSD condition. Mean UCLA PTSD Index scores for severity and number of symptoms and impairment of psychosocial functioning are given in Table 1. The highest PTSD severity score for the sample was 64 (of 68), the greatest number of symptoms was 17 (of 17), and the largest number of areas of impaired functioning was four out of a possible six. Table 2 gives the frequency of child lifetime exposure to potentially traumatic events for children with and those without a PTSD diagnosis. According to the PDS data, 55.2% of the mothers were exposed to at least one lifetime traumatic event and 47% were exposed to two or more events (M = 1.79, SD = 1.86). Mothers most frequently reported exposure to a natural disaster (25.3%) followed by being in or witnessing a serious accident, fire, or explosion (23%). Significantly higher rates of exposure to sexual assault by a family member or someone known were found in mothers of children with PTSD than in mothers whose child Please cite this article in press as: Erolin, K. S., et al. Family violence exposure and associated risk factors for child PTSD in a Mexican sample. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.04.011



7

Table 1 UCLA PTSD Index scores for severity, number of symptoms, and impairment of psychosocial functioning in children in Nuevo León, Mexico (n = 36). All M

Gender SD

Boys (n = 20) M

PTSD severity score PTSD symptoms Impaired areas of psychosocial functioning

24.94 11.42 1.72

12.70 3.03 1.32

21.65 10.80 1.45

Girls (n = 16) SD

M

10.25 2.53 1.32

29.10 12.19 2.06

SD 14.52 3.49 1.29

Note. PTSD = posttraumatic stress disorder.

did not have PTSD (33.3% vs. 9.8%), 2 (1, N = 87) = 7.43, p = .006. No differences were found between groups for other events types. Twelve mothers met diagnostic criteria for PTSD of which seven had children with PTSD. Data from the Family Violence Checklist indicated that 87.4% of the children had been exposed to one or more types of violence in the home (76 children: 46 boys, 30 girls). The mean lifetime number of events was 3.85 (SD = 3.43), and the greatest number was 18. Children most often experienced emotional abuse (70.1%) and physical abuse (67.8%). About one third witnessed violence toward a family member (34.5%), 19.5% experienced neglect, and 3.4% experienced sexual abuse. Continuous violence in the home was common; 65.5% (57 children: 36 boys, 21 girls) experienced at least one event type in the past month. The mean number of events was 2.02 (SD = 2.22), and the maximum number was 10. Ongoing exposure to emotional abuse was reported in 50.6%, physical abuse in 48.3%, witnessing violence in 11.5%, neglect in 9.2%, and sexual abuse in 1.1%. Table 3 gives an overview of the children’s lifetime exposure to different types of family violence in relation to PTSD diagnosis and gender. Analysis of the CAS data showed that 78.2% of mothers had been exposed to intimate partner violence. The percentages of mothers with exposure to various types of intimate partner violence are presented in relation to the child’s diagnosis in Table 4. The mean of number of events was 14.21 (SD = 20.38), with a maximum number of 108. Mothers with a child diagnosed with PTSD (M = 19.42, SD = 26.98) had experienced more violence than mothers whose child did not meet PTSD criteria (M = 10.53, SD = 13.10), but the difference was not significant. The most frequent experience was emotional abuse, and 52.9% had experienced 3 or more episodes. Nearly half (49.9%) had undergone 1 or more episodes of physical abuse and severe combined abuse, and 31% had experienced 2 or more episodes of harassment. Predicting Child PTSD Diagnosis A 5-predictor model of child PTSD was used: (1) child lifetime exposure to all types of family violence, (2) child seeing family member being hit, punched, or kicked very hard at home, (3) child having adult touch private sexual body parts, (4) mother sexual assault by family member/someone known, and (5) mother sexual assault by stranger. Multiple indicators were used to provide a comprehensive assessment of the binary logistic regression results including: (a) overall model fit, (b) statistical tests of individual predictors, (c) an assessment of the predicted probabilities, and (d) model validation (Peng, Lee, & Ingersoll, 2002). The results of logistic regression are presented in terms of the odds of an outcome. The 5-predictor model was statistically significant and fit the data better than the null (constant only) model in distinguishing between children with PTSD and children without PTSD, 2 (5, N = 87) = 39.65. The coefficient of determination (Nagelkerke R2 ) was 0.49, indicating that approximately half of the variance in the PTSD status of the children was predicted from the linear combination of the five independent variables. The statistical significance of the individual regression coefficients (Bs) of each predictor was tested using the Wald chi-square statistic. The variables “child having adult touch private sexual body parts” and “mother sexual assault by stranger” were not significant contributors to the regression model. The remaining three covariates were significant predictors of child PTSD diagnosis as indicated by the odds ratios (ORs) in Table 5. A one-unit increase in the variable “child lifetime exposure to all types of family violence” was associated with a 26% increase in the odds of child PTSD (OR = 1.26). This factor had the least impact on the odds of a child being diagnosed with Table 2 Percentages of Child Lifetime Exposure to Traumatic Events (UCLA PTSD Index) in relation to PTSD Diagnosis (N = 87).

Being in another kind of disaster (fire, tornado, flood or hurricane) Being in a very serious accident Being hit, punched, or kicked very hard at home Seeing family member hit, punched, or kicked very hard at home Being beaten up, shot at or threatened to be hurt badly in town Seeing someone being beaten, shot at or killed in town Seeing a dead body in town Having adult touch private sexual body parts Hearing about a violent death/serious injury of a loved one Having painful or scary medical treatment in hospital

PTSD (n = 36)

No-PTSD (n = 51)

13.9 11.1 25.0 36.1 11.1 25.0 16.7 8.3 22.2 27.8

3.9 3.9 3.9 3.9 2.0 15.7 17.6 0.0 3.9 3.9



ARTICLE IN PRESS

8


Table 3 Percentages of Child Lifetime Exposure to Family Violence (Family Violence Checklist) in relation to PTSD diagnosis and gender (N = 87). Types of family violence

Physical abuse Hit with object Slapped Arms twisted/pulled by hair Punched or kicked Things being thrown at Tied or locked up Burned/attempted to burn or strangle Injured Emotional abuse Shouted, screamed or sworn at Verbally threatened Told you are not good Made fun of in front of others Threatened to be hurt or killed Witnessing violence Family member hit, punched or kicked Family member threatened to be killed Family member injured Neglect Ignored by parents Had to wear dirty or ragged clothes Not given enough food or drink Sexual abuse Intimately touched by older person Forced to intimately touch older person * ** ***

Diagnosis

Gender

PTSD (n = 36)

No-PTSD (n = 51)

Boys (n = 50)

Girls (n = 37)

75.0 55.6 47.2 41.7 33.3 30.6 13.9 8.3 5.6 72.2 58.3 30.6* 25.0** 22.2 13.9 41.7 36.1 13.9** 8.3 30.6* 13.9 8.3 11.1 8.3 8.3 2.8

62.7 41.2 39.2 25.5 31.4 15.7 5.9 3.9 0.0 68.6 62.7 9.8 3.9 11.8 2.0 29.4 21.6 0.0 2.0 11.8 2.0 5.9 5.9 0.0 0.0 0.0

86.0*** 58.0* 58.0** 42.0* 42.0* 26.0 12.0 2.0 0.0 74.0 60.0 22.0 10.0 16.0 6.0 38.0 26.0 2.0 4.0 22.0 10.0 10.0 6.0 0.0 0.0 0.0

43.2 32.4 21.6 18.9 18.9 16.2 5.4 10.8 5.4 64.9 62.2 13.5 16.2 16.2 8.1 29.7 29.7 10.8 5.4 16.2 2.7 2.7 10.8 8.1 8.1 2.7

p < .05. p < .01. p < .001.

PTSD. The variable “child seeing family member being hit, punched or kicked very hard at home” had the greatest impact; every one-unit increase in this predictor was associated with a 92% increase in the odds of the child not having a PTSD diagnosis (OR = 0.08). Mother sexual assault by family member/someone known also had a large effect, and every one-unit increase in this predictor was related to an 82% increase in the odds of the child not having a PTSD diagnosis (OR = 0.18). Additionally, the confidence intervals (CIs) for these three predictors did not cross the value of 1.0 suggesting that these covariates were significant contributors to child PTSD in the sample (Stoltzfus, 2011). The predictive power of the model was evaluated according to the classification table (c-index), which indicates the discrimination ability of the model and how well the combination of variables predicts child PTSD. The c-statistic (concordance) for sensitivity (non-event) and specificity (event) were both above the 0.50 cutoff, indicating positive results; however, the covariates were better at predicting children without PTSD than children with a PTSD diagnosis. The overall percentage correctly predicted by the model was 80.5%, indicating a 21.9% increase in the predictive power of the model with the five predictors compared to the null model (58.6%), a substantial improvement over the chance level. Discussion Consistent with the overall goal of this study, we wanted to establish a more culturally grounded knowledge base. Therefore, the local cultural context in Monterrey was the starting point for conceptualizing, developing, and designing the study from which these data are drawn. We used a combined approach, integrating knowledge from within the culture (emic) with knowledge from outside the culture (etic) to understand and assess the impact of trauma on families in Mexico (Bernal, Bonilla, & Bellido, 1995). Multiple ecological systems of interaction were considered, and families and researchers were all part of the scientific process. This study contributes to the international literature by illuminating the impact of traumatic exposure on families (child and mother) and related risk factors predicting child PTSD within a specific cultural and trauma context in Mexico. Our subjects shared many similarities with traumatized populations in the United States and with older-age samples in Mexico. As expected, lifetime exposure to at least one traumatic event was common, and many children and mothers were exposed to two or more events. PTSD severity, number of symptoms, and impairment of psychosocial functioning in the children in our sample were similar to trends previously documented for overall prevalence and gender differences, with more girls than boys meeting diagnostic criteria, experiencing more severe symptomatology, and exhibiting more impairment in psychosocial functioning. The children and mothers in this study also had some unique characteristics, particularly the high prevalence of family and community violence; high exposure rates were found across the entire sample, Please cite this article in press as: Erolin, K. S., et al. Family violence exposure and associated risk factors for child PTSD in a Mexican sample. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.04.011



9

Table 4 Percentages of Mother Lifetime Exposure to Intimate Partner Violence (Composite Abuse Scale) in relation Child’s PTSD diagnosis (N = 87). Types of violence

Child’s diagnosis

Emotional abuse Told me I wasn’t good enough Tried to turn family, friends, and children against me Told me I was ugly Tried to keep me from seeing/talking to my family Blamed me for causing their violent behavior Became upset if dinner/housework wasn’t done Told me I was crazy Told me no one would ever want me Did not want me to socialize with female friends Tried to convince my friends, family or children I was crazy Told me I was stupid Physical abuse Slapped me Threw me Shook me Pushed, grabbed or shoved me Hit/tried to hit me with something Kicked, bit or hit me with a fist Beat me up Severe combined abuse Kept me from medical care Locked me in the bedroom Raped me Tried to rape me Used a knife or gun or other weapon Took my wallet and left me stranded Put foreign objects in my vagina Refused to let me work outside the home Harassment Followed me Hung around outside my house Harassed me over the phone Harassed me at work * **

PTSD (n = 36)

No-PTSD (n = 51)

75.0 41.7 27.8 30.6* 27.8 50.0 36.1 25.0 22.2 30.6 13.9 30.6* 50.0 25.0 44.4 33.3 41.7 33.3 25.0 16.7** 50.0 22.2 13.9 13.9 11.1 13.9 16.7 2.8 19.4 36.1 25.0 25.0 11.1 16.7

72.5 23.5 21.6 9.8 13.7 51.0 25.5 21.6 21.6 39.2 13.7 11.8 49.0 31.4 35.3 27.5 37.3 15.7 13.7 0.0 49.0 11.8 9.8 3.9 5.9 5.9 7.8 0.0 33.3 33.3 17.6 13.7 9.8 11.8

p < .05. p < .01.

including in children and mothers who did not meet diagnostic criteria for PTSD. The most frequent traumatic events experienced by children in this sample were intense, most likely as a result of the interpersonal nature and chronicity of the violence. The mothers’ lifetime exposure to intimate partner violence was considerably higher than previous estimates in the United States and Mexico; however, PTSD prevalence in the mothers was much lower than in previous studies. Diagnostic assessment was based on the worst event reported by mothers across a number of traumatic stressors rather than experiences solely with intimate partner violence, which could help explain the low PTSD prevalence in mothers. It appears that children and mothers in this study are a particularly vulnerable population. Not only were they exposed to multiple incidences of abuse in the home, but they also encountered daily violence outside of their homes. “Seeing someone being beaten up, shot at, or killed in town” and “seeing a dead body in town (not including funerals)” were two of the event types children most frequently reported. Sadly, public executions by shooting and hanging, assault, and bodily mutilations are becoming commonplace in Monterrey. Families in this sample were extremely impoverished, living on an average monthly income of Mex$5,075 (US$492) for six people compared to the Mexican national average of Mex$8,422 MXN (US$817) (Gardner, 2005). The drug cartels are quickly becoming a last resort out of crippling poverty as a growing number of youth look toward a future with little hope (del Bosque, 2011). Table 5 Logistic Regression Model Predicting Child PTSD diagnosis (N = 87). Predictor

B

C-all types of family violence C-seeing family hit, punched or kicked at home C-adult touched private sexual parts M-sexual assault family/someone known M-sexual assault stranger Constant

0.23 −2.59 −21.37 −1.73 21.70 2.01

SE

OR [95% CI]

0.09 0.86 22,057.84 0.79 17,658.15 28,255.24

1.23 [1.05–1.51] 0.08 [0.01–0.40] 0.00 [0.00–infinity] 0.18 [0.04–0.83] 2.66 [0.00–infinity] 7.43 [NA]

Wald 2 6.38 9.14 0.00 4.85 0.00 0.00

p .01 .99

Note. CI = confidence interval for odds ratio (OR). df = 1. C = child variables. M = mother variables.



ARTICLE IN PRESS

10


Our findings show the need for a deeper investigation to differentiate factors involved in the development of PTSD in contexts where everyone is exposed to high levels of family and community violence. Recent studies conducted in Sri Lanka (Sriskandarajah et al., 2014) and Uganda (Saile et al., 2014) with children exposed to traumatic events related to war and family violence indicated that event types within the family had a greater likelihood of being associated with child PTSD, and positive parenting and close parental bond were the most significant factors mediating children’s development of PTSD. The findings in this sample seem to point to a similar association between parental factors affecting children’s PTSD symptomatology. Some limitations of the study include the small size and non-representative nature of the sample and the inability to externally validate the results with a different community-based sample, which limit the generalizability of the findings to the larger population and clinical application in the real world. Furthermore, given the nature of family violence and the secrecy that often surrounds the perpetuation of violence in the home, it can be challenging to obtain accurate and representative prevalence rates of trauma exposure and PTSD.

Challenges and Successes of Cross Cultural Research This study was conducted within the framework of a larger research project involving multiple stakeholders, including families, researchers, clinicians, and state and national institutions. Provision of services for child maltreatment in families is a recognized area of need in Mexico, and this study was partially funded by the Mexican DCPS because of their interest in developing more effective ways to assess and treat affected families. Several structural and organizational challenges at the institutional level influenced the design and implementation of this research. The organizational structure of DCPS is very hierarchical, no consistent assessment or intervention guidelines exist, and the sustainability of social services programs is limited. These factors made it difficult for clinicians and staff to stick to a highly structured assessment protocol and data entry procedures. The clinicians only had professional training in psychoanalysis, with few opportunities for continuing education related to family/relational assessments and interventions. Many of the clinicians and staff felt demoralized, expressing feelings of hopelessness because they wanted to help the families they worked with but felt inadequately prepared to do so. Limited resources, low wages, exorbitant workloads, and tremendous time demands lead to burnout. At the family level, mothers and children were faced with the challenges of extreme poverty, inability to fulfill basic needs, high stress levels, illiteracy, lack of accessible transportation, and high levels of family and community violence. These families are entrenched in multigenerational family and contextual systems that are difficult to change. The result is an isomorphic process in which parents do not know how to help their children and professionals do not know how to help the families. Nevertheless, several successes have been achieved throughout the larger project. The level of commitment from administrators, clinicians, and staff at DCPS and CIFAC was impressive. The few resources available were used to support the research efforts across various research phases. Internal structural changes within the organizations were made in order to accommodate the training schedule. For example, time spent in training sessions and in interviewing families was included as a part of regular work hours. The training sessions also served as continuing education, which increased the morale of the clinicians and staff, helped them to feel more confident and better equipped to help families, and also reinforced their eagerness and commitment to the research process. It is important to note that research is not typically integrated into government institutions in Mexico. DCPS directors were progressive in acknowledging the importance of investing in research regarding alternative ways of approaching trauma related to maltreatment of children. The U.S. research team provided significant training related to research in order to get buy-in from administrators, clinicians, staff, and families. Mothers also demonstrated their commitment to the health and wellbeing of their children by making the time and necessary sacrifices to participate in the study.

Implications for Child Mental Health This study highlights the importance of family and ecological factors influencing children’s wellbeing and mental health. Children living in a recovery environment characterized by high levels of ongoing violence in the home and community, extreme poverty, and parental psychological distress are at elevated risk for PTSD and/or exacerbation of ongoing disorders. Moreover, cultural factors may affect the meaning and expression of trauma symptoms and how individuals, families, and communities respond to and cope with traumatic stress and violence. Some cultures may be more permissive than others in tolerating mistreatment of vulnerable groups and the social and legal climate may affect perceptions and reporting of violence (Díaz-Olavarrieta et al., 2002). Preventive interventions targeting parents and caregivers might be one of the most direct paths to interrupting the escalation of children’s mental health disorders and potential intergenerational transmission of violence. Gewirtz, Forgatch, and Wieling (2008) underscored the need for more attention to the proximal mechanism of parenting practices as potential mediators between children’s adjustment after exposure to traumatic events. More family-based research, especially with school age and younger children, is needed to examine intra- and inter-familial influences on the parent–child relationship. Understanding how parenting and family interventions might mitigate children’s PTSD symptomatology and Please cite this article in press as: Erolin, K. S., et al. Family violence exposure and associated risk factors for child PTSD in a Mexican sample. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.04.011



11

other psychological difficulties has important implications for institutional systems such as DCPS—in Mexico and around the world—in addressing children’s mental health. References Altman, D. G., & Royston, P. (2000). What do we mean by validating a prognostic model? Statistics in Medicine, 19, 453–473. Appleyard, K., & Osofsky, J. D. (2003). Parenting after trauma: Supporting parents and caregivers in the treatment of children impacted by violence. Infant Mental Health Journal, 24(2), 111–125. Baker, C. K., Norris, F. H., Diaz, D. M., Perilla, J. L., Murphy, A. D., & Hill, E. G. (2005). Violence and PTSD in Mexico: Gender and regional differences. Social Psychiatry and Psychiatric Epidemiology, 40, 519–528. Bernal, G., Bonilla, J., & Bellido, C. (1995). Ecological validity and cultural sensitivity for outcome research: Issues for the cultural adaptation and development of psychosocial treatments with Hispanics. Journal of Abnormal Child Psychology, 23(1), 67–82. Catani, C., Jacob, N., Schauer, E., Kohila, M., & Neuner, F. (2008). Family violence, war, and natural disasters: A study of the effect of extreme stress on children’s mental health in Sri Lanka. BMC Psychiatry, 8(33), 1–28. Catani, C., Schauer, E., Elbert, T., Missmahl, I., Bette, J. P., & Nuener, F. (2009). War trauma, child labor, and family violence: Life adversities and PTSD in a sample of school children in Kabul. Journal of Traumatic Stress, 22(3), 163–171. Cauffman, E., Feldman, S., Waterman, J., & Steiner, H. (1998). Posttraumatic stress disorder among female juvenile offenders. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 1209–1215. Cicchetti, D., & Toth, S. L. (1995). A developmental psychopathology perspective on child abuse and neglect. Journal of the American Academy of Child and Adolescent Psychiatry, 34(5), 541–565. Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research & Evaluation, 10(7), 1–9. Cox, C. M., Kenardy, J. A., & Hendrikz, J. K. (2008). A meta-analysis of risk factors that predict psychopathology following accidental trauma. Journal for Specialists in Pediatric Nursing, 13(2), 98–110. Creamer, M., Burgess, P., & McFarlane, A. C. (2001). Posttraumatic stress disorder: Findings from the Australian National Survey of Mental Health and Well-Being. Psychological Medicine, 31, 1237–1247. Davis, L., & Siegel, L. J. (2000). Posttraumatic stress disorder in children and adolescents: A review and analysis. Clinical Child and Family Psychology Review, 3(3), 135–154. De Girolamo, G., & McFarlane, A. C. (1996). The epidemiology of PTSD: A comprehensive review of the international literature. In A. J. Marsella, M. J. Friedman, E. T. Gerrity, & R. M. Scurfield (Eds.), Ethnocultural aspects of posttraumatic stress disorder: issues, research, and clinical applications (pp. 33–85). Washington, DC: American Psychological Association. de Jong, J. T., Komproe, I. H., Ommeren, M. V., El Masri, M., Araya, M., Khaled, N., van De Put, W., & Somasundaram, D. (2001). Lifetime events and posttraumatic stress disorder in four postconflict settings. Journal of the American Medical Association, 286(5), 555–562. Dekel, R., & Goldblatt, H. (2008). Is there intergenerational transmission of trauma? The case of combat veterans’ children. American Journal of Orthopsychiatry, 78, 281–289. del Bosque, M. (2011). Can Monterrey save itself from the bloody drug war? Texas observer. Retrieved from: http://www.texasobserver.org/ lalinea/can-monterrey-save-itself-from-the-bloody-drug-war Díaz-Olavarrieta, C., Ellertson, C., Paz, F., Ponce de Leon, S., & Alarcon-Segovia, D. (2002). Prevalence of battering among 1780 outpatients at an internal medicine institution in Mexico. Social Science and Medicine, 55, 1589–1603. Elbert, T., Schauer, M., Huschka, E., Hirth, M., & Neuner, F. (2009). Trauma-related impairment in children: An epidemiological survey in Sri Lankan provinces affected by armed conflict. Child Abuse & Neglect, 33(4), 238–246. Ertl, V., Pfeiffer, A., Saile, R., Schauer, E., Elbert, T., & Neuner, F. (2011). Validation of a mental health assessment in an African conflict population. Psychological Assessment, 22(2), 318–324. Flaherty, J. A., Gavieria, F. M., Pathak, D., Mitchell, T., Wintrob, R., Richman, J. A., & Birz, S. (1988). Developing instruments for cross-cultural psychiatric research. Journal of Nervous and Mental Disease, 176(5), 257–263. Foa, E. B. (1993). Posttraumatic Stress Diagnostic Scale manual. Minneapolis, MN: National Computer Systems, Inc. Foa, E. B., Cashman, L., Jaycox, L., & Perry, K. (1997). The validation of a self-report measure of posttraumatic stress disorder: The Posttraumatic Stress Diagnostic Scale. Psychological Assessment, 9(4), 445–451. Gardner, P. (2005). Household income and expenditure statistics (HIES): A comparison of eight countries. International Labour Office Bureau of Statistics on-line database LABORSTA. Retrieved from: http://laborsta.ilo.org Gewirtz, A., Forgatch, M., & Wieling, E. (2008). Parenting practices as potential mechanisms for child adjustment following mass trauma. Journal of Marital and Family Therapy, 34(2), 177–192. Hegarty, K. (2007). Composite Abuse Scale manual. Melbourne, Australia: Department of General Practice, University of Melbourne. Hegarty, K., Bush, R., & Sheehan, M. (2005). The Composite Abuse Scale: Further development and assessment of reliability and validity of a multidimensional partner abuse measure in clinical settings. Violence and Victims, 20(5), 529–547. Keane, T. M., Marshall, A. D., & Taft, C. T. (2006). Posttraumatic stress disorder: Etiology, epidemiology, and treatment outcome. Annual Review of Clinical Psychology, 2, 161–197. Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005). Lifetime prevalence of age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62, 593–602. Lambert, J. E., Holzer, J., & Hasbun, A. (2014). Association between parents’ PTSD severity and children’s psychological distress: A meta-analysis. Journal of Traumatic Stress, 27, 9–17. Lehmann, P. (2000). Posttraumatic stress disorder (PTSD) and child witnesses to mother-assault: A summary and review. Children and Youth Services Review, 22(3/4), 275–306. Lindal, E., & Stefansson, J. G. (1993). The lifetime prevalence of anxiety disorders in Iceland as estimated by the U.S. National Institute of Mental Health Diagnostic Interview Schedule. Acta Psychiatry Scandinavia, 88, 29–34. McCloskey, L. A., & Walker, M. (2000). Posttraumatic stress in children exposed to family violence and single-event trauma. Journal of the Academy of Child and Adolescent Psychiatry, 39(1), 108–115. McDonald, J. H. (2009). Fischer’s exact test of independence. In Handbook of biological statistics (2nd ed., pp. 70–75). Baltimore, MD: Sparky House Publishing. Norris, F. H., Murphy, A. D., Baker, C. K., Perilla, J. L., Gutiérrez Rodriguez, F., & de Jesús Gutiérrez Rodriguez, J. (2003). Epidemiology of trauma and posttraumatic stress disorder in Mexico. Journal of Abnormal Psychology, 112(4), 646–656. Orozco, R., Borges, G., Benjet, C., Medina-Mora, M. E., & López-Carrillo, L. (2007). Traumatic life events and posttraumatic stress disorder among Mexican adolescents: Results from a survey. Salud Pública de México, 50(1), S29–S37. Peng, C. J., Lee, K. L., & Ingersoll, G. M. (2002). An introduction to logistic regression analysis and reporting. Journal of Educational Research, 96(1), 3–14. Pine, D. S., & Cohen, J. A. (2002). Trauma in children and adolescents: Risk and treatment of psychiatric sequelae. Biological Psychiatry, 51, 519–531. Pineda-Lucatero, A. G., Trujilllo-Hernández, B., Millán-Guerrero, R. O., & Vásquez, C. (2009). Prevalence of child sexual abuse among Mexican adolescents. Child Care, Health and Development, 35(2), 184–189. Pynoos, R. S., Rodriguez, N., Steinberg, A., Stuber, M. L., & Frederick, C. (1998). The UCLA PTSD Reaction Index for DSM-IV (Revision 1). Los Angeles: UCLA Trauma Psychiatry Program.



ARTICLE IN PRESS

12


Rosner, R., Powell, S., & Butollo, W. (2003). Posttraumatic stress disorder three years after the siege in Sarajevo. Journal of Clinical Psychology, 59, 41–55. Saile, R., Neuner, F., & Catani, C. (2014). Children of the post-war years: A two-generational multilevel risk assessment of child psychopathology in Northern Uganda. (submitted for publication). Scheeringa, M. S., & Zeanah, C. H. (2001). A relational perspective of PTSD in early childhood. Journal of Traumatic Stress, 14(4), 799–815. Scholte, W. F., Olff, M., Ventevogel, P., de Vries, G. J., Jansveld, E., Cardozo, B. L., & Crawford, C. A. (2004). Mental health symptoms following war and repression in Eastern Afghanistan. Journal of the American Medical Association, 292, 585–593. Sriskandarajah, V., Neuner, F., & Catani, C. (2014). Parental care moderates the association between trauma and mental health in Sri Lankan children (submitted for publication). Steinberg, A. M., Brymer, M. J., Decker, K. B., & Pynoos, R. S. (2004). The University of California at Los Angeles Posttraumatic Stress Reaction Index. Current Psychiatry Reports, 6, 96–100. Suliman, S., Kaminer, D., & Seedat, S. (2005). Assessment techniques and South African community studies of trauma and posttraumatic stress disorder on children and adolescents. Journal of Child and Adolescent Mental Health, 17, 55–62. Stoltzfus, J. C. (2011). Logistic regression: A brief primer. Research Methods and Statistics, 18(10), 1099–1104. Trickey, D., Siddaway, A. P., Meiser-Stedman, R., Serpell, L., & Field, A. P. (2012). A meta-analysis of risk factors for post-traumatic stress disorder in children and adolescents. Clinical Psychology Review, 32, 122–138. Udwin, O., Boyle, S., Yule, W., Bolton, D., & O’Ryan, D. (2000). Risk factors for long-term psychological effects of a disaster experienced in adolescence: Predictors of posttraumatic stress disorder. Journal of Child Psychology and Psychiatry, 41(8), 969–979. United States Department of State, Bureau of Diplomatic Security. (2012). Mexico 2012 OSAC crime and safety report: Monterrey. Retrieved from: https://www.osac.gov/pages/contentreportdetails.aspx?cid=11962 Wilt, S., & Olson, S. (1996). Prevalence of domestic violence in the United States. Journal of the American Medical Women’s Association, 51, 77–82.


Family interaction and child outcome in a high-risk sample.

Subtypes of exposure to intimate partner violence within a Canadian child welfare sample: associated risks and child maladjustment.

Academic Failure and Child-to-Parent Violence: Family Protective Factors.

Risk factors for PTSD and other diagnoses in a general sample of Vietnam veterans.

Family environment factors associated with child abuse.

Child sex tourism - prevalence of and risk factors for its use in a German community sample.

Symptoms of PTSD in a sample of female victims of sexual violence in post-earthquake Haiti.

Child victims and poly-victims in China: are they more at-risk of family violence?

Risk factors for premature termination of treatment at a child and family mental health clinic.

Risk factors for the development of PTSD and depression among child and adolescent victims following a 7.4 magnitude earthquake.

A National Study on Nurses' Exposure to Occupational Violence in Lebanon: Prevalence, Consequences and Associated Factors.

Exposure therapy for PTSD.

Sunlight exposure and cardiovascular risk factors in the REGARDS study: a cross-sectional split-sample analysis.

Child exposure to parental violence and psychological distress associated with delayed milestones.

Prevalence of Family Violence and Associated Factors Among In-School Adolescents in São Paulo, SP, Brazil.

Community violence exposure and severe posttraumatic stress in suburban American youth: risk and protective factors.

[Intimate partner violence and family dysfunction among Mexican women seen a Primary Care Unit].

Association between intimate partner violence and child maltreatment in a representative student sample in Hong Kong.

Examining cumulative victimization, community violence exposure, and stigma as contributors to PTSD symptoms among high-risk young women.

Family factors affecting child development.

Child Problems as a Moderator of Relations Between Maternal Impulsivity and Family Environment in a High-Risk Sample.

Risk factors for severe intimate partner violence and violence-related injuries among women in India.

Childrearing Violence and Child Adjustment Following Exposure to Kenyan Post-election Violence.

Exposure to Violence and Carotid Artery Intima-Media Thickness in Mexican Women.