Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Jan 1.
Published in final edited form as: Vulnerable Child Youth Stud. 2013 Nov 6;9(3):220–233. doi: 10.1080/17450128.2013.855346

Child and Caregiver Concordance of Potentially Traumatic Events Experienced by Orphaned and Abandoned Children

Divya Guru Rajan 1, Kristen Shirey 2, Jan Ostermann 3, Rachel Whetten 4, Karen O’Donnell 5, Kathryn Whetten 6
PMCID: PMC4217223  NIHMSID: NIHMS537959  PMID: 25379051

Abstract

Exposure to trauma is associated with significant emotional and behavioral difficulties among children (Perepletchikova & Kaufman, 2010). Overall, reports of trauma and violence experienced by children are discrepant from those of their caregivers (Lewis et al., 2012). Even less is known about the extent of concordance between orphans and their caregivers. This study examines the correlates of concordance in reported traumatic experiences between 1,269 orphaned and abandoned children (OAC) and their caregivers. The OAC lived in family-settings in 5 low and middle income countries and were part of a longitudinal study, “Positive Outcomes for Orphans” (POFO) that enrolled children aged 6 to 12 at baseline. By examining concordance with respect to specific types of trauma reported, this study expands the understanding of who reports which types of traumas experienced by orphaned and abandoned children, thereby improving the potential to provide targeted interventions for children who have experienced such events. In this study, children and caregivers were asked separately if the child had experienced different types of potentially traumatic events. Children were significantly more likely to report physical abuse, sexual abuse and family violence than were caregivers. Caregivers were significantly more likely than children to report natural disasters and accidents. High levels of concordance were found in the reporting of wars, riots, killings, and deaths in the family.

The impacts of trauma on behavior and mental health are profound, and highly effective interventions targeting sequelae of childhood trauma are currently being developed for use in low resource areas. Findings from this study demonstrate that it is feasible to conduct screening for potentially traumatic events utilizing child self-report in resource limited settings and that child self-report is crucial in evaluating trauma, particularly family violence and physical or sexual assault.

Keywords: concordance, trauma, orphans, low-income countries, POFO

Background

Childhood exposure to trauma creates profound emotional and behavioral sequelae and impacts overall well-being (Perepletchikova & Kaufman, 2010). Recent studies of orphaned and abandoned children in low-and middle income countries (LMICs) demonstrate that orphaned children are at higher risk than those not orphaned for experiencing potentially traumatic events (PTEs) due to lack of adequate adult protection (Ahmad, et al., 2005; Cluver, Fincham & Seedat, 2009; Cluver & Gardner, 2006; Cluver, Gardner & Operario, 2007).

Detecting children’s exposure to different types of traumatic events poses a significant clinical challenge, as there are marked discrepancies between reports of children and their caregivers (Lewis et al., 2012; Yule & Canterbury, 1994). Caregivers may significantly under-report the behavioral effects of such trauma (Almqvist and Brandell-Forsberg, 1997). Obtaining accurate reports of exposure among orphans and vulnerable children in low-resource settings is an even greater challenge, as little is known about discrepancies between reports of caregivers and orphans in these settings.

In the Positive Outcomes for Orphans (POFO) study, Whetten et al (2011) describe rates of potentially traumatic events (PTEs) and associated emotional and behavioral difficulties among orphaned and abandoned children (OAC) in five LMICs. Ninety-eight percent of children surveyed reported having experienced PTEs in addition to the loss of a parent or abandonment, and more than half had experienced four or more such events. Higher numbers of PTEs were linked to statistically significant increases in emotional and behavioral difficulties. Such difficulties in children can make advancement in school, creation of positive social networks, and employment more challenging (Rapport et al, 2001).

Screening for a condition is generally recommended when there is a reasonable likelihood that the condition screened for will result in negative outcomes, is highly prevalent in a population, and if care alternatives are available (Raffle & Gray, 2007). Given the detrimental impact of PTEs on children’s well-being, mental health, and future ability to succeed (Schilling et al., 2007, Spertus et al., 2003), that OACs are particularly vulnerable to exposure to PTEs (Whetten et al., 2011), and that effective, feasible treatment options for mental health sequelae of PTEs are being developed in and for low-resource settings (Gupta & Zimmer 2008, Ertl et al., 2011, Bolton et al., 2003, Bolton et al., 2007), developing accurate screening techniques for trauma exposure in these populations is critical. In addition to enhancing the ability to screen and enroll children into programs that treat the anxiety and depression that may result from PTEs, screening at the individual and population levels may also result in primary prevention by identifying high risk families and/or communities in need of intervention or child protection. Effective treatment, such as group interpersonal psychotherapy or cognitive behavioral therapy (CBT), tested among trauma-exposed children (Patel et al., 2010; Bolton et al., 2003; Bolton et al., 2007; Patel et al., 2007; Layne et al., 2008) in LMICs relies on having knowledge of the potentially traumatic events to which the child was exposed.

Traditionally, caregiver report has been the primary source of information regarding a child’s exposure to PTEs. Including reports from children rather than relying on reports from caregivers alone yields a more comprehensive assessment of children’s exposure to violence (Kolko, Kazdin & Day 1996). Previous studies evaluating concordance between caregivers and children have found discrepancies in reports of domestic violence (Knutson et al., 2009, Thomson et al., 2002); the child’s age and gender are significant associated with greater reporting of PTEs by children (Kuo et al., 2000). The clinical significance of this study is that it evaluates variability in concordance patterns between caregiver and child reports in a multi-year study of vulnerable children situated in low resource settings.

Study Participants

Details of the POFO study’s sampling strategy and the demographic characteristics of participating caregivers and children have been reported elsewhere (The Positive Outcomes for Orphans Research Team, 2010; Whetten et al., 2009). Here, we describe elements of the sampling approach most germane to this analysis.

Six sites were chosen: Battambang District, Cambodia; Addis Ababa, Ethiopia; Hyderabad, India; Dimapur and Kohima Districts in Nagaland, India; Bungoma District, Kenya; and Kilimanjaro Region, Tanzania. A two-stage random sampling survey methodology was employed to obtain a sample of 1,480 OAC aged 6–12 at baseline who resided in family dwellings. The definition of an orphaned child for this study included both single and double orphans, consistent with the definition developed and used by UNICEF and other global policy makers (UNICEF, 2012). To be considered abandoned for purposes of this study, children had to have been left by both parents with caregivers reporting that they had no knowledge of the parents’ whereabouts and had no expectation of parental return. The sampling strategy in each study area involved the selection of 50 sampling areas (“clusters”) with five OAC randomly selected per cluster. Geographic or administrative boundaries were used to define sampling areas. Up to five eligible children were selected in each cluster either at random, from available lists, or through a household census. In households with multiple age-eligible children, the child whose first name started with the earliest letter of the alphabet was selected.

To provide a qualitative reference group, 301 children who were not orphaned or abandoned at baseline were enrolled into the study. For each cluster of five OAC, one household without OAC was randomly selected from which a dyad of one child aged 6–12 and his/her primary caregiver was enrolled. At each site, study enrollment and data collection were conducted over a 6 to 8 month period between May 2006 and February 2008.

Procedures

Every enrolled child and self-identified primary caregiver dyad was contacted twice yearly for a period of up to 3 years. Children’s exposure to PTEs were assessed annually by caregiver report, and, starting from age 10, also by child self-report. The primary source of information for this manuscript is caregiver and child information from the 3-year follow-up assessment.

Measures

The measures utilized in this study have been validated in numerous countries and cultural contexts. Focus groups and pilot interviews were conducted to test measures in all study settings to ensure that concepts were understood similarly across sites. Assessment tools were translated and back-translated, and were field-tested by trained local interviewers fluent in both English and the local language.

Potentially Traumatic Events

The Life Events Checklist (LEC) is a list of experienced or witnessed events that have been found to be predictive of the diagnoses of posttraumatic stress disorder (PTSD), anxiety, and depression (Gray et al., 2004). The measure was created by the National Center for PTSD to facilitate the diagnosis of PTSD (Gray, et al., 2004) and is one of the most frequently used instruments by researchers for PTE exposure (Elhai et al., 2005). The LEC assesses exposure to potentially traumatic events including natural disasters, witnessing someone hurt or killed, experiencing physical or sexual assault, or being forced to leave home. Caregivers and children ages 10 and older were asked if the child had ever witnessed or experienced each type of event, and, if yes, whether the event had occurred more than once. On follow-up, an additional question was included about whether the event(s) happened more than one year ago, less than one year ago, or both. Children were reported as having experienced a PTE if either the caregiver or child reported a PTE at any point prior to or during the 3-year study period.

One PTE, “hearing about a family member who died,” was not included in this analysis because a positive response was nearly ubiquitous among the children surveyed. The item “had a brother or sister die” was also excluded because it could not be determined whether the event had been witnessed by the child. There were 19 remaining PTEs which were grouped into 6 conceptually related clusters for analysis: witnessing family death; physical or sexual abuse; witnessing family violence; being forced to leave home; war, riots or killings; and disasters or accidents. PTEs have been grouped similarly in other studies of childhood and adult PTEs (Mugavero et al., 2006). In addition, parental death and abandonment were included as separate PTE categories. As study inclusion criteria included having experienced parental death or abandonment, all OAC included in this study necessarily had experienced at least one PTE.

Wealth Index

The World Bank’s Child Needs Assessment (CNA) Toolkit (World Bank, 2008) was used to assess household characteristics. This measure was administered to all participating caregivers at baseline, and selected elements were asked at annual follow-up assessments. In addition, on at least one of the annual follow-up assessments an asset checklist, several elements of which have been used by Demographic and Health Surveys (DHS), conducted in over 90 countries, in the construction of Wealth Indices (Rutstein & Johnson, 2004), was administered to all caregivers. Using elements of the asset checklist and the CNA Toolkit that were comparable to variables used in the construction of DHS wealth indices, a wealth index was constructed for each participating POFO household (The Positive Outcomes for Orphans Research Team, 2010).

Analysis

Wilcoxon signed rank tests for paired samples were used to determine the statistical significance of discordance between the reporting of potentially traumatic events by children and caregivers. Bivariate probit models were constructed to investigate the predictors of reports of different types of PTE by OAC and their caregivers. The extent of correlation between the two outcomes is given by the correlation coefficient, ‘rho’, with the associated Wald test statistic testing the null hypothesis of zero correlation between the two reports. Robust clustered (at the site level) standard errors were used in all regressions. All models included site fixed effects.

Regression models were conducted for four of the five trauma categories: disasters and accidents, war and riots, physical and sexual abuse, family violence, and family death. The category ‘forced to leave home’ was not included as it was commonly experienced by children across the study sites and therefore provided insufficient variation. We present only models for PTEs in which significant discordance was observed.

Results

The gender distribution of the children is 46% female and 54% male participants (Table 1). Nearly one-fifth (19%) of the children are double orphans while more than half are paternal orphans (55%). Grandparents care for nearly one-fifth of the children (19%), and other relatives and non-relatives care for slightly more than one-third (36%). The vast majority of caregivers are female (85%), and most are widowed (58%). The sample is predominantly (69%) urban living (Table 1).

Table 1.

Demographic Characteristics

Total Sample Size of Child-Caregiver Pairs 1269
Child Gender %
 Percent Female 46.49%
 Percent Male 53.51%
Child Orphan Status
 No Dead Parent 10.01%
 Maternal Orphan 16.00%
 Paternal Orphan 55.08%
 Both Parents Dead 18.91%
Child and Caregiver Relationship
 Biological Parent 46.89%
 Grandparent 17.10%
 Other 36.01%
Caregiver’s Gender
 Percent Female 85.42%
 Percent Male 14.58%
Caregiver Marital Status
 Married 24.27%
 Widowed 57.68%
 Other 18.05%
Residential Location
 Rural 31.36%
 Urban 68.56%

The overall prevalence of having reported experiencing physical or sexual abuse is 26%; severe family violence is 16%; war, riots or killings is 17%, family death other than the death of the parent is 57%; and natural disasters or severe accidents is 14% (Table 2). Examining differences in mere proportions of reporting without accounting for the effect of demographic characteristics, it appears children are more likely to report physical or sexual abuse (p=0.0273) while caregivers are more likely to report disasters or accidents (p=0.003) and family death (p=0.0001).

Table 2.

Frequency of Reports of Trauma (3-year follow-up) by Report Type (N=1269 child-caregiver pairs)

Trauma Type Either the caregiver or child or both reported PTE Trauma incidencea Both Reported PTE Caregiver only Reported PTE % All Trauma Reported by Caregiverb Child only Reported PTE % All Trauma Reported by Childc Z-score from Wilcoxon Signed Rank Testd p-value of Z-score
Physical or Sexual Abuse 326 26% 60 115 54% 151 65% −2.207* 0.027
Family violence 208 16% 34 81 55% 93 61% −0.910 0.363
War, Riots, Killings 210 17% 108 44 72% 58 79% −1.386 0.166
Family death 726 57% 371 214 81% 141 71% 3.874** 0.000
Disasters or Accidents 180 14% 39 92 73% 49 49% 3.621** 0.003
a

Trauma Incidence = (Both Report Trauma + Report by Caregiver Only + Report by Child Only) / Total Sample Size of Child-Caregiver Pairs

b

% All Trauma Reported by Caregiver = (Both Report Trauma + Report by Caregiver Only) / (Both Report Trauma + Report by Caregiver Only + Report by Child Only)

c

% All Trauma Reported by Child = (Both Report Trauma + Report by Child Only) / (Both Report Trauma + Report by Caregiver Only + Report by Child Only)

d

Null Hypothesis Test of Caregiver Report = Child Report

*

p<0.05.

**

p<0.01

We further examine the differences in reporting through regression models that take into account the effect of demographic characteristics. Table 3 and Table 4 provide the primary results of the bivariate probit regression models for those categories of PTEs where significant discordance was found either in the descriptive statistics or regressions. Table 3 illustrates that the older the child, the less likely the caregiver is to report incident physical or sexual abuse (p <0.05). Further, the wealthier the household, the less likely the child is to report physical or sexual abuse that had been reported by the caregiver (p<0.001). The caregiver is less likely to report female and older children witnessing family violence (p<0.05; p<0.001) and when they are the grandparent (p<0.001) or widowed (p<0.05). When the caregiver is not related to the child, the child reports significantly less family violence (p<0.05).

Table 3.

Bivariate Probit Models of Cumulative Trauma: 3-year follow-up

Variable Physical or Sexual Abuse Family Violence
Caregiver Report Self Report Chi-Sq Testa Caregiver Report Self Report Chi-Sq Testa
Child is Female Coefficient (SE) Chi-Sq Coefficient (SE) Chi-Sq
−0.134 (0.145) −0.127 (0.147) 0.01 −0.225* (0.094) −0.153 (0.096) 0.65
Age of child −0.058* (0.025) 0.035 (0.035) 3.73* −0.091*** (0.021) 0.055 (0.029) 47.98***
Caregiver is Grandparent (vs. Biological Parent) −0.123 (0.097) 0.071 (0.153) 1.11 −0.581*** (0.160) −0.097 (0.109) 3.8*
Caregiver is Other (not a grandparent and not a biological parent vs. Biological Parent) −0.116 (0.219) −0.124 (0.249) 0.00 −0.353 (0.333) −0.507* (0.229) 0.12
Orphan Status: No Dead Parent (vs. Both Parents Dead) −0.074 (0.129) −0.069 (0.181) 0.00 0.124 (0.166) −0.129 (0.081) 2.24
Orphan Status: Maternal Orphan (vs. Both Parents Dead) 0.003 (0.099) −0.079 (0.223) 0.19 0.084 (0.226) 0.023 (0.186) 0.04
Orphan Status: Paternal Orphan (vs. Both Parents Dead) −0.342 (0.254) −0.164 (0.359) 0.43 0.478 (0.401) 0.212 (0.351) 0.59
Caregiver is Female −0.040 (0.068) −0.007 (0.134) 0.04 0.194 (0.141) −0.131 (0.126) 8.24***
Caregiver is Widowed (vs. Married) −0.113 (0.151) −0.094 (0.156) 0.01 −0.269* (0.132) −0.176 (0.164) 0.12
Caregiver is Any Other Marital Status (vs. Married) 0.073 (0.143) 0.154 (0.145) 0.10 0.095 (0.143) −0.279 (0.201) 4.18**
Age of caregiver 0.0002 (0.003) −0.0001 (0.004) 0.01 0.004 (0.004) −0.011 (0.007) 2.8*
Wealth Index 0.182 (0.197) −0.140*** (0.041) 2.42 0.152 (0.147) −0.022 (0.075) 1.10
Urban Location 0.154 (0.245) −0.109 (0.119) 0.76 0.135 (0.100) −0.306 (0.208) 5.75**
Site Fixed Effects Yes Yes Yes Yes
Log Pseudo-Likelihood −816.84 −589.49
Rho 0.12 0.28
Wald (Chi-sq)
Test Rho = 0
0.46 3.86
N 1025 1025

Robust clustered standard errors in parentheses;

*

p<0.05,

**

p<0.01,

***

p<0.001

a

Test statistic of equality of coefficients across models

Table 4.

Bivariate Probit Models of Cumulative Trauma: 3-year follow-up

Variable Disasters or Accidents Family Death
Caregiver Report Self Report Chi-Sq Testa Caregiver Report Self Report Chi-Sq Testa
Child is Female Coefficient (SE) Chi-Sq Coefficient (SE) Chi-Sq
−0.112 (0.091) −0.088 (0.108) 0.07 −0.070 (0.083) 0.064 (0.046) 5.48**
Age of child −0.094*** (0.027) 0.046 (0.024) 15.24*** −0.009 (0.019) 0.062*** (0.013) 11.43***
Caregiver is Grandparent (vs. Biological Parent) 0.014 (0.109) 0.008 (0.274) 0.00 −0.071 (0.150) −0.192 (0.156) 5.35**
Caregiver is Other (not a grandparent and not a biological parent vs. Biological Parent) −0.164 (0.241) −0.513*** (0.127) 1.74 −0.233 (0.137) −0.172 (0.122) 0.12
Orphan Status: No Dead Parent (vs. Both Parents Dead) 0.058 (0.178) −0.106 (0.107) 0.51 0.979*** (0.101) 0.477 (0.297) 2.53
Orphan Status: Maternal Orphan (vs. Both Parents Dead) 0.131 (0.166) −0.112 (0.231) 1.56 0.884*** (0.229) 0.525 (0.288) 1.54
Orphan Status: Paternal Orphan (vs. Both Parents Dead) 0.0560 (0.274) 0.021 (0.170) 0.01 0.385* (0.188) 0.365 (0.288) 0.01
Caregiver is Female −0.002 (0.162) −0.153 (0.163) 0.78 −0.100 (0.213) −0.409*** (0.099) 2.62
Caregiver is Widowed (vs. Married) −0.073 (0.062) −0.310** (0.095) 10.98*** −0.176 (0.139) −0.0223 (0.160) 0.58
Caregiver is Any Other Marital Status (vs. Married) 0.082 (0.128) −0.613** (0.195) 9.52*** −0.139 (0.077) −0.208 (0.177) 0.18
Age of caregiver −0.002 (0.006) −0.017 (0.010) 2.51 −0.002 (0.005) −0.007 (0.005) 1.85
Wealth Index 0.003 (0.078) −0.239 (0.127) 2.71 −0.069 (0.058) −0.0307 (0.084) 0.16
Urban Location 0.053 (0.126) −0.195 (0.131) 2.22 0.118 (0.069) −0.255 (0.150) 4.07**
Site Fixed Effects Yes Yes Yes Yes
Log Pseudo-Likelihood −584.44 −1235.40
Rho 0.58 0.44
Wald (Chi-sq) Test Rho = 0 65.94*** 40.58***
N 1025 1025

Robust clustered standard errors in parentheses;

*

p<0.05,

**

p<0.01,

***

p<0.001

a

Test statistic of equality of coefficients across models

Caregivers are significantly more likely to report incident disasters and accidents, (Table 4) and this difference is larger when the caregiver is not related to the child (p <0.01), widowed (p <0.01), or of “other” marital status (e.g. never married or divorced; p <0.01).

In the categories of disasters and accidents (Correlation statistic =0.58; Wald X2 statistic = 65.94) and family death (Correlation statistic =0.44; Wald X2 statistic = 40.58), reports of trauma by the caregiver and the child are highly correlated (Table 4). The correlations for disasters and accidents and family death are contrary to the results from the descriptive statistics shown in Table 2, which compares PTEs reported by children and their caregivers without controlling for the effect of demographic variables.

In the category of family violence, contrary to the high correlation in Table 2, there appears to be significantly less correlation (Correlation statistic =0.28; Wald X2 statistic = 3.86; p = 0.049) after controlling for demographic characteristic (Table 3). In the physical or sexual abuse trauma category (Table 3) there is low correlation (Correlation statistic =0.12; Wald X2 statistic = 0.46) between the reports of PTEs by children and their caregivers after controlling for demographic characteristics in the regression (as seen in Table 2). Among predictors of reports of family violence (Table 5 provides a summary of the test results from Table 3), the effect of the gender of the caregiver, marital status of the caregiver, the age of the caregiver, and the household’s location in a rural vs. urban areas differ between the two types of reports.

Table 5.

Comparison of Predictors Across Bivariate Probit Models (in Table 3)

Variable Physical or Sexual Abuse Family Violence
Association with… Consistency of associations across reports Association with… Consistency of associations across reports
Caregiver report Child report Caregiver report Child report
Child Characteristics
Female gender Null Null Consistent Negative Null Consistent
Age of child Negative Null Inconsistent Negative Null Inconsistent
Caregiver is Grandparent (vs. Biological Parent) Null Null Consistent Negative Null Inconsistent
Caregiver is Other (not a grandparent and not a biological parent vs. Biological Parent) Null Null Consistent Null Negative Consistent
Caregiver Characteristics
Caregiver is Female Null Null Consistent Null Null Inconsistent
Caregiver is Widowed (vs. Married) Null Null Consistent Negative Null Consistent
Caregiver is Any Other Marital Status (vs. Married) Null Null Consistent Null Null Inconsistent
Age of caregiver Null Null Consistent Null Null Inconsistent
Household Characteristics
Wealth Index Null Negative Consistent Null Null Consistent
Urban Location Null Null Consistent Null Null Inconsistent

In the category of war, riots or killings (Table 6), there is significant concordance (Correlation statistic = .59; Wald X2 statistic = 58.59; p < 0.001). Caregivers who were older, or were taking care of older children were less likely to report such PTEs while children living in urban areas and with caregivers who are not married and not widowed are less likely to report such PTEs.

Table 6.

Bivariate Probit Models of Cumulative Trauma: 3-year follow-up

Variable War, Riots or Killings
Caregiver Report Self Report Chi-Sq Testa
Child is Female Coefficient (SE) Chi-Sq
0.170 (0.238) −0.036 (0.096) 1.46
Age of child −0.074** (0.0223) 0.021 (0.033) 3.06*
Caregiver is Grandparent (vs. Biological Parent) 0.322* (0.138) 0.014 (0.173) 3.87**
Caregiver is Other (not a grandparent and not a biological parent vs. Biological Parent) 0.053 (0.059) −0.432* (0.190) 4.04**
Orphan Status: No Dead Parent (vs. Both Parents Dead) 0.188 (0.343) 0.143 (0.078) 0.02
Orphan Status: Maternal Orphan (vs. Both Parents Dead) −0.044 (0.266) −0.180 (0.219) 0.09
Orphan Status: Paternal Orphan (vs. Both Parents Dead) −0.185 (0.333) −0.035 (0.188) 0.12
Caregiver is Female −0.0004 (0.178) −0.155 (0.255) 0.22
Caregiver is Widowed (vs. Married) 0.324* (0.128) −0.231 (0.196) 5.53**
Caregiver is Any Other Marital Status (vs. Married) 0.202 (0.186) −0.700*** (0.149) 8.26**
Age of caregiver −0.018** (0.006) −0.013 (0.007) 0.22
Wealth Index −0.213 (0.133) −0.106 (0.111) 0.32
Urban Location −0.273* (0.117) −0.370** (0.129) 1.63
Site Fixed Effects Yes Yes
Log Pseudo-Likelihood −430.03
Rho 0.59
Wald (Chi-sq) Test Rho = 0 58.59***
N 1025

Robust clustered standard errors in parentheses;

*

p<0.05,

**

p<0.01,

***

p<0.001

a

Test statistic of equality of coefficients across models

Sensitivity Analyses

To check the sensitivity of estimates from the main regression, the same model was run after restricting the data in the following ways: a) With only previous rounds of interviews that included the full set of trauma-related questions, i.e. 1-year follow-up and 2-year follow-up interviews; (b) First-time interviewees in the 1-year and 3-year follow-up rounds; and c) Interviewees who were part of both the 1-year follow-up and the 3-year follow-up.

Sensitivity analyses (results available on request) indicate that the significance of predictors remains consistent across rounds for each trauma category and the extent of concordance across trauma categories is consistent with that found in the main regression results. The correlation coefficients from regressions with different samples at the 3-year follow-up are reported for each trauma category in Table 7.

Table 7.

Correlation Coefficients by Model at the 3-year follow-up

Model Type Disasters / Accidents War / Riots / Killings Physical / Sexual Abuse Family Violence Family Death
Full Sample .58** .59** 0.12 .28* .44**
First-time Interviews .57** .55** 0.03 0.16 .4**
Interviewed in both 1-year & 3-year follow-ups .55** .56** 0.11 0.17 .51**
*

p<0.05.

**

p<0.01

Discussion

This study demonstrates that concordance between childrens’ and caregivers’ reports are significantly lower in certain types of PTEs – those that may either be associated with fear or shame responses on the part of the caregiver (i.e., physical / sexual abuse and witnessing family violence).

In the simple descriptive analyses in Table 1 where we do not adjust for different demographic characteristics, we find that there are larger differences in the reports of PTEs related to witnessing disasters and accidents or family deaths. However, after adjusting for demographic characteristics and obtaining regression-based measures of concordance (as seen in Table 4), we do not find any significant differences in the reports by children and caregivers of PTEs related to witnessing disasters and accidents or family deaths. The difference between child and caregiver in the reports of these PTEs (of witnessing disasters and accidents or family deaths) seen in the initial descriptive analyses in Table 1 is then significantly associated with specific demographic characteristics after controlling for which, the differences are no longer present. The relevant demographic characteristics are those for which the coefficients of are significantly different between the two types of reports in Table 4 (columns with the Chi-sq test results) – specifically, they are the age and gender of the child, the caregiver’s relationship to the child, the caregiver’s marital status (widowed and other) and urban location of residence.

Since we do not assess whether the PTEs occurred at the current location of care or in a previous setting, it is possible that current caregivers are simply unaware of some PTEs. This then may also contribute to lower levels of concordance for PTEs. Consequently, obtaining reports from multiple sources, especially directly from children, regarding exposure to specific types of trauma is crucial from a policy and programmatic standpoint. Through this study, we demonstrate that conducting such interviews is feasible across a variety of settings in low and middle income countries that vary widely with respect to cultural, political, and historical backgrounds.

The child’s age, gender and the marital status of the caregiver are significant predictors of differences between reports of PTEs of physical / sexual abuse and family violence by the child and the caregiver. These findings point to the need to target older children and female children in screening for exposure to PTEs and to develop improved opportunities for OAC to disclose PTE exposure. Larger-scale screening may identify more OACs at risk for sequelae of trauma including emotional and behavioral difficulties. Conducing screening on a large scale without targeting specific demographic groups may help to reduce stigma by normalizing the disclosure of PTE exposure and may reduce some of the shame and fear associated with disclosure.

There are several limitations of this study. Reporting bias of PTEs is likely in the direction of underreporting, which could alter the estimated effects of PTEs on emotional and behavioral difficulties and overestimate the concordance between child and caregiver. While exposure to different types of PTEs was assessed, this study does not look at the total number of PTEs children were exposed to and whether this has a relationship with disclosure or concordance of reporting between child and caregiver. With respect to traumatic events that typically occur in the home environment including physical or sexual abuse and family violence, we did not assess if these exposures occurred in the current care setting or in previous care settings. Whether the child was living with the current caregiver would certainly impact the concordance in trauma reporting, either positively as the caregiver is aware of PTEs in the child’s life, or negatively if there is abuse or family violence in the care setting potentially involving the caregiver and associated with shame and fear of disclosure.

Orphanhood is associated with serious mental health sequelae including anxiety, depression, behavioral difficulties, and suicidal ideation. OACs are at increased risk of exposure to additional PTEs, which further heighten the risk of mental health symptomatology. Psychosocial interventions to protect OACs from further exposure to PTEs and mental health care for those exposed to PTEs may protect these children from further PTEs and attenuate the effects of trauma on emotional well-being, affording them more opportunities to succeed in school, work, and relationships, and in becoming healthy and productive adults. As such interventions are being developed it is important to ensure that PTE exposure is accurately assessed by incorporating caregivers’ PTE reports but also providing children with the opportunity to report certain types of trauma in the absence of their caregiver. This study demonstrates that children are able to report these PTEs in a wide variety of settings and are willing to do so when asked by trained interviewers.

Footnotes

*

In addition to the listed authors, the POFO Team consists of Chris Bernard Agala, Rajeswara Rao Konjarla, Neville Selhore, Frehiwot Alebachew, Mao Lang, John Shao, Sisay W. Ameya, Dean Lewis, Amani Sizya, Robin Briggs, Ira Madan, Vanroth Vann, Sopheak Chan, Cyrilla Manya, Augustine Wasonga, Haimanot Diro, Restituta Mvungi, Belaynesh Engadawork, Laura K. Murphy-McMillan, Dafrosa Itemba, Agnes Ngowi, Venkata Gopala Krishna Kaza, Imliyanger Pongen, Becky Kinoti, Pelevinuo Rai. This work was supported by the National Institute of Child Health and Development (NICHD), grant No. 5R01HD046345-04. We thank all the children and caregivers who participated in this study. We appreciate the support that has been provided by the partner organizations: ACE Africa, Bungoma, Kenya; Homeland Meahto Phum Ko’Mah in Battambang, Cambodia; KIWAKKUKI and TAWREF, Moshi, Tanzania; SaveLives Ethiopia and Stand for the Vulnerable Organization, Addis Ababa, Ethiopia; and Sahara Centre for Rehabilitation and Residential Care in Delhi, Hyderabad and Nagaland, India.

Disclaimer

The project described was supported by grant # 5R01HD046345 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD). The content is solely the responsibility of the authors and does not necessarily represent the official views of the Eunice Kennedy Shriver National Institute of Child Health and Human Development or the National Institutes of Health.

Contributor Information

Divya Guru Rajan, Sanford School of Public Policy, Duke University, Durham, North Carolina, United States of America.

Kristen Shirey, Center for Health Policy & Inequalities Research, Duke Global Health Institute, Departments of Psychiatry and Medicine, Duke University, Durham, North Carolina, United States of America.

Jan Ostermann, Center for Health Policy & Inequalities Research, Duke Global Health Institute, Duke University, Durham, North Carolina, United States of America.

Rachel Whetten, Center for Health Policy & Inequalities Research, Duke University, Durham, North Carolina, United States of America.

Karen O’Donnell, Center for Health Policy & Inequalities Research, Departments of Psychiatry and Pediatrics, Center for Child and Family Health, Duke University, Durham, North Carolina, United States of America.

Kathryn Whetten, Center for Health Policy & Inequalities Research, Duke Global Health Institute, Sanford School of Public Policy, Duke University, Durham, North Carolina, United States of America.

References

  1. Ahmad A, Qahar J, Siddiq A, Majeed A, Rasheed J, Jabar F, et al. A 2-year follow-up of orphans’ competence, socioemotional problems and post-traumatic stress symptoms in traditional foster care and orphanages in Iraqi Kurdistan. Child Care Health Dev. 2005;31(2):203–215. doi: 10.1111/j.1365-2214.2004.00477.x. [DOI] [PubMed] [Google Scholar]
  2. Almqvist K, Brandell-Forsberg M. Refugee children in Sweden: Post-traumatic stress disorder in Iranian pre-school children exposed to organized violence. Child Abuse and Neglect. 1997;24(1):351–366. doi: 10.1016/s0145-2134(96)00176-7. [DOI] [PubMed] [Google Scholar]
  3. Blanchard EB, Jones-Alexander J, Buckley TC, Forneris CA. Psychometric properties of the PTSD Checklist (PCL) Behav Res Ther. 1996;34(8):669–673. doi: 10.1016/0005-7967(96)00033-2. [DOI] [PubMed] [Google Scholar]
  4. Bolton P, Bass J, Neugebauer R, Clougherty K, Verdeli H, Ndogoni L, Wickramaratne P, Weissman M. Results of a clinical trial of a group intervention for depression in rural Uganda. Journal of the American Medical Association. 2003;279:3117–3124. [Google Scholar]
  5. Bolton P, Bass J, Betancourt T, Speelman L, Neugebauer R, Murray L, Onyango G, Clougherty K, Verdeli H. Interventions for depressions symptoms among adolescent survivors of war and displacement in northern Uganda: A randomized controlled trial. Journal of the American Medical Association. 2007;298:519–527. doi: 10.1001/jama.298.5.519. [DOI] [PubMed] [Google Scholar]
  6. Cluver L, Fincham DS, Seedat S. Posttraumatic stress in AIDS-orphaned children exposed to high levels of trauma: the protective role of perceived social support. Journal of Traumatic Stress. 2009;22(2):106–112. doi: 10.1002/jts.20396. [DOI] [PubMed] [Google Scholar]
  7. Cluver L, Gardner F. The psychological well-being of children orphaned by AIDS in Cape Town, South Africa. Ann Gen Psychiatry. 2006;5:8. doi: 10.1186/1744-859X-5-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cluver L, Gardner F, Operario D. Psychological distress amongst AIDS-orphaned children in urban South Africa. Journal of Child Psychology and Psychiatry. 2007;48(8):755–763. doi: 10.1111/j.1469-7610.2007.01757.x. [DOI] [PubMed] [Google Scholar]
  9. Cluver L, Orkin M. Cumulative risk and AIDS-orphanhood: interactions of stigma, bullying and poverty on child mental health in South Africa. Soc Sci Med. 2009;69(8):1186–1193. doi: 10.1016/j.socscimed.2009.07.033. [DOI] [PubMed] [Google Scholar]
  10. Cluver LD, Gardner F, Operario D. Effects of stigma on the mental health of adolescents orphaned by AIDS. Journal of Adolescent Health. 2008;42(4):410–417. doi: 10.1016/j.jadohealth.2007.09.022. [DOI] [PubMed] [Google Scholar]
  11. Elhai JD, Gray MJ, Kashdan TB, Franklin CL. Which instruments are most commonly used to assess traumatic event exposure and posttraumatic effects?: A survey of traumatic stress professionals. Journal of Traumatic Stress. 2005;18(5):541–545. doi: 10.1002/jts.20062. [DOI] [PubMed] [Google Scholar]
  12. Ertl V, Pfeiffer A, Schauer E, Elbert T, Neuner F. Community-implemented trauma therapy for former child soldiers in Northern Uganda: a randomized controlled trial. Journal of the American Medical Association. 2011;3;306(5):503–12. doi: 10.1001/jama.2011.1060. [DOI] [PubMed] [Google Scholar]
  13. Foster G, Makufa C, Drew R, Mashumba S, Kambeu S. Perceptions of children and community members concerning the circumstances of orphans in rural Zimbabwe. AIDS Care. 1997;9(4):391–405. doi: 10.1080/713613166. [DOI] [PubMed] [Google Scholar]
  14. Convention on the Rights of the Child, Resolution 44/25, in accordance with article 49. C.F.R. (1989).
  15. Goodman R. The Strengths and Difficulties Questionnaire: a research note. J Child Psychol Psychiatry. 1997;38(5):581–586. doi: 10.1111/j.1469-7610.1997.tb01545.x. [DOI] [PubMed] [Google Scholar]
  16. Gray MJ, Litz BT, Hsu JL, Lombardo TW. Psychometric properties of the life events checklist. Assessment. 2004;11(4):330–341. doi: 10.1177/1073191104269954. [DOI] [PubMed] [Google Scholar]
  17. Gupta L, Zimmer C. Psychosocial intervention for war-affected children in Sierra Leone. British Journal of Psychiatry. 2008;192(3):212–6. doi: 10.1192/bjp.bp.107.038182. [DOI] [PubMed] [Google Scholar]
  18. Knutson J, Lawrence E, Taber S, Bank L, DeGarmon D. Assessing Children’s Exposure to Intimate Partner Violence. Clinical Child Family Psychology Review. 2009;12:157–173. doi: 10.1007/s10567-009-0048-1. [DOI] [PubMed] [Google Scholar]
  19. Koenig L, Doll L, O’Leary A, Pequegnat W, editors. From Child Sexual Abuse to Adult Sexual Risk: Trauma, Revictimization, and Intervention. Washington, D.C: American Psychological Association; 2004. [Google Scholar]
  20. Kolko D, Kazdin A, Day B. Children’s Perspectives in the Assessment of Family Violence: Psychometric Characteristics and Comparison to Parent Reports. Child Maltreatment. 1996;1(2):156–167. [Google Scholar]
  21. Kuo M, Mohler B, Raudenbush S, Earls F. Assessing Exposure to Violence Using Multiple Informants: Application of Hierarchical Linear Model. Journal of Child Psychology and Psychiatry. 2000;41(8):1049–1056. [PubMed] [Google Scholar]
  22. Layne CM, Saltzman WR, Poppleton L, Burlingame GM, Pasali A, Durakovic E. Effectiveness of a school-based group psychotherapy program for war-exposed adolescents: A randomized controlled trial. Journal of the American Academy of Child and Adolescent Psychiatry 2008. 2008;47:1048–1063. doi: 10.1097/CHI.0b013e31817eecae. [DOI] [PubMed] [Google Scholar]
  23. Lewis T, Thompson T, Kotch JB, Proctor L, Litrownik A, English D, Runyan D, Wiley T, Dubowitz H. Parent–youth discordance about youth-witnessed violence: Associations with trauma symptoms and service use in an at-risk sample. Child Abuse and Neglect. 2012;36(2012):790–797. doi: 10.1016/j.chiabu.2012.09.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lin KK, Sandler IN, Ayers TS, Wolchik SA, Luecken LJ. Resilience in parentally bereaved children and adolescents seeking preventive services. J Clin Child Adolesc Psychol. 2004;33(4):673–683. doi: 10.1207/s15374424jccp3304_3. [DOI] [PubMed] [Google Scholar]
  25. Maslow A. A theory of human motivation. Psychol Rev. 1943;50:370–396. [Google Scholar]
  26. Matashalanga NR. Mass orphanhood in the era of HIV/AIDS: Bold Support for alleviation of poverty and education may avert a social disaster. British Medical Journal. 2002;324(26):185–186. [Google Scholar]
  27. Mugavero M, Ostermann J, Whetten K, Leserman J, Swartz M, Stangl D, et al. Barriers to antiretroviral adherence: the importance of depression, abuse, and other traumatic events. AIDS Patient Care STDS. 2006;20(6):418–428. doi: 10.1089/apc.2006.20.418. [DOI] [PubMed] [Google Scholar]
  28. Ntozi JP, Lubaale YM, Nakanaabi IM. AIDS mortality in Uganda: circumstances, factors and impact of death. Health Transit Rev. 1997;7(Suppl):207–224. [PubMed] [Google Scholar]
  29. Patel V, Weiss HA, Chowdhary N, Naik S, Pednekar S, Chatterjee S, De Silva M, Bhat B, Araya R, King M, Simon G, Verdeli H, Kirkwood B. Effectiveness of an intervention led by lay health counsellors for depressive and anxiety disorders in primary care in Goa, India (MANAS): a cluster randomised controlled trial. The Lancet. 2010;376(9758):2086–2095. doi: 10.1016/S0140-6736(10)61508-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Patel V, Araya R, Chatteriee S, Chisholm D, Cohen A, De Silva M, Hosman C, McGuire H, Rojas G, van Ommeren M. Treatment and prevention of mental disorders in low income and middle-income countries. The Lancet. 2007;370(9591):991–1005. doi: 10.1016/S0140-6736(07)61240-9. [DOI] [PubMed] [Google Scholar]
  31. Perepletchikova F, Kaufman J. Emotional and behavioral sequelae of childhood maltreatment. Current Opinion in Pediatrics. 2010 Oct;22(5):610–5. doi: 10.1097/MOP.0b013e32833e148a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pianta RC. Child-Parent Relationship Scale. University of Virginia; 1992. [Google Scholar]
  33. Raffle AE, Gray JAM. Screening. Vol. 2007 Oxford University Press; 2007. [Google Scholar]
  34. Rapport MD, Denney CB, Chung KM, Hustace K. Internalizing behavior problems and scholastic achievement in children: Cognitive and behavioral pathways as mediators of outcome. Journal of Clinical Child Psychology 2001. 2001;30:536–551. doi: 10.1207/S15374424JCCP3004_10. [DOI] [PubMed] [Google Scholar]
  35. Rutstein SO, Johnson K. Unpublished manuscript. Calverton, Maryland: 2004. The DHS Wealth Index. [Google Scholar]
  36. Assistance for Orphans and Other Vulnerable Children in Developing Countries Act of 2005, Public Law 109-95 C.F.R. (2005).
  37. Schilling EA, Aseltine RH, Jr, Gore S. Adverse childhood experiences and mental health in young adults: a longitudinal survey. BMC Public Health. 2007;7:30. doi: 10.1186/1471-2458-7-30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sengendo J, Nambi J. The psychological effect of orphanhood: a study of orphans in Rakai district. Health Transit Rev. 1997;7(Suppl):105–124. [PubMed] [Google Scholar]
  39. Spertus IL, Yehuda R, Wong CM, Halligan S, Seremetis SV. Childhood emotional abuse and neglect as predictors of psychological and physical symptoms in women presenting to a primary care practice. Child Abuse & Neglect. 2003;27(11):1247–58. doi: 10.1016/j.chiabu.2003.05.001. [DOI] [PubMed] [Google Scholar]
  40. The Positive Outcomes for Orphans Research Team. Positive Outcomes for Orphans (POFO) - A five country study of the wellbeing of orphaned and abandoned children. 2010 See Publications page of http://globalhealth.duke.edu/dghi-fieldwork/open-projects/pofo.
  41. Thompson C, Roberts K, Curran A, Ryan L, Wright R. Caretaker-Child Concordance for Child’s Exposure to Violence in a Preadolescent Inner-City Population. Archives of Pediatric and Adolescent Medicine. 2002;156:818–823. doi: 10.1001/archpedi.156.8.818. [DOI] [PubMed] [Google Scholar]
  42. UNAIDS. Investing in our future: psychosocial support for children affected by HIV/AIDS: a case study of Zimbabwe and Tanzania. Geneva: 2001. [Google Scholar]
  43. UNICEF. State of the world’s children. New York: 2009. [Google Scholar]
  44. UNICEF. Facts on Children. 2010 Retrieved 12 Feb 2010, from http://www.unicef.org/media/media_45451.html.
  45. UNICEF, UNAIDS, & USAID. Children on the Brink: A joint report of new orphan estimates and a framework for action. 2004. [Google Scholar]
  46. UNICEF and UNAIDS. Framework for the Protection, Care and Support of Orphans and Vulnerable Children Living in a World with HIV and AIDS. New York: UNICEF and UNAIDS; 2004. [Google Scholar]
  47. UNICEF. [last accessed on Feb. 14th, 2012];Orphans. 2012 http://www.unicef.org/media/media_45279.html.
  48. Urassa M, Boerma JT, Ng’weshemi JZL, Isingo R, Schapink D, Kumogola Y. Orphanhood, child fostering and the AIDS epidemic in rural Tanzania. Health Transit Rev. 1997;7(Suppl 2):141–153. [Google Scholar]
  49. Verdeli H, Clougherty K, Bolton P, Speelman L, Lincoln N, Bass J, et al. Adapting group interpersonal psychotherapy for a developing country: experience in rural Uganda. World Psychiatry. 2003;2(2):114–120. [PMC free article] [PubMed] [Google Scholar]
  50. Weathers FW, Litz BT, Herman DS, Huska JA, Keane TM. Unpublished manuscript. 1991. The PTSD Checklist - Civilian Version (PCLC) [Google Scholar]
  51. Weathers FW, Litz BT, Herman DS, Huska JA, Keane TM. The PTSD checklist (PCL): Reliability, validity, and diagnostic utility. Paper presented at the Trauma, Coping and Adaptation ISTSS: 9th Annual Meeting.1993. [Google Scholar]
  52. Whetten K, Leserman J, Whetten R, Ostermann J, Thielman N, Swartz M, et al. Exploring lack of trust in care providers and the government as a barrier to health service use. Am J Public Health. 2006;96(4):716–721. doi: 10.2105/AJPH.2005.063255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Whetten K, Ostermann J, Whetten RA, Pence BW, O’Donnell K, Messer LC, et al. A comparison of the wellbeing of orphans and abandoned children ages 6–12 in institutional and community-based care settings in 5 less wealthy nations. PLoS One. 2009;4(12):e8169. doi: 10.1371/journal.pone.0008169. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Whetten K, Ostermann J, Whetten R, O’Donnell K, Thielman N and the Positive Outcomes for Orphans Research Team. More than the Loss of a Parent: Potentially Traumatic Events Among Orphaned and Abandoned Children. Journal of Traumatic Stress. 2011;24(2):174–82. doi: 10.1002/jts.20625. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Yule W, Canterbury R. The treatment of post traumatic stress disorder in children and adolescents. International Review of Psychiatry. 1994;6(2/3) [Google Scholar]

RESOURCES