The association of depression and child maltreatments among Indonesian adolescents

Background: Depression is one of the most prevalent emotional mental health problem among adolescents. Mental health problem might be the result of child maltreatment considering their prevalence are increasing simultaneously in Indonesia. The aim of this study is determining the association between depression and history of maltreatment among adolescents. Methods: An analytic cross-sectional study was conducted to 786 junior high school students of Bandung City, West Java, Indonesia. Subject was selected using two stage cluster sampling. The questionnaire of Children’s Depression Inventory (CDI) and ISPCAN Child Abuse Screening Tool (ICAST) were applied to assess depression and history of maltreatment respectively. Depression confirmation was diagnosed by psychiatry based on scored positive in CDI. Data were analyzed using chi-square and multiple regression test. Results: History of child maltreatment was associated with depressive disorder in adolescents (p = 0.03). All dimension of child maltreatment had significant association with depression (p <0.05). Psychological violence had the highest risk factor for the occurrence of depressive disorders (PR = 6.51), followed by violence exposure and physical violence. Sexual violence was not a common dimension of child maltreatment among students. The history of psychological violence had the strongest association with depression, which three times more likely to develop depression (POR = 3.302, p = 0.004) Conclusion: Psychological violence was proven as a strong risk factor in developing depression symptom for adolescents student. Early intervention to prevent maltreatment and its consequences is critical, as well.


Study design
A cross-sectional study was conducted on junior high school students. A two-stage cluster sampling was done to determine the school and the subjects. Firstly, several schools were selected, and then the adequate number of students was determined by simple random sampling method. A minimum sample size of 770 students was needed in this study (99% power and 95% significance interval). The study was conducted from May to December 2016.
A letter of approval from the Provincial Directorate of National Education of the city where the research took place was obtained prior to the study. An ethical approval was issued by the Health Research Ethics Committee Faculty of Medicine, Universitas Padjadjaran

Tools
Depression was assessed using the Children's Depression Inventory (CDI) 16,17 as well as interviews for the enforcement of a diagnosis based on the criteria of the DSM V diagnostic criteria. Depression criteria were fulfilled with students' filled questionnaire. Later on, with the determination of the subsequent CDI score ³19, the interview was carried out by psychiatrist based on the DSM V diagnostic criteria. The CDI instrument itself had been validated in Indonesian version. 18 To assess any history of child maltreatment, we used the ICAST-C questionnaire 19,20 had been validated in Indonesian version 21 . ICAST-C consists of 55 questions from 5 dimensions of maltreatment: violence exposure (9 questions), physical (18 questions), psychological (18 questions), neglect ( 6 questions), and sexual (4 questions). Score for each question were interpreted as follow: 1=if there is a history of violence and 0=if there is no history of violence. A cut-off point was made using the mean value of the history of child maltreatment data. Subjects with a total score of child maltreatment dimensions below the cut-off point were not categorized as experiencing child maltreatment.
For correlation validity test, we compared the items correlation value with reference value by taking 5% with the number of respondents (45) as ɑ value. It turned out to be 0.294. The results showed that all items in ICAST-C had adequate validity. Based on Kuder-Richarson reliability test method, the ICAST-C instrument showed strong reliability (KR20=0.92 and KR21=0.87).

Data analysis
Descriptive tests with numeric and percentage value presentation were used to analyze the results of CDI and I-CAST. Analytic Chi-square test was used to analyze the difference of socio-demographic characteristics of students among depression subjects and association between scores of both instruments. Bivariate analysis between the histories of child maltreatment with depression was tested with Prevalence Ratio (PR). Multivariate test was proceded to analyze which type of maltreatment was the most correlated to depression. Results were considered significant if p value <0.05. Data management and analysis were done using SPSS (Statistical Package for Social Science) 15.0.

Results
A total of 845 students from 23 junior high schools gave their consent, however only 835 filled a questionnaire. After examining the questionnaire, a number of 786 students participated in this study ( Figure 1). Students were from 7 th grade (34.86%), 8 th grade (36.51%), and 9 th grade (28.63%) of junior high school. Both male and female students had almost equal proportion of 441(56.11%) and (345) 43.89% respectively. The age range was 12-16 years old, with mean age 13 years old. A 43 (5.47%) subjects with CDI score ³19, had subsequently interviewed by psychiatrist and met the DSM-V criteria of depression. There were no significant differences of students' socio-demographic characteristic among depression and non-depression subjects (Table 1). A number of 27 (3.43%) subjects with depression had experience of child maltreatment. Significant association was found between depression and child maltreatment history ( p=0.03) (see Table 2). Chi-square p <0.05 Based on bivariate analysis, all of 5 dimensions of child maltreatment history were associated with depression. Subjects with psychological victimization history were 6.51 times more likely to develop depression. While neglect was experienced by total of 315 subjects, with 28 of them developed depression. (Table 3)

Discussion
This study revelaed that child maltreatment was quite common in junior high school students in Bandung, West Java, Indonesia. About 46,69% of our subjects had history of child maltreatment. Indonesia itself has no definite data of child maltreatment or child abuse so far. However, this current study showed the history of victimization in psychological dimensions was the most widely experienced by the students (46.31%), followed with violence exposure (45.04%), and physical victimization (40.84%). On the other hand, sexual victimization was uncommon (21.88%). Epidemiological data from a study in India showed that the highest prevalence of violence experienced was psychological (61.9%), physical (21.43%) and sexual violence (16.67%). 23 Meanwhile data from the United States showed 686.000 cases of children violence was neglect (78.6%), physical abuse (18.3%) and sexual violence (9.3%). 23 However, several studies showed the cumulative prevalence based on a survey of communities were about 15-30% in girls and 5-15% in boys for sexual violence, 5-35% for physical violence, 4-9% psychological violence and 6-12% neglect 23 .
Most of subjects were seventh grade students who were in early adolescent phase (ages 11 to 14) were typically very egocentric with poor self-regulation. 24 Those young adolescents were most emotionally victimized at school. School violence played an important role in developing depression. 12,25 Meanwhile, conduct disorder had the highest prevalence in adolescents, namely 7% in adolescents aged 12 to 16 years old. 26 This study showed all child maltreatment dimensions had significant association with depression. This is similar to prior studies that asserted the existence of correlation between the histories of violence in children with depression. 27,28 Subjects with history of psychological child maltreatment had 6.51 times higher risk of depression. This finding is similar to a meta-analysis that stated abuse or maltreatment and neglect were strongly associated between child abuse history with depressive disorder in adolescence. 25 Moreover, Pirdehghan's study in Iran showed a correlation between mental disorder and violence (Spearman rho:0.2;pvalue <0.001) as well. 29 We found that depressive symptoms on the CDI were strongly associated with all dimension of child maltreatment, particularly psychological violence and neglect. In one review of 124 studies, psychological violence increased the risk of depression by an odds ratio of 3.06, whereas physical abuse increased the risk of depression by only half that amount.Furthermore, psychological violence maltreatment was more closely related to depression severity than sexual or physical abuse. 10 Emotional abuse and neglect may alter the development of brain reward and oxytocin systems in children leading to impaired parental care giving in the subsequent generation. 31 These findings were confirmed in a recent systematic review and meta-analysis, although most of the data came from retrospective cross-sectional studies or longitudinal designs that relied on selfreported abuse. 32 Sexual violence showed a far weaker association. There are several possible explanations. may be underreported because of stigma. 33 Practitioners should be aware that violence during childhood might result in negative consequences in adolescents. Therefore, a good understanding is required to prevent violence acts against children hence might increase effective intervention for violence issue in adolescents. 30

Conclusions
History of child abuse has correlation with depressive disorder in adolescents. Psychological child abuse was the highest risk factor for the onset of disorders of depression compared to other violent dimensions.

Limitations
We recognized those questionnaires might cause depression diagnosis was overrepresented in adolescents. Thus, we involve some psychiatrists to give interview to subjects in order to solve this issue. The number of subjects that met the depression criteria was also small which should be carefully considered to represent the broader population. Limitation of this research was also in the process of fulfilling the questionnaires. Some students met some challenges in answering the questions, thus the authors accompaniment was required to get the foremost results. The use of legally defined cases is likely to have over or underestimated the true prevalence of childhood maltreatment.

Consent for publication
Not applicable.

Availability of data and materials
All data and materials of this study are available at Faculty of Medicine, Universitas Padjadjaran, Bandung, Indonesia, by contacting the corresponding author Meita Dhamayanti, email meita.dhamayanti@unpad.ac.id

Competing interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt Internal Grant of Universitas Padjadjaran for financial support the research, authorship, and/or publication of this article Figure 1 Sample selection

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download. STROBEchecklistcrosssectionalCANDepresBMC1.doc