Association of Childhood Intrafamilial Aggression and Childhood Peer Bullying With Adult Depressive Symptoms in China

This cross-sectional study examines the mediating role of childhood peer bullying in the association between childhood intrafamilial aggression and depression in Chinese adults.


Introduction
Intrafamilial aggression and peer bullying in childhood are recognized as major social issues worldwide, conferring considerable risk for life-course mental health problems. [1][2][3] A growing number of studies [4][5][6][7][8][9][10][11][12][13] have established a robust association between earlier-life exposure to adverse experience (eg, parental physical maltreatment, sibling aggression, or peer bullying) and later-life psychological outcomes, such as anxiety, depression, self-harm, and attempt or completion of suicide. Although our knowledge of the consequences of adverse experiences is primarily based on studies performed in highly industrialized countries with societies considered to be individualistic, such as North America, Europe, and Australia, similar associations began to be found in more collectivistic or Confucian cultures, such as China, where harsh parenting (eg, "spare the rod and spoil the child") and sibling hierarchical relationships (eg, older siblings get greater respect, but also take on the responsibility of providing care for younger siblings) are standard. 9,12 In addition, dozens of studies provide support for the view that children experiencing intrafamilial aggression were at a higher risk for peer bullying. 3,[14][15][16][17][18][19] Given the pairwise association among childhood intrafamilial aggression, peer bullying, and adulthood mental health, the experience of childhood peer bullying could be a mediator in explaining the association between childhood intrafamilial aggression and adult mental health that has yet to be assessed.
Nevertheless, a recent study highlighted that childhood parental maltreatment and peer bullying exposure has an independent effect on young adults' mental health, and bullying has a stronger effect on adult mental health in comparison with childhood maltreatment. 20 Consistent with the finding, childhood intrafamilial aggression and peer bullying are dealt with by 2 different departments in some countries (eg, China). If childhood intrafamilial aggression is associated with childhood peer bullying and adult depressive symptoms, dealing with intrafamilial aggression and peer bullying independently may neglect the potential association between intrafamilial aggression and peer bullying.
To clarify the association of childhood intrafamilial aggression and peer bullying with depression at a later age and enhance the effectiveness of related policy, this study aims to quantify the mediating role of exposure to childhood peer bullying in understanding the association between childhood intrafamilial aggression and depression symptoms at a later age in China. Childhood intrafamilial aggression included parental physical maltreatment and sibling aggression in this study.
The former refers to any act or series of acts of physical aggression by a parent or a caregiver that results in harm, potential for harm, or threat of harm to a child. The latter is similar but conducted by siblings and is often seen as a normative and harmless component of sibling relationships. 3,18 This study measured sibling aggression, which is often neglected in research.

Measures Childhood Intrafamilial Aggression
Parental physical maltreatment was identified in response to the following questions: "When you were growing up, did your parents or guardian ever hit you? Was that often, sometimes, rarely, or never?" Following the rule of Chapman et al, 23 participants were defined as experiencing physically adverse behavior by parents or guardians if they responded often or sometimes to the questions.
Sibling aggression was identified in a similar way with the questions, "When you were growing up, did your siblings ever hit you? Was that never, rarely, sometimes, or often?" Participants were defined as experiencing sibling aggression if they responded sometimes or often to the questions. 23

Peer Bullying
Participants were defined as being bullied by peers as a child by the following questions 15 : "When you were a child, how often were you picked on or bullied by kids in your neighborhood (never, rarely, sometimes, or often)?" and "When you were a child, how often were you picked on or bullied by kids in your school (never, rarely, sometimes, or often)?" A response of often or sometimes to either question was defined as peer bullying.

Adult Depression Symptoms
A shortened modification of the CES-D scale including 7 items was used to measure depression symptoms. 24 The items were evaluated as follows: (1) "was bothered by things," (2) "had trouble keeping mind on tasks," (3) "felt depressed," (4) "felt everything he/she did was an effort," (5) "felt fearful," (6) "restless sleep," and (7) "felt lonely." The frequency in experiencing such symptoms in the previous week before the survey was encoded from 1 to 4, where 1 indicates none or rarely; 2, some or little; 3, occasionally or a moderate amount; and 4, most or all of the time. Summed scores ranged from 7 to 28, with higher scores indicating more depressive disorders and therefore worse mental health. A categorical variable for the CES-D score was created based on a usual cutoff score of 12. 25 The variable equaled 1 if the CES-D score was at least 12, and 0 if otherwise. Organization, demographic characteristics, socioeconomic status, and level of physical health were controlled for the regression analysis. 28 Adulthood socioeconomic status was measured by educational attainment, which was a binary measure for the upper secondary school level and above.
Childhood socioeconomic status was captured by both parents' educational attainment and household financial status during the respondent's childhood. For the father's or mother's educational attainment, upper secondary school or higher was encoded as 1; otherwise educational attainment was encoded as 0. In addition, respondents were asked to classify household financial status during their childhood period into 2 categories: worse than others or better than others.
Demographic variables included sex (reference group: female), marital status (reference group: married with spouse present, including common-law marriage; unmarried included single, divorced, or separated), and 65 years or older. Physical health was assessed by asking respondents if a physician had diagnosed any chronic disease. If the answer was "yes," the variable was labeled as 1.
Odds ratios (ORs) and 95% CIs were reported for the logistic model. Weighted regression models with robust variance estimates were derived from generalized estimating equations to adjust the SEs for the stratified sampling design and response rate. The initial weight consisted of crosssectional weights from CHARLS 2015. Then, the sample attrition adjustment method using the response propensity model was applied to obtain the weight of our sample. 29 Because 1796 respondents did not complete the 7 questions of the CES-D scale, the final weights further accounted for the response probability of the CES-D question. The inverse probability weight factor is calculated by the inverse predicted probability of completing the assessment of CES-D for everyone.
As a robust check, the multiple imputation method was considered to impute missing data by creating 20 imputed data sets, and logistic models were then applied. The hypothesis that the association between childhood intrafamilial aggression and mental health at later age will be mediated by being bullied by peers was tested using a 4-step analysis with the Sobel approach. 30,31 The method involves testing a direct path between childhood intrafamilial aggression and adult depression symptoms and then estimating how much the association is reduced by the inclusion of childhood peer bullying. Following Buis, 32 total effects were calculated and decomposed into direct and indirect effects. The details of the analysis are presented in the eMethods in the Supplement.

Results
The mean (SD) age of the 15 450 respondents in the present study was 59.5 (9.9) years, with 7987 women (51.7%) and 7463 men (48.3%). The mean (SD) CES-D score was 12. . When incomplete data were imputed with the multiple imputation method, the results were consistent (eTables 2 and 3 in the Supplement). Table 3 presents the indirect contribution of being bullied by peers in the association of childhood intrafamilial aggression and adulthood mental health. Childhood peer bullying was a mediator of this association. It explained the association in part; the contribution of peer bullying was 30% (95% CI, 19%-42%) of the association between childhood parental maltreatment and adult depression symptoms and 35% (95% CI, 15%-54%) of the association between sibling aggression and adult depression symptoms.

Discussion
In this large, population-based, cross-sectional study, we found that exposure to intrafamilial aggression or peer bullying during childhood was associated with adult risk of depression in China,     [4][5][6][7][8][9][10][11][12][13]20,[33][34][35] that have found a negative association between childhood adverse experience and adult mental health, despite culture differences. This study adds to the emerging evidence that sibling aggression, a less-studied type of intrafamilial aggression, was also associated with depression symptoms in adulthood. 3 As expected, analyses indicated that the association of childhood intrafamilial aggression and adult depression was partially mediated by being bullied by peers. Childhood intrafamilial aggression was associated with elevated levels of depressive symptoms at a later age through increased likelihood of peer bullying. The results were consistent with those of previous research, in which peer bullying was found to be associated with prior experiences and subsequent mental health problems. [35][36][37][38] Furthermore, a study showed the similar view that the association between maltreatment and depression at the same stage was mediated by peer bullying among US children from low socioeconomic backgrounds. 17 The mediational findings could be explained by organizational theories of development suggesting continuity in relationships. 38,39 Experience of intrafamilial aggression may lead children to develop negative expectations pertaining to themselves and others and a concept of relationships involving bullying and agression. 40 In subsequent relationships, these children may continuously recreate familiar social environments to validate their expectations so that they could maintain a coherent sense of self, which may help them adapt to maltreated and neglected homes. 41 In this case, children experiencing intrafamilial aggression might be vulnerable to bullying behaviors, which may put them at risk for mental health issues at an older age.
To our knowledge, this study is the first to establish the mediating role of childhood peer bullying in the association between adult mental health and childhood exposure to intrafamilial aggression, including parental physical maltreatment and particularly sibling aggression. The present study is unique in demonstrating to what extent childhood exposure to intrafamilial aggression is associated with adulthood mental health through peer bullying exposure.
Our investigation sheds light on the creation and implementation of prevention and intervention programs to mitigate the effect of early-life stress and to promote life-course mental health. First, to pursue the premise of maximizing mental health throughout one's life, a life-course policy on health promotion should be adopted instead of only targeting the mental health of specific age groups. The possibility that the influences of childhood intrafamilial aggression extend to health in later age implies that policy interventions should work throughout the entire life cycle, beginning from childhood. Those growing up within an environment of intrafamilial aggression may not only be vulnerable to poor mental health but may also experience peer bullying; thus, mental health interventions or policies should be tailored to focus on these people, recognizing early warning signs of parental physical maltreatment and in particular sibling aggression.
Second, antibullying efforts at school should also account for children's parent and sibling relationships. Because peer bullying is a mediator of the association between intrafamilial aggression and mental health in later age, it is important for schools, health services, and other agencies to coordinate their responses to intrafamilial aggression and peer bullying.

Limitations
When we studied childhood adverse experiences, our measurements were crude owing to data constraints of CHARLS. The effects of onset and severity of childhood adverse experience were not investigated in this study and should be in future studies. In addition, the indicators to measure childhood adverse experience were retrospective self-evaluation with potential measurement error issues, whereas prospective evidence suggests that effects of childhood adverse experience reach this far. 4,7,42 However, self-evaluation could reveal participants' own perceptions of their internal states and has been found to be congruent with peer evaluation. 43,44 Finally, girls are subjected more