The Moderating Effect of Contact with Children on the Relationship between Adverse Childhood Experiences and Depression in Adulthood among a Chinese Adult Population

The effect of adverse childhood experiences (ACEs) on depression in adulthood has been identified in many studies; however, the underlying mechanisms remain unclear. To understand the moderating effect of ACEs on depression, a moderation analysis using the interaction effect model was performed based on data obtained from the China Health and Retirement Longitudinal Study. This study found that people with ACEs had significantly lower depression scores than those without ACEs, particularly in categories such as physical abuse, emotional neglect, sibling death, parental illness/disability, parental depression, hunger, violence, and bullying. In addition, the results indicated that contact with children moderated the relationship between ACEs and depression in adulthood. Increased levels of contact with children reduced the adverse effects of parental drug abuse and the experience of starvation, but not physical abuse. This study highlights the role of family support in eliminating health disparities, which can reduce the effects of ACEs on depression in adulthood.


Introduction
Substantial empirical evidence from multiple countries has demonstrated that adverse childhood experiences (ACEs) are highly correlated with depression in adulthood [1][2][3]. For example, scholars have long believed that ACEs may enhance vulnerability to psychological development throughout the course of a person's life, specifically depression [4][5][6][7][8]. In addition, some meta-analyses confirm that a history of ACEs significantly increases the risk of depression and anxiety in adulthood [9][10][11][12][13]. These results underscore the importance of evaluating the negative effects of ACEs.
The existing literature has also examined the effects of different types of ACEs on depression. Of the various ACEs in the literature, childhood physical and sexual abuse and a history of family violence showed the strongest relationships with anxiety disorders later in life [10,12,14]. Researchers found that childhood verbal and emotional abuse increased the risk of depression [15][16][17]. Sachs-Ericsson et al. [18] also found people who were verbally abused had 1.6 times as many symptoms of depression and anxiety as those who had not been verbally abused and were twice as likely to have experienced a mood or anxiety disorder in their lifetime. Adolescents who experienced peer isolation and emotional neglect were found to be most at risk of symptoms of depression and anxiety [19]. Symptom severity and major depressive episodes were greater for those who experienced community violence and family distress [20]. In addition, Cavanaugh and Nelson [21] found that Black American adults who reported any of the five types of child abuse/neglect

Data
This study used data from the China Health and Retirement Longitudinal Study (more detailed information about CHARLS can be found at http://charls.pku.edu.cn/zh-CN (accessed on 3 May 2022)) conducted by the National School of Development at Peking University. The survey adopted the method of multi-stage sampling. Respondents were selected from 28 provinces (autonomous regions and municipalities) in China in 2011, and follow-up surveys were conducted in 2013, 2014, 2015, and 2018. Information on the respondents' ACEs was obtained from the 2014 survey. Respondents in this survey retrospectively reported all ACEs, such as physical abuse, emotional neglect, experiencing parental death, divorce, and substance abuse. In addition, the 2015 follow-up survey provided data on mental health in adulthood, and we pooled the two parts of the variables together. Therefore, this survey provided a reliable database to analyze the effect of ACEs on individuals' depression. After excluding cases with missing data, the final valid sample for this study was 11,564. The average age of respondents was 57.49, with the youngest being 30 and the oldest being 84. The distribution of the selected samples is shown in Table 1. The dependent variable of this study was the respondents' depression scores in adulthood based on the depression scores at the time of the survey. This study used the Depression Scale (CES-D) developed by Sirodff [37], which is a self-report scale designed to measure depressive symptomatology in the general population. Depression in adulthood was measured using a 10-item scale, including "I was bothered by things that don't usually bother me", "I had trouble keeping my mind on what I was doing", "I felt depressed", "I felt everything I did was an effort", "I felt hopeful about the future", "I felt fearful", "My sleep was restless", "I was happy", "I felt lonely", and "I could not get 'going'." The answers were rated on a four-point Likert scale (1 = "rarely or none of the time"; 4 = "most or all of the time"). Cronbach's alpha coefficient in this study was 0.795. The depression score was the total score of the above 10 items, with a minimum value of 2 and a maximum value of 40. Respondents with higher scores indicated higher levels of depression.

Independent and Control Variables
The independent variable was ACEs. Previous studies often used retrospective indicators that allow respondents to recall ACEs that occurred during childhood [38,39]. This study measured the following types of ACEs: physical abuse, emotional neglect, parental divorce, parental death, sibling death, parental drug abuse, parental alcoholism, parental crime, parental illness/disability, parental depression, the experience of starvation, the experience of violence, and being bullied by schoolmates. All questions were dichotomous, with "Yes/No" response options to indicate whether the adverse event happened during childhood.

Moderating Variable
Contact with children was included in the model as a moderating variable measured by the question "How often do you have contact with your own children either by phone, text message, mail, or email?" Options for response were "Almost every day", "2-3 times a week", "Once a week", "Every two weeks", "Once a month", "Once every three months", "Once every six months", "Once a year", and "Almost never". Higher values indicated more frequent contact with children. The maximum value was 9, and the minimum value was 1.

Statistical Analysis
This study used a t-test and an F-test to examine differences in depression scores among populations with different characteristics and people with and without ACEs. Using depression scores as the dependent variable, ordinary least squares (OLS) regression was applied to analyze the effect of childhood ACEs on depression scores in adulthood. Moreover, to examine the moderating effect of contact with children on ACEs and depression outcomes during adulthood, this study established the interaction of ACEs and contact with children. Model 1 included only control variables; Model 2 was the main effect model; Model 3 was the interaction effect model. All statistical analyses were conducted using STATA 15.0 (StataCorp LLC, College Station, TX, USA). Table 2 shows the average depression scores for different characteristic groups. Female depression scores were significantly higher than male scores. The depression scores of rural residents were significantly higher than those of urban residents. Respondents with a level of education of elementary school and below had the highest depression scores, followed by middle school, whereas respondents with a college or above education background had the lowest depression scores. The depression scores of minority ethnic groups were slightly higher than those of Han ethnic groups. However, there was no significant difference in depression scores between residents with and without religious beliefs and between party members and non-party members. Table 3 presents the average depression scores for male and female respondents with different ACEs. We found that respondents who experienced physical abuse, emotional neglect, parental death, sibling death, parental illness/disability, parental depression, starvation, violence, and bullying had higher depression scores than those who did not. Male respondents with parental alcoholism and parental criminality scored significantly higher than those without these experiences. Female respondents who experienced parental divorce scored higher on depression without this experience. In addition, parental drug abuse was not significant among male or female respondents.

Male Female
All Note: *** p < 0.001, ** p < 0.01, * p < 0.05, † p < 0.1. Table 4 presents the average depression scores for urban and rural residents. Among both urban and rural residents, those who experienced physical abuse, emotional neglect, parental depression, starvation, violence, and bullying had higher depression scores than those who did not. However, parental divorce, parental drug use, and parental alcoholism were not significant among either urban or rural residents. Among urban residents, only parental depression, starvation, violence, and bullying were significant, whereas among rural residents, physical abuse, emotional neglect, parental death, sibling death, parental crime, parental illness/disability, parental depression, starvation, violence, and being bullied were significant. Note: *** p < 0.001, ** p < 0.01, * p < 0.05, † p < 0.1. Table 5 presents the results of the OLS regression analysis. Model 1 only included control variables, Model 2 included independent variables of ACEs on the basis of control variables, and Model 3 included the interaction term of ACEs and contact with children. From Model 1 to Model 3, control variables-namely sex, Hukou type, age, religious belief, political status, and education level-had significant effects on respondents' depression scores.

The Effects of ACDs on Depression
Among the independent variables in Model 2, the effects of physical abuse, emotional neglect, sibling death, parental illness/disability, parental depression, the experience of starvation, the experience of violence, and being bullied by schoolmates were significant. Specifically, respondents who experienced physical abuse had a depression score that was 0.565 higher than those who did not; those who experienced emotional neglect had depression scores 0.43 higher than those who did not; those who experienced the death of a sibling had depression scores 0.406 higher than those who did not; those whose parents were sick or disabled had depression scores 1.237 higher than those whose parents were not; those whose parents were depressed had depression scores 2.168 higher than those whose parents were not; those who experienced starvation in childhood had depression scores 0.958 higher than those who did not; those who experienced violence had depression scores 0.991 higher than those who did not; those who experienced bullying by classmates had depression scores 1.230 higher than those who did not.  Note: standard errors in parentheses; *** p < 0.001, ** p < 0.01, * p < 0.01, † p < 0.1.
In Model 3, after incorporating the interaction items of ACEs and contact with children, parental illness/disability, parental depression, the experience of starvation, the experience of violence, and being bullied by schoolmates remained significant; however, physical abuse, emotional neglect, and sibling death were no longer significant. The interaction terms of physical abuse, parental drug abuse, the experience of starvation, and contact with children were significant. For every one-unit increase in the frequency of contact with children, those who experienced physical abuse had depression scores 1.862 lower than those who did not. This suggests that contact with children had a significant moderating effect on the depression scores between those who experienced physical abuse and those who did not. We calculated the marginal effect to further show the moderating effect of contact with children, as shown in Figure 1. We found that with increased frequency of contact with their own children, those who were physically abused had significantly higher depression scores than those who were not. In other words, the increased frequency of contact with their children did not eliminate the difference in the depression scores between these individuals.
For parental drug abuse, for every one-unit increase in the frequency of contact with children, respondents whose parents used drugs scored 0.941 higher on depression than those whose parents did not. Figure 2 shows the moderating effect of contact with children on depression scores between people whose parents were drug users and those whose parents were not. As the frequency of contact with their children increased, the depression scores of those whose parents used drugs reduced more than those whose parents did not. In other words, increased bonding with their children contributed more to the improvement in depressive outcomes for those individuals whose parents were drug users than for those whose parents were non-drug users.
ating effect on the depression scores between those who experienced physical abuse and those who did not. We calculated the marginal effect to further show the moderating effect of contact with children, as shown in Figure 1. We found that with increased frequency of contact with their own children, those who were physically abused had significantly higher depression scores than those who were not. In other words, the increased frequency of contact with their children did not eliminate the difference in the depression scores between these individuals. For parental drug abuse, for every one-unit increase in the frequency of contact with children, respondents whose parents used drugs scored 0.941 higher on depression than those whose parents did not. Figure 2 shows the moderating effect of contact with children on depression scores between people whose parents were drug users and those whose parents were not. As the frequency of contact with their children increased, the depression scores of those whose parents used drugs reduced more than those whose parents did not. In other words, increased bonding with their children contributed more to the improvement in depressive outcomes for those individuals whose parents were drug users than for those whose parents were non-drug users. For every one-unit increase in the frequency of contact with children, respondents who experienced starvation in childhood scored 0.116 higher on depression than those who did not. Figure 3 shows the marginal effect of the interaction term between experiencing starvation and contact with their children. Increased bonding with their children was more beneficial for reducing depression among respondents who experienced child- For every one-unit increase in the frequency of contact with children, respondents who experienced starvation in childhood scored 0.116 higher on depression than those who did not. Figure 3 shows the marginal effect of the interaction term between experiencing starvation and contact with their children. Increased bonding with their children was more beneficial for reducing depression among respondents who experienced childhood starvation than those who did not. In addition, the interaction terms for parental divorce, parental death, and contact with one's children were marginally significant. Increased connection with their children improved depression among people who experienced parental divorce and death (see Figures A1 and A2).

Discussion
This study examined the effect of ACEs on depression and the moderating role of contact with children based on CHARLS data. First, this study further confirmed the fact that most ACEs contribute to depression in adulthood. The present study examined 13 types of ACEs. Eight of these were significant, accounting for 62%. More specifically, adults who experienced physical abuse, emotional neglect, sibling death, parental illness/disability, parental depression, starvation, violence, and being bullied by schoolmates in childhood were more likely to experience depression in adulthood than those who did not. This finding was consistent with extant studies [40][41][42].
Second, the current study demonstrated that not all ACEs cause depression in adulthood and further illustrated that different ACEs have different outcomes. Although the ever-increasing literature indicates that ACEs may trigger mental health problems, such as depression [2,4], not all individuals who experience ACEs develop depression. Five ACEs (parental divorce, parental death, parental drug abuse, parental alcoholism, and parental crime) did not cause depression in adulthood. However, this does not mean that these ACEs did not cause other harm. For instance, parental divorce has been shown to lead to lower self-rated health [43]. The reasons and mechanisms behind only certain ACEs leading to depression and not others require further research.
Third, this study determined the role of contact with children in mitigating the effects of ACEs on depression risks in adulthood. The results revealed that an increase in contact with their children resulted in lower depression scores in adulthood among individuals who experienced parental drug abuse and starvation in childhood, indicating that contact

Discussion
This study examined the effect of ACEs on depression and the moderating role of contact with children based on CHARLS data. First, this study further confirmed the fact that most ACEs contribute to depression in adulthood. The present study examined 13 types of ACEs. Eight of these were significant, accounting for 62%. More specifically, adults who experienced physical abuse, emotional neglect, sibling death, parental illness/disability, parental depression, starvation, violence, and being bullied by schoolmates in childhood were more likely to experience depression in adulthood than those who did not. This finding was consistent with extant studies [40][41][42].
Second, the current study demonstrated that not all ACEs cause depression in adulthood and further illustrated that different ACEs have different outcomes. Although the ever-increasing literature indicates that ACEs may trigger mental health problems, such as depression [2,4], not all individuals who experience ACEs develop depression. Five ACEs (parental divorce, parental death, parental drug abuse, parental alcoholism, and parental crime) did not cause depression in adulthood. However, this does not mean that these ACEs did not cause other harm. For instance, parental divorce has been shown to lead to lower self-rated health [43]. The reasons and mechanisms behind only certain ACEs leading to depression and not others require further research.
Third, this study determined the role of contact with children in mitigating the effects of ACEs on depression risks in adulthood. The results revealed that an increase in contact with their children resulted in lower depression scores in adulthood among individuals who experienced parental drug abuse and starvation in childhood, indicating that contact with children mitigated these adverse effects. Childhood adversity should be healed in adulthood, and the family is an important arena for such healing. Contact with children helps ACE victims experience the family warmth that they did not experience as children, thereby reducing the risk of depression. Starvation is a physical experience, and it develops into a psychological experience as an individual grows, evolving into a fear of starvation and a sense of insecurity. Contact with children enables individuals suffering from hunger to gain a sense of security through interaction with family members. Some scholars have noted that resilience [25] and stress management [26] as moderating variables are particularly important when it comes to reducing the effect of ACEs on depression. However, resilience and stress management are both intrinsic qualities of the individual and do not reveal how family intergenerational relationships affect people's mental health. This study revealed that contact with children moderates the relationship between ACEs and depression, further underscoring the role of family intergenerational relationships in reducing the vulnerability effect of ACEs.
The effects of ACEs on adult depression further indicated the importance of family support. With the moderation of contact with children, parental divorce and death had significant marginal effects. Parental divorce and death are family issues. As mentioned above, Chinese families are becoming increasingly diverse, with rising divorce rates and an increase in single-parent families. The results of this study demonstrated that better communication with their children could reduce the impact of parental divorce and death on depression in an individual's adult life. In other words, pain caused by the family should be treated by the family. Furthermore, this sheds light on the importance of family policy in reducing the negative impact of childhood adversity [44][45][46]. These findings suggest that China, which has experienced diverse family forms, should strive to implement policies that facilitate family communication. For instance, public services, such as education and medical care, that are consistent with the local registered population should be provided in the permanent residence of migrant workers, which would allow left-behind children to live in cities with their parents. Moreover, the government introduced policies to prevent family breakdowns, such as a cooling-off period for divorce; however, this policy lacks sufficient flexibility and influence, and the effect is not satisfactory. The development of modern communication technology has brought people more convenient communication conditions and facilitated communication between family members. Social policy should be family oriented and systematically integrated.
The results of the present study have certain limitations. First, although the survey data used in this study are nationally representative, the sample age was mainly over 40 years old, and cannot reflect the situation of ACEs and depression among adults under 40 years old. Moreover, people born in the 1970s or 1960s experienced historical events, such as the Cultural Revolution and the transition from a planned economy to a market economy in their early childhood, which are unique and lack universal significance. Second, this study used a self-rated depression scale, which was not confirmed by a clinical evaluation by a licensed mental health provider and may have certain errors. Third, regarding the mechanism of ACEs and depression in adulthood, this study only examined the role of contact with children and did not include social support variables, such as contact with friends. Future studies can investigate additional moderator variables to identify more variables that can mitigate the effects of ACEs on depression in adults.

Conclusions
The present study explored the role of contact with children as a moderator in the effect of ACEs on depression in adulthood and found that enhancing contact with children could reduce the effects of at least two ACEs (parental drug abuse and starvation) on depression in adulthood.
These findings highlight the need to focus on the role of the family in individual mental health. First, we should develop a good family culture. As an important element of a social organism, the family carries out basic functions, such as providing guidance on social norms, moral education, cultural inheritance, and emotional comfort. Family culture has a direct, lasting, and subtle influence on the healthy growth of individual family members. Second, we must consolidate and cultivate family education in the youth. It is necessary to provide guidance around the implementation of the fundamental task of moral education and guide parents to use correct actions, thoughts, and methods to cultivate good thoughts, conduct, and habits in children. Family education is the foundation of all education, and it shoulders the important responsibility of cultivating values in young people. Finally, we should improve family services and promote social harmony. Harmony should be created between the society at large and small families, thus facilitating happiness and a sense of well-being in every family. Social harmony is ultimately reflected in the happiness of thousands of families and in the continuous improvement of the lives of hundreds of millions of people.       Figure A2. The marginal effects of contact with children on depression and parental death.