This study is theoretically grounded in Bronfenbrenner's (1) ecological systems theory, which posits that human development occurs through dynamic interactions between individuals and the multiple environmental systems surrounding them. These systems include the microsystem, mesosystem, exosystem, macrosystem, and chronosystem, each influencing and interacting with one another throughout the developmental process.

From this perspective, a mother's ACEs can be considered a distal risk factor originating in the chronosystem and early microsystem. ACE can leave enduring psychological and social consequences into adulthood, particularly affecting a mother's QoL and psychosocial resources such as self-esteem. These resources, in turn, have direct and indirect impacts on the development and well-being of her children.

In the conceptual model of this study, the mother's QoL is conceptualized as a mediating variable reflecting current psychosocial functioning. It is hypothesized that the effects of ACE on maternal mental health, physical health, social relationships, and life satisfaction are transmitted to children through QoL. Conversely, maternal self-esteem is posited as a protective factor buffering the intergenerational transmission of adversity. Mothers with high self-esteem are more likely to maintain parenting efficacy, remain consistent, and engage in positive parenting behaviors even under stressful circumstances.

This study also incorporates socio-ecological covariates, as proposed in Bronfenbrenner's model. For example, variables such as education level, household income, and perceived health status classified within the exosystem or macrosystem interact with mothers' internal resources (QoL, self-esteem) and influence child development. By statistically controlling for these covariates, the study aims to isolate the unique effects of maternal psychosocial characteristics on children's QoL.

In conclusion, the theoretical framework of this study presents an integrative model that explains how ACE, QoL, self-esteem, and socio-environmental factors interact within the multilayered structure of the ecological system to influence children's QoL. This framework not only elucidates the pathways of intergenerational transmission of adversity but also provides scientific evidence for designing multi-level intervention strategies. Figure 1 illustrates the conceptual framework of this study, adapted from Bronfenbrenner's ecological systems theory. It visually depicts the multiple nested systems individual, microsystem, mesosystem, exosystem, and macrosystem showing how maternal ACE, QoL, and self-esteem are embedded within and influenced by broader socio-environmental contexts. The diagram highlights the directional relationships between these variables and situates them within the ecological layers, clarifying the hypothesized pathways through which maternal characteristics and contextual factors impact children's QoL.

Although maternal ACEs have been widely recognized as risk factors for adult psychological distress and maladaptive parenting, their intergenerational impact on children's QoL remains underexplored. Specifically, few studies have examined the mechanisms through which maternal ACEs influence children's QoL, particularly in relation to maternal psychosocial resources such as quality of life and self-esteem.

Previous studies have often treated parental ACEs as static background variables, overlooking the dynamic roles of maternal well-being and internal psychological factors. Moreover, inconsistent findings remain regarding the extent to which maternal self-esteem serves as a protective factor, and limited research has empirically tested both mediation and moderation models in a single analytic framework.

Therefore, this study aims to investigate how maternal QoL mediates the relationship between maternal ACEs and children's QoL, and whether maternal self-esteem moderates this relationship. By clarifying these pathways, the study seeks to enhance the understanding of intergenerational transmission of adversity and inform targeted intervention strategies that support maternal psychological health to improve child outcomes.

The study was a descriptive, correlational research conducted to verify the effects of mothers' ACEs and QoL on the QoL of their children and validate the moderating effect of mothers' self-esteem and the mediating effect of QoL.

This secondary data analysis utilized raw data from the 2018 Childhood Life Experience Survey conducted by the Korea Institute for Health and Social Affairs. A two-stage stratified sampling method was applied to select survey districts and households to ensure the representativeness of the sample.

In the sample extraction process, enumeration districts were selected based on the results of the Population and Housing Census, and samples were allocated based on adults aged 19–59. Subsequently, the systematic sampling method was used to select households with children aged 9 to under 18 in each enumeration district for the survey, thereby constructing nationally representative data (23).

From the total of 1,515 respondents, the present study specifically targeted married mothers with an ACE score of one or more. The inclusion of only married mothers was a deliberate methodological choice aimed at ensuring comparability in family structure and controlling for potential confounding effects of marital status on maternal psychological resources, parenting behaviors, and child outcomes. Restricting the sample to mothers living with their spouse and children allowed for a clearer examination of the intergenerational transmission of adversity within relatively stable family environments. Respondents with missing values in any of the variables used in the analyses were excluded. The final analytic sample comprised 930 participants.

Children's QoL was assessed through self-report questionnaires completed directly by children aged 9–17 years. In households with multiple children, each eligible child within this age range was invited to respond to the child survey, rather than a pre-designated “target child”.

The general characteristics of the study participants were classified as maternal age, education level, dual- income status, average monthly income, subjective health status, number of children, age at first childbirth, child gender, child age, and child's subjective health status. Maternal age was categorized as under 40, in the 40s, and over 50. Educational level was categorized as high school or below and college or above. Dual-income status was categorized as single-income and dual-income. Average monthly income and maternal subjective health status were categorized as poor, average, good, and very good. The number of children was categorized as one, two, three, and four. Children's gender was categorized as male and female. Children's ages were categorized as under 13 years old or 13 years or older. Children's subjective health status was categorized as poor, average, good, and very good.

The Adverse Childhood Experiences International Questionnaire (ACE-IQ) is a scale used to assess the impact of ACEs on adult life (24). It employs a retrospective self-reporting approach for adults, reflecting on experiences before the age of 18. The questionnaire includes 13 items covering emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, family violence, and other family issues. According to guidance for analyzing ACE-IQ, any experience reported at least once is recorded as “present” (1), and no experience as “absent” (0). The scores of each item were summed, with a possible range from 1–13 points. A higher score indicates a greater number of adverse childhood experiences.

Self-esteem was measured using the 10-item Rosenberg Self-Esteem Scale. The self-esteem scale includes statements such as “I feel that I am at least as worthy as other people”, “I feel that I have a number of good qualities”, and “I tend to feel that I am a failure”. The respondents indicated how much they agreed with each statement, and the responses were categorized into four levels (1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree). Negatively worded items were reverse-coded to calculate the total score for each item. A higher score indicates higher self-esteem. Cronbach's alpha for self-esteem measured in this study was.69.

The overall QoL was measured using seven items. It assesses self-reported satisfaction with life and personal circumstances, considering aspects such as standard of living, health, achievements in life, personal relationships, feelings of safety, sense of community belonging, and security about the future. The responses for each item are categorized on a scale from 0 (not satisfied at all) to 10 (very satisfied). A higher score indicates a higher level of QoL. Cronbach's alpha for QoL measured in this study was.89.

QoL was measured using six items. It assesses self-reported overall satisfaction with life through statements such as “I enjoy my life”, “My life is going well”, “I am living a good life”, “Many good things happen in my life”, “I like my life”, and “I feel happiness about my life”. The responses for each item are categorized on a scale from 0 (strongly disagree) to 10 (strongly agree). A higher score indicates a higher level of QoL. Cronbach's alpha for QoL measured in this study was .95.

The negative impact of maternal ACEs on both maternal and child QoL was analyzed, as well as the transmission processes involved. Additionally, the moderating effect of maternal self-esteem in this relationship was examined using SPSS 26 and the PROCESS macro program. Initially, the demographic characteristics of the subjects were determined through frequency and descriptive statistical analysis. Pearson correlation analysis was conducted to assess the characteristics of the main variables and their interrelationships. Analyses using Models 4 and 5 of the PROCESS macro were performed to validate the research model. The significance of the mediating effects and conditional direct effects was confirmed using bootstrapping tests. The significance of the mediating effects and conditional direct effects was confirmed using bootstrapping tests. PROCESS model 4 (mediation) and model 5 (moderation) were selected to examine the hypothesized mechanisms separately. While models 8 and 59 allow simultaneous estimation of mediated moderation or moderated mediation, they assume a reciprocal and dynamic interplay between maternal self-esteem and maternal QoL. Conceptually, self-esteem and QoL are indeed interrelated; however, our theoretical framework, grounded in Bronfenbrenner's ecological model, positioned maternal QoL as an intrapersonal resource shaped by ACEs, and self-esteem as a personal protective factor moderating this pathway. Therefore, we intentionally tested the two processes independently to preserve model parsimony, avoid overfitting given sample size limitations, and to align with the conceptual model of unidirectional influence hypothesized in this study.

A frequency analysis was conducted to examine the general characteristics of the study subjects (Table 1). The maternal age distribution showed that 205 participants (22.0%) were under 40 years old, 636 (68.4%) were in their 40s, and 89 (9.6%) were over 50, with an average age of 43.37 years. Four hundred and forty-nine (48.3%) had high school educations or below and 481 (51.7%) had college or above. Regarding employment status, 371 participants (39.9%) were in single-income families, while 559 participants (60.1%) were in dual-income families. The average monthly income was 4,801,800 KRW (approximately 3,693.69 USD as of March 4, 2025). Maternal subjective health levels were reported as poor by 10 participants (1.1%), average by 187 participants (20.1%), good by 634 participants (68.2%), and very good by 99 participants (10.6%). Of the participants, 269 (28.9%) had one child, 479 (51.5%) had two children, 161 (17.3%) had three children, and 21 (2.3%) had four or more. The average maternal age at the birth of the first child was 28.61 years.

Examining the characteristics of the children, 482 (51.8%) were male and 448 (48.2%) were female. Of them, 396 (42.6%) were aged 12 or below, and 534 (57.4%) were aged 13 or above, with an average age of 13.16 years. Children's subjective health status was reported as poor by two children (.2%), average by 63 children (6.8%), good by 496 children (53.3%), and very good by 369 children (39.7%). The levels of adverse childhood experiences, self-esteem, quality of life, and children's quality of life among the study participants were as follows (Table 2). The average for adverse childhood experiences was 3.62 ± 1.89, for self-esteem 2.96 ± 0.35, for quality of life 7.18 ± 0.93, and for children's quality of life 7.19 ± 1.21.

ACEs were found to have a significant negative correlation with maternal QoL (r = −.184, p < .001), children's QoL (r = −.204, p < .001), and maternal self-esteem (r = −.102, p < .01). Conversely, maternal QoL showed a significant positive correlation with children's QoL (r = .470, p < .001) and self-esteem (r = .316, p < .001). Children's QoL also demonstrated a significant positive correlation with maternal self-esteem (r = .318, p < .001) (Table 2). Descriptive statistical analysis of the main variables confirmed that the skewness and kurtosis did not exceed the absolute values of 3 and 10, respectively, indicating that the data met the criteria for normal distribution (25).

The pathways through which maternal ACEs impacted their QoL, and subsequently their children's QoL, were analyzed using Model 4 of the PROCESS macro. The control variables included demographic characteristics such as maternal age, education, dual-income status, income, subjective health, number of children, age at first childbirth, child's gender, age, and subjective health (Table 3). The first model assessed the impact of the independent variable (ACE) on the mediator (maternal QoL) and showed significant results. The F-value indicated that this model explained 16.8% of the variance. The analysis revealed that, even after controlling for demographic characteristics, ACE had a significant negative impact on maternal QoL (β = −.177, p < .001), indicating that higher levels of ACE were associated with lower QoL.

The second model analyzed the effects of both the independent variable and the mediator on the dependent variable (children's QoL), again showing a significant F-value and explaining 31.8% of the variance. The results again found that ACEs had a significant negative impact (β = −.119, p < .001), while maternal QoL had a significant positive impact (β = .386, p < .001). These findings suggest that higher maternal ACE levels correlate with lower children's QoL, whereas higher maternal QoL corresponds with higher children's QoL.

The third model explored the direct relationship between the independent variable and the dependent variable, showing significant results, with an F-value and 19.4% of the variance explained. ACEs were found to have a direct negative impact on children's QoL (β = −.187, p < .001). The significance of both the direct and mediated effects indicates partial mediation. The significance of the mediating effect was further confirmed by the bootstrapping test, which showed that the confidence interval did not include zero, verifying that maternal ACEs indirectly affect children's QoL through maternal QoL (Table 4).

The moderating effect of maternal self-esteem on the pathway through which maternal ACEs impact children's QoL was analyzed using Model 5 of the PROCESS macro, as shown in Table 5. This model also controlled for the demographic characteristics of both the mothers and the children and showed significant results, with an F- value and a high explanatory power of 33.8%. The analysis revealed that ACEs had significant negative impacts on children's QoL (B = −.068, p < .001), while both maternal QoL (B = .458, p < .001) and maternal self-esteem (B = .506, p < .001) had significant positive effects on children's QoL. Notably, the interaction term ACE maternal self-esteem also showed a significant positive impact (B = .103, p < .05), indicating that high maternal self-esteem can mitigate the negative effects of ACEs on children's QoL. In the negative relationship shown as decreases in children's QoL by maternal ACEs, maternal self-esteem acted as a protective factor. Figure 2 shows the results of a simple slope test. The solid line in Figure 3 illustrates low maternal self-esteem, where higher levels of ACEs are significantly associated with lower child QoL. In contrast, the dotted line represents high maternal self-esteem, where the negative association between ACEs and child QoL is attenuated and not statistically significant.

Among the covariates, maternal education was significantly associated with higher child QoL. This finding suggests that contextual factors such as maternal education level may play an important role in shaping child outcomes, beyond the effects of maternal ACEs, QoL, and self-esteem.

These findings have several practical implications for intervention and policy. At the individual and family level, efforts should focus on enhancing maternal self-esteem and mental well-being. Evidence-based programs such as cognitive-behavioral therapy and parenting interventions (e.g., the Triple P program) have been shown to improve mothers' psychological health, parenting skills, and mother–child relationships (20, 39–41). At the community level, expanding access to maternal mental health services and parent support programs (such as peer support groups or home visitation for at-risk families) is critical (36). Healthcare and educational professionals should also be trained to recognize signs of parental trauma and connect families with appropriate resources for early intervention.

At the policy level, broader structural supports and preventive measures are needed. Raising public awareness about the long-term impacts of ACEs and integrating ACE screening into routine prenatal and pediatric care could facilitate early identification of high-risk families (CDC, 2025); (8, 42, 43). Additionally, implementing socio-economic support policies (e.g., financial assistance, childcare subsidies, and affordable mental health services) alongside family-friendly workplace arrangements (e.g., flexible hours, parental leave) would help alleviate stress on vulnerable parents. Finally, strengthening social services and legal protections to prevent child abuse and domestic violence remains imperative to safeguard at-risk children. Taken together, a comprehensive multi-level approach addressing both individual psychological resources and broader socio-economic factors is needed to effectively break the intergenerational cycle of adversity and improve children's QoL (8, 36). Relevant institutions should provide tailored support programs for high-risk mothers.

This cross-sectional study limits the ability to determine causal relationships between maternal ACE, QoL, self- esteem, and children's QoL, highlighting the need for longitudinal research to clarify temporal changes and causal pathways. Reliance on maternal self-reports may introduce recall and social desirability biases, particularly in retrospective ACE reporting, which can be influenced by current psychological status. The sample's restriction to a single cultural and regional context reduces generalisability to populations with differing values, family structures, and policy environments. Potential confounders such as paternal ACE, marital relationship quality, family conflict, and discrimination experiences were not measured, underscoring the need for future multivariate models encompassing the developmental histories and interactions of all family members. Furthermore, the use of subjective measures for QoL and self-esteem warrants integration of objective health indicators or observational data to enhance interpretative validity. Finally, the control of only a subset of socio- ecological covariates leaves open the possibility of residual confounding from unmeasured factors, including community safety, housing conditions, and access to health and welfare services, which may further influence children's QoL.