Exposure to an accumulation of adverse childhood experiences (ACEs) of abuse, neglect, and family dysfunction predicts poorer physical and mental health in adulthood (Felitti et al., Reference Felitti, Anda, Nordenberg, Williamson, Spitz, Edwards, Koss and Marks1998; Hughes et al., Reference Hughes, Bellis, Hardcastle, Sethi, Butchart, Mikton, Jones and Dunne2017). Furthermore, the children of parents who were exposed to a larger number of ACEs are at risk of increased behavior problems (Cooke et al., Reference Cooke, Racine, Plamondon, Tough and Madigan2019, Reference Cooke, Racine, Pador and Madigan2021). Yet most studies of the intergenerational effects of parents’ ACEs have been conducted in North America, and the generalizability of their findings to other countries and cultures is limited (Cooke et al., Reference Cooke, Racine, Pador and Madigan2021). In addition, the effects of ACEs on parenting and the pathways by which parents’ ACEs affect their parenting and their children’s behavior problems have received relatively little attention (e.g., Harris et al., Reference Harris, MacMillan, Andrews, Atkinson, Kimber, England-Mason and Gonzalez2021).
To begin to address these gaps, we focused on mothers and their infants and toddlers from three ethnocultural groups living in Israel: non-ultra-Orthodox Jewish, ultra-Orthodox Jewish, and Arab Muslim. We studied whether mothers’ ACEs would be associated with children’s behavior problems and two aspects of parenting known to have a profound impact on development: maternal sensitivity, which reflects mothers’ appropriate and prompt responsiveness to their children during interactions (Tarabulsy et al., Reference Tarabulsy, Provost, Bordeleau, Trudel-Fitzgerald, Moran, Pederson, Trabelsi, Lemelin and Pierce2009), and the home environment, measured as the quality and quantity of stimulation, support, and structure mothers provide their children (Caldwell & Bradley, Reference Caldwell and Bradley2016). To shed light on the process by which the accumulative effects of parents’ ACEs may occur, we examined whether mothers’ psychological distress mediated these associations. We also explored whether maternal sensitivity and a better home environment would buffer the mediated link between mothers’ ACEs and children’s behavior problems.
This type of research can illuminate the universality of the consequences of parents’ ACEs and the mechanisms by which parents’ ACEs shape the development of their offspring. As such, it may guide prevention and early intervention efforts to break the intergenerational transmission of adversity and trauma.
ACEs
In 1998, Felitti and colleagues published their seminal epidemiological study on the detrimental effects of exposure to an accumulation of ACEs (Felitti et al., Reference Felitti, Anda, Nordenberg, Williamson, Spitz, Edwards, Koss and Marks1998). The researchers surveyed clients of one of the largest health care companies in the United States for ten ACEs, reflecting childhood maltreatment of abuse (i.e., physical, emotional, and sexual abuse) and neglect (i.e., emotional and physical neglect), as well as household dysfunction (e.g., mental illness in the household, exposure to substance abuse in the household). The researchers found that ACEs increased the risk of poor physical and mental health conditions in adulthood in a dose-response fashion (Felitti et al., Reference Felitti, Anda, Nordenberg, Williamson, Spitz, Edwards, Koss and Marks1998).
Subsequent studies gave further evidence of the lifelong additive effects of exposure to multiple ACEs (Felitti & Anda, Reference Felitti, Anda, Lanius and Vermetten2010). ACEs were found to be associated with physiological changes in the immune, endocrine, cardiovascular, and nervous systems and to affect inflammatory and metabolic functioning, as well as brain structure and functioning (see meta-analysis in Cooke et al., Reference Cooke, Connolly, Boisvert and Hayes2023). As such, ACEs take a toll not only on health but also on the ability to function at cognitive, social, and emotional levels (Hales et al., Reference Hales, Saribaz, Debowska and Rowe2022; Hughes et al., Reference Hughes, Bellis, Hardcastle, Sethi, Butchart, Mikton, Jones and Dunne2017). Studies consistently show that exposure to a higher number of ACEs is predictive of health-harming behaviors (e.g., Bellis et al., Reference Bellis, Hughes, Leckenby, Jones, Baban, Kachaeva, Povilaitis, Pudule, Qirjako, Ulukol, Raleva and Terzic2014), physical diseases, higher rates of mortality, increased psychological distress and psychopathology, lower educational attainment, and poorer socioeconomic outcomes (see meta-analyses in Hales et al., Reference Hales, Saribaz, Debowska and Rowe2022; Hughes et al., Reference Hughes, Bellis, Hardcastle, Sethi, Butchart, Mikton, Jones and Dunne2017).
Parents’ ACEs, parenting, and children’s adjustment
In light of these pervasive effects, it is not surprising that ACEs also influence the next generation. Children whose parents were exposed to higher numbers of ACEs are at increased risk of poorer health and reduced social-emotional adjustment, particularly increased behavior problems from infancy (Madigan et al., Reference Madigan, Wade, Plamondon, Maguire and Jenkins2017) through adolescence (Doi et al., Reference Doi, Isumi and Fujiwara2022; see meta-analysis in Cooke et al., Reference Cooke, Racine, Pador and Madigan2021). To deepen the understanding of these associations, researchers have begun to explore pathways underlying these effects, and suggested mothers’ psychological distress as a potential mediating mechanism (e.g., Zhang et al., Reference Zhang, Mersky and Lee2023). Psychological distress may shape mothers’ emotional expressions and social behaviors, and these, in turn, might be modeled by the child. In addition, mothers’ psychological distress may hamper mothers’ marital relationships and other family and non-familial social relationships, with negative effects on children’s behavior (Connell & Goodman, Reference Connell and Goodman2002). Supporting these notions, a few studies have found an indirect link between mothers’ ACEs and children’s behavior problems via mothers’ elevated psychological distress (Cooke et al., Reference Cooke, Racine, Plamondon, Tough and Madigan2019; Doi et al., Reference Doi, Isumi and Fujiwara2022; Rieder et al., Reference Rieder, Roth, Musyimi, Ndetei, Sassi, Mutiso, Hall and Gonzalez2019; Zhang et al., Reference Zhang, Mersky and Lee2023). Except for one study in Japan (Doi et al., Reference Doi, Isumi and Fujiwara2022) and another in Kenya (Rieder et al., Reference Rieder, Roth, Musyimi, Ndetei, Sassi, Mutiso, Hall and Gonzalez2019), however, results were obtained in research conducted in the US and Canada, thus limiting their applicability to other cultural contexts (Cooke et al., Reference Cooke, Racine, Pador and Madigan2021).
Parenting is also considered to be a central mechanism in the intergenerational transmission of ACEs’ effect. ACEs may shape parenting in different direct and indirect ways. They may diminish the set of skills parents have and lead to distorted expectations of parenting and to distorted beliefs about how to scaffold the interactions with the child and create a developmentally appropriate environment that fosters child development (Sheffield Morris et al., Reference Sheffield Morris, Hays-Grudo, Zapata, Treat and Kerr2021). Furthermore, parents’ experiences as children with their parents are thought to shape their mental representations of attachment and caregiving relationships, and these, in turn, guide their caregiving to their child (Crowell et al., Reference Crowell, Warner, Davis, Marraccini and Dearing2010). A higher number of experiences of witnessing or experiencing abuse and a lack of emotional support from parents due to neglect or family dysfunction might lead to more negative and disorganized mental representations; parents may not be aware of these, but they are likely to color their interpretations of their child’s signals and their behavioral responses to the child (Crowell et al., Reference Crowell, Warner, Davis, Marraccini and Dearing2010; Murphy et al., Reference Murphy, Steele, Dube, Bate, Bonuck, Meissner, Goldman and Steele2014; Thomson & Jaque, Reference Thomson and Jaque2017). The increased risk of psychological distress associated with ACEs (Hales et al., Reference Hales, Saribaz, Debowska and Rowe2022) may also impede parents’ ability to self-regulate and remain emotionally available to their child, particularly when the child is distressed. This, in turn, may hamper the development of the child’s self-regulation skills and lead to increased behavior problems (Cooke et al., Reference Cooke, Racine, Plamondon, Tough and Madigan2019). At the same time, however, when parents, despite childhood adversities, succeed in providing sensitive caregiving, it may mitigate the negative effects of their child’s exposure to adversities in the family, such as those related to parental psychological distress (Narayan et al., Reference Narayan, Merrick, Lane and Larson2023).
Numerous studies suggest childhood maltreatment, one of the two core ACE types, is associated with an increased risk of abusive or neglectful parenting, insensitive caregiving, and poorer child-rearing practices (see a review in Greene et al., Reference Greene, Haisley, Wallace and Ford2020 and a meta-analysis in Madigan et al., Reference Madigan, Cyr, Eirich, Fearon, Ly, Rash, Poole and Alink2019). However, relatively few studies have examined the effects of cumulative ACEs on parenting (for a review, see Zhang et al., Reference Zhang, Mersky, Gruber and Kim2022), and these have had mixed results. While some found mothers’ multiple ACEs were associated with self-reports of poorer parenting, directly or indirectly through mothers’ psychological distress (Doi et al., Reference Doi, Isumi and Fujiwara2022; Racine et al., Reference Racine, Plamondon, Madigan, McDonald and Tough2018; Sheffield Morris et al., Reference Sheffield Morris, Hays-Grudo, Zapata, Treat and Kerr2021; Yoon et al, Reference Yoon, Cederbaum, Mennen, Traube, Chou and Lee2019), others did not find significant links (Thomas-Giyer & Keesler, Reference Thomas-Giyer and Keesler2021) or found evidence of such associations in some countries but not others (Brown et al., Reference Brown, Eisner, Walker, Tomlinson, Fearon, Dunne, Valdebenito, Hughes, Ward, Sikander, Osafo, Madrid, Baban, Van Thang, Fernando and Murray2021).
Furthermore, only a small number of studies have gone beyond parents’ self-reports of their parenting to evaluate the effects of ACEs on the quality of mother-child observed interactions, and these also yielded inconsistent results. One found mothers’ ACEs were associated with lower emotional availability (i.e., maternal sensitivity, structuring, non-intrusiveness, non-hostility; Biringen et al., Reference Biringen, Derscheid, Vliegen, Closson and Easterbrooks2014), but only when children were 18 months old, not when they were 60 months old (Harris et al., Reference Harris, MacMillan, Andrews, Atkinson, Kimber, England-Mason and Gonzalez2021). Two others did not find significant links between mothers’ ACEs and maternal sensitivity towards their infants (Coe et al., Reference Coe, Huffhines, Contente, Seifer and Parade2020) or emotional availability towards their preschoolers (Ziv et al., Reference Ziv, Umphlet, Olarte and Venza2018). Additional studies have reported that mothers’ ACEs indirectly led to lower maternal sensitivity, through its effect on mothers’ incoherent narratives on childhood attachment experiences (Crowell et al., Reference Crowell, Warner, Davis, Marraccini and Dearing2010) and increased psychological distress (Bouvette-Turcot et al., Reference Bouvette-Turcot, Fleming, Unternaehrer, Gonzalez, Atkinson, Gaudreau, Steiner and Meaney2020). Finally, a recent study showed that although mothers’ ACEs were not associated with their emotional availability, they were indirectly associated with children’s behavior problems via mothers’ psychological distress, and this mediated path was moderated by mothers’ emotional availability; it was significant for mothers who showed low emotional availability but not for those who showed high emotional availability (Wurster et al., Reference Wurster, Sarche, Trucksess, Morse and Biringen2020). Given the inconsistent findings, further research on the relations between mothers’ ACEs and observed mother-child interactions is called for.
Another core aspect of observed parenting known to have a significant effect on children’s development that is possibly affected by ACEs is the home environment (see a review in Bradley, Reference Bradley2015). An adequate home environment includes developmentally appropriate home transactions (e.g., avoiding harsh punishment), experiences with objects (e.g., toys and books), events (e.g., visiting family and friends), and arrangements (e.g., being taken regularly to the doctor; Bradley, Reference Bradley2015). Parents who experienced abuse, neglect, and family dysfunction as children may unintentionally provide their own children with an inadequate home environment. This link could be indirect via parental psychological distress. At the same time, if parents manage to provide an adequate home environment, it may protect the child from the effects of exposure to negative experiences in the family, such as exposure to elevated parental distress. Studies outside the realm of ACE research have indicated that parental psychological distress is associated with less adequate home environments (Bradley, Reference Bradley2015). Yet we are aware of only one study on the associations between ACEs and the home environment. The study assessed mothers’ childhood experiences of maltreatment and neglect, but not family dysfunction, and did not find support for either a direct association or an indirect link through mothers’ psychological distress (Ammerman et al., Reference Ammerman, Shenk, Teeters, Noll, Putnam and Van Ginkel2012).
Overall, more studies on the effect of mothers’ ACEs on maternal sensitivity, the home environment, and children’s behavior problems and the processes by which these effects occur are needed, especially in a non-North American context. Our study begins to fill the gap.
The context of the study: ethnocultural groups in Israel
The study focused on mothers and children from three ethnocultural groups in Israel. The first group comprised non-ultra-Orthodox Jews, the majority in Israel. Members of this group adhere to individualist values of autonomy, equality, and self-fulfillment (Sher-Censor, Reference Sher-Censor2015). The two additional groups were ultra-Orthodox Jews and Arab Muslims, who constitute 13.3% and 18.1% of the Israeli population, respectively (Israel Central Bureau of Statistics, 2023). The two groups are close-knit and follow collectivist, patriarchal, and authoritarian values (Gemara & Nadan, Reference Gemara and Nadan2022; Mizrachi & Weiss, Reference Mizrachi and Weiss2020). Marriage and transition to parenting occur at earlier ages than in the non-ultra-Orthodox Jewish group, and rates of single parenting and divorce are lower (Israel Central Bureau of Statistics, 2023).
Ultra-Orthodox Jews have intense religious beliefs and adhere to strict ritual laws and cultural mores. They reside in self-isolated communities and neighborhoods, maintain a separate educational system, isolate themselves from secular media, and strive to provide for their own needs using internal organizations to protect the religious Jewish heritage from secular influences. In many ultra-Orthodox families, women provide the primary financial support, as their spouses refrain from working and engage in full-time religious studies (Gemara & Nadan, Reference Gemara and Nadan2022; Simhi et al., Reference Simhi, Yoselis, Sarid and Cwikel2020).
The majority of Arab Muslims are self-defined as traditional religious or religious (Israel Central Bureau of Statistics, 2023). They reside in separate towns, villages, and neighborhoods and maintain a separate education system (Israel Central Bureau of Statistics, 2023). Arab Muslim communities are organized around patriarchal extended families (Mizrachi & Weiss, Reference Mizrachi and Weiss2020). Yet this society is undergoing a process of modernization that involves a change of values towards an increased emphasis on autonomy and equality, including a gradual increase in women’s empowerment (Shoshana & Shchada, Reference Shoshana and Shchada2018).
There is little research on mother-child relationships and children’s adjustment among ultra-Orthodox Jews and Arab Muslims (Feldman & Masalha, Reference Feldman and Masalha2010; Sher-Censor et al., Reference Sher-Censor, Dolev, Said, Baransi and Amara2017, Reference Sher-Censor, Dan Ram-On, Rudstein-Sabbag, Watemberg and Oppenheim2020; Zreik et al., Reference Zreik, Oppenheim and Sagi-Schwartz2017). Furthermore, past studies raised concerns about the tendency of both groups to conceal or ignore abuse, neglect, and family dysfunction, partly to protect family members from stigma and partly because of mistrust in social services (Nadan et al., Reference Nadan, Gemara, Keesing, Bamberger, Roer-Strier and Korbin2019). Our study provided an opportunity to examine these understudied ethnocultural groups in a comparative context.
Study hypotheses
The aim of the study was to examine whether and how Israeli mothers’ ACEs shape their parenting and their children’s behavioral problems and explore the pathways by which these effects occur. We examined two models. The first focused on parenting and hypothesized that mothers’ exposure to more ACEs would be associated with lower maternal sensitivity and poorer quality of the home environment. We expected that these associations would be indirect, through mothers’ psychological distress. The second model focused on children’s behavior problems. We hypothesized that the link between mothers’ ACEs and higher levels of children’s behavior problems would be indirect through maternal psychological distress, and this indirect effect would be moderated by maternal sensitivity and the home environment. Specifically, we expected the link between mothers’ psychological distress and children’s behavior problems would be evident only when mothers showed lower sensitivity or provided their children with a poorer home environment. Finally, given the absence of prior relevant research on ultra-Orthodox Jews and Arab Muslims, we examined ethnocultural differences with no a-priori hypothesis.
Method
Participants
Participants were 232 Israeli mothers and their children, referred to a national early intervention project for parents at risk of providing poor parenting. In the sample, 63.36% of the mothers were non-ultra-Orthodox Jewish, 17.24% were ultra-Orthodox Jewish, and 19.40% were Arab Muslim. Mothers’ ages ranged from 19 to 44 years (Mean = 28.23, SD = 5.59). Mothers’ years of education ranged from 6 to 18 years (Mean = 13.07, SD = 1.91). Most mothers were married or cohabiting (65.95%). The average number of children in the family was 1.39 (SD = .58, Range = 1–3), and 50.86% of the children were female. Children’s ages ranged from 1 to 36 months (Mean = 18.40, SD = 10.76). Thirty-one (13.36%) children had a medical (e.g., heart defects, hydrocephalus, developmental dysplasia of the hip) or developmental (e.g., language or developmental delay) diagnosis. Families were referred to the intervention for various reasons, including single parenting, living in extreme poverty, or mothers’ reports of difficulty parenting the child or significant marital discord. Mothers with psychiatric diagnoses, those who were currently experiencing marital violence, and those who were identified by child protection services as neglecting or abusing their child were referred to other intervention programs and were excluded from the study.
Procedure
The study was part of larger longitudinal research evaluating the intervention project. Data presented herein were collected during 2020–2022, before mothers received the intervention. Information on ACEs was collected by therapists as part of the intake process. Research assistants collected mothers’ self-reports, including reports on children’s behavior problems and family demographics. A home visit was scheduled next to collect observational data. In line with the Home Observation for Measurement of the Environment (HOME) protocol (Caldwell & Bradley, Reference Caldwell and Bradley2016), visits were scheduled at a time when the mother was at the home and the child was awake. The home observation was conducted first. Then, mothers were asked to play with their child for 10 minutes, as they usually played, to evaluate maternal sensitivity. To maintain the blindness of the coders of the home observation to maternal sensitivity, the home observation was conducted before mother-child play interaction. Play interactions were videotaped and coded later by independent blind coders, except for 19 interactions that were coded during the home visit, because mothers refused to be videotaped. Therapists and research assistants were native speakers of the native language of the family. Tasks were completed in the mothers’ native language. The study was approved by the institutional review board of the University of Haifa (Approval #424/19).
Measures
Maternal ACEs
Therapists used the widely used Adverse Childhood Experiences Checklist (Center for Disease Control, 2016; Felitti et al., Reference Felitti, Anda, Nordenberg, Williamson, Spitz, Edwards, Koss and Marks1998) to ask mothers, as part of the intake interview, whether before the age of 18 years, they experienced the following 10 ACEs: physical abuse, emotional abuse, sexual abuse, physical neglect, emotional neglect, parental separation or divorce, domestic violence, addictions in the household, mental illness in the household, or incarceration of a member of the household. Scores were calculated by adding the total number of “yes” responses.
Maternal sensitivity
The Maternal Behaviour Q-sort - Short Version (MBQS; Tarabulsy et al., Reference Tarabulsy, Provost, Bordeleau, Trudel-Fitzgerald, Moran, Pederson, Trabelsi, Lemelin and Pierce2009) was used to assess maternal sensitivity. The 25-item MBQS is an adaptation of the original 90-item Maternal Behaviour Q-sort (MBQ; Pederson & Moran, Reference Pederson and Moran1996). It is designed to code maternal sensitivity from video recorded or live interactions. Trained coders rated 25 items describing mothers’ appropriate and warm responses to the child (e.g., “Respects child as an individual, i.e., able to accept child’s behaviour even if it is not consistent with her wishes”). Coding involved sorting the items into five piles, with five items placed in each pile. Piles ranged from most to least characteristic of the mother, yielding a score of 1–5 for each item.
The sensitivity score of each mother was calculated based on the correlation between the mother’s sort scores and the criterion sort of the prototypically sensitive mother provided by the developers of the MBQ (Pederson & Moran, Reference Pederson and Moran1996; Tarabulsy et al., Reference Tarabulsy, Provost, Bordeleau, Trudel-Fitzgerald, Moran, Pederson, Trabelsi, Lemelin and Pierce2009). Thus, scores could theoretically range from −1.00 to 1.00. A higher correlation reflected that the mother was more sensitive. The validity of the MBQ and MBQS has been supported in several studies (e.g., Booth et al., Reference Booth, Greenwood, Youssef, McIntosh, Nguyen, Letcher, Edwards, Hutchinson, Sanson, Olsson and Macdonald2023), including among Arab-Israeli families (Zreik et al., Reference Zreik, Oppenheim and Sagi-Schwartz2017). To establish reliability in the current study, 32% of the interactions were rated by two independent blind coders. The average correlation between the two coders' arrays of scores was .83 (SD = .10, range = .50–1.00). Disagreements between coders were resolved through discussion.
The home environment
To assess the home environment provided by the mothers, we used the HOME (Caldwell & Bradley, Reference Caldwell and Bradley1976, 2016). The HOME combines a natural home observation with a parental interview. The infant and toddler version of the HOME used in this study includes 45 items tapping six domains: (1) mothers’ responsivity to the child (e.g. “Parent spontaneously vocalizes to child at least twice”); (2) mothers’ acceptance of the child (e.g., “Parent does not scold or criticize the child during the visit”); (3) organization of the child’s environment in terms of a safe physical environment and regular and predictable schedule (e.g., “Child’s play environment is safe”; “Child is taken regularly to doctor’s office or clinic”); (4) provision of appropriate play and learning materials that can stimulate child development (e.g., “Push or pull toys are available to the child”); (5) maternal involvement in child learning (e.g. “Parent keeps child in visual range, looks at often”); and (6) variety or the extent to which mothers include a variety of people and events in the child’s daily life (e.g., “Family visits relatives or receives visits once a month or so”). Observers rated each item as 1 (“yes”) or 0 (“no”), and scores were summed. Higher scores reflected a better home environment.
The HOME has been used in numerous studies, including in Israel (e.g., Schwartz & Bilsky, Reference Schwartz and Bilsky1990). Cronbach’s alpha of the HOME items in our study was good, alpha = .82. Blind research assistants conducted and coded the HOME. As part of their training, they achieved at least 90% agreement on five mother-child pilot cases. To ensure coders’ reliability, seven (5.98%) home visits were conducted by pairs of blind research assistants. Interrater reliability of their HOME scores was excellent (Intraclass Correlation Coefficient = .99). Coding disagreements were resolved by consensus.
Children’s behavior problems
Mothers were invited to complete the Problem Scale of the Brief Infant Toddler Social Emotional Assessment (BITSEA; Briggs-Gowan & Carter, Reference Briggs-Gowan and Carter2006). The scale assesses social-emotional problems of toddlers aged 12–36 months (Briggs-Gowan & Carter, Reference Briggs-Gowan and Carter2006). It includes 31 items tapping externalizing and internalizing problems, dysregulation, maladaptive behaviors, and atypical behaviors (e.g., “Seems nervous, tense, or fearful”; “Is destructive. Breaks or ruins things on purpose”). Mothers were asked to rate their child’s behavior in the last month on a scale ranging from 0 (“not true/rarely”) to 2 (“very true/often”). For some items, an option of N (“no opportunity”) is available. The scale showed good internal consistency (Cronbach’s alpha = .79). Scores were summed, with higher scores reflecting more behavior problems.
Maternal psychological distress
Mothers completed the Depression and Anxiety scales of the Depression, Anxiety and Stress Scale - 21 Items (DASS-21; Lovibond & Lovibond, Reference Lovibond and Lovibond1995) to assess their psychological distress. They rated the extent to which they experienced seven symptoms of depression (e.g., “I felt down-hearted and blue”) and seven symptoms of anxiety (e.g., “I felt scared without any good reason”) over the past week on a 4-point Likert scale ranging from 0 (“does not apply”) to 3 (“applies very much”). Scores on each scale were summed. The scales demonstrated good internal consistency (Cronbach’s alpha Depression = .82; Cronbach’s alpha Anxiety = .82). The DASS is widely used, including in its Hebrew and Arabic versions (e.g., Ali et al., Reference Ali, Ahmed, Sharaf, Kawakami, Abdeldayem and Green2017; Sher-Censor et al., Reference Sher-Censor, Dan Ram-On, Rudstein-Sabbag, Watemberg and Oppenheim2020). It was used in previous Israeli research on ultra-Orthodox Jews (Sher-Censor et al., Reference Sher-Censor, Dan Ram-On, Rudstein-Sabbag, Watemberg and Oppenheim2020) and Arab Muslims (e.g., Alfayumi-Zeadna et al., Reference Alfayumi-Zeadna, Gnaim-Abu Touma, Weinreich and O’Rourke2022). Depression and anxiety scores were significantly and strongly correlated (r = .66, p < .001) and thus were averaged to form mothers’ psychological distress. Higher scores reflected higher psychological distress.
Missingness and analytic plan
Percentage of missing questionnaire data due to technical errors ranged from 1.29% for marital status to 7.33% for mothers’ years of education. In addition, as the minimal age for which the BITSEA is valid is 12 months, 73 mothers (31.47%) whose children were younger than 12 months did not report their child’s behavior problems. These families did not differ from families with a child aged 12 months and up in any of the background variables (all ps > .098) except for the number of children in the family. Families with children older than 12 months had more children (Mean = 1.53, SD = .64) than families with children younger than 12 months (Mean = 1.10, SD = .30), t (164.23) = 5.88, p < .001.
Due to resource limitations, only 50% of the participating families, selected randomly, were visited at home. Accordingly, maternal sensitivity was observed for 123 families (53.02%). HOME observations were collected for 107 of these 123 families (46.12%). The remaining 16 families (6.90%) joined the intervention during periods of COVID-19 lockdowns, so a home visit could not be conducted. For these families, maternal sensitivity was recorded via a Zoom meeting. Families who did not participate in the maternal sensitivity assessment did not differ from those who participated in any background variables (all ps > .064) except for one unexpected difference. Mothers who did not participate in the maternal sensitivity assessment had fewer years of education (Mean = 12.61, SD = 1.87) than mothers who participated (Mean = 13.41, SD = 1.87), t (213) = 3.08, p = .002.
Preliminary analyses involving correlations, t-tests, and ANOVA informed the inclusion of background variables in hypotheses testing. Descriptive statistics for the prevalence of each ACE were examined next, followed by correlations analysis to determine whether the conditions for models with indirect effects were met, namely a significant link between mothers’ ACEs and psychological distress and between psychological distress and the outcome variables (maternal sensitivity, home environment, children’s behavior problems). If the conditions were not met, subsequent mediation analysis was not performed. A regression analysis was conducted instead to examine whether the bivariate link between mothers’ ACEs and the outcome variable was significant when controlling for ethnocultural background and relevant covariates identified in the preliminary analyses. If the conditions for models with indirect effects were met, mediation and moderated mediation models were analyzed using Hayes’s (2018) SPSS PROCESS routine. This routine estimates the significance of indirect effects with a bootstrap approach, more specifically, a nonparametric method based on repeated random resampling with replacement, yielding 95% bootstrapped confidence intervals (CIs) for the indirect effect. A significant indirect effect is evident when zero is outside the 95% CIs, indicating the mediation effect is statistically different from zero at p < .05. Continuous variables were centered, and analyses controlled for ethnocultural background and relevant demographic variables.
Results
Preliminary analyses
The distribution of all variables was sufficiently normal to render parametric statistics valid (Afifi et al., Reference Afifi, Kotlerman, Ettner and Cowan2007). Descriptive statistics of study variables are shown in Table 1.
Table 1. Descriptive statistics of study variables
Background and study variables did not vary by ethnocultural group (all ps > .210), except for two differences, which were expected and in line with reports of the Israeli Central Bureau of Statistics (2023). First, single-parent status was more prevalent among non-ultra-Orthodox Jewish mothers than among ultra-Orthodox Jewish and Arab Muslim mothers, χ 2 (2) = 13.04, p < .001. Second, non-ultra-Orthodox Jewish mothers were older (M = 29.93, SD = 5.92) than ultra-Orthodox Jewish mothers (M = 24.89, SD = 2.73, p < .001) and Arab Muslim mothers (M = 25.37, SD = 3.44, p < .001; F [2, 222] = 22.58, p < .001). It should be noted that maternal years of education and the home environment also varied by ethnocultural background, F (2, 212) = 3.03, p = .050 and F (2, 104) = 3.77, p = .026 respectively. Yet none of the post hoc comparisons was significant (ps > .051 and ps > .108, respectively).
There were no significant associations between other background variables and the study variables (all ps > .081), except for the following. Mothers who reported more ACEs had fewer years of education (r = −.25, p < .001). Older mothers and children, as well as more years of education, were associated with better HOME scores (r = .36, p > .001, r = .32, p <= .001, and r = .26, p = .007 for older mothers, older children, and more education, respectively). Finally, mothers of boys (Mean = 13.87, SD = 6.84) reported more child behavior problems than mothers of girls (Mean = 10.67, SD = 5.83), t (140) = 3.01, p = .034.
Mothers’ age was significantly related to two covariates: ethnocultural background (mentioned above) and mothers’ years of education (r = .23, p < .001). Therefore, to minimize collinearity concerns, we did not include maternal age in further analyses. In accordance with the preliminary analyses, ethnocultural background, mothers’ years of education, child age, and child gender were included as covariates in subsequent analyses.
Prevalence of mothers’ ACEs
As shown in Table 1, mothers reported, on average, 2.56 adverse experiences. The majority (72.41%) experienced at least one ACE and a third (32.33%) reported four or more ACEs. As can be seen in Figure 1, the most frequently reported ACEs were emotional abuse and emotional neglect. The least frequently reported were physical neglect and an incarcerated relative.
Figure 1. Percentage of mothers’ adverse childhood experiences in the sample (N = 232).
Predicting mothers’ psychological distress, maternal sensitivity, home environment, and children’s behavior problems from mothers’ ACEs
Correlation analyses
As shown in Table 2 and in line with the study’s hypothesis, mothers who experienced more ACEs reported increased psychological distress, showed lower sensitivity during play interactions, provided a poorer home environment, and reported more child behavior problems. As hypothesized, mothers’ psychological distress was associated with their reports of more child behavior problems. Unexpectedly, mothers’ psychological distress was not significantly correlated with their sensitivity to the child or with the home environment. Thus, there was no justification for examining the mediating role of mothers’ psychological distress in the links between mothers’ ACEs and their sensitivity and the home environment.
Table 2. Inter-correlations of study variables
Note. *p < .05. **p < .01.
Regression analyses
We ran two regression analyses to predict mothers’ sensitivity and the home environment from mothers’ ACEs while controlling for ethnocultural background and relevant covariates identified in preliminary analyses. Ethnocultural background was included as two dummy variables, Jewish/Arab and ultra-Orthodox/non-ultra-Orthodox. Psychological distress was included as a covariate because, as noted above, it was significantly correlated with mothers’ ACEs. As hypothesized, the regression predicting maternal sensitivity indicated that after controlling for mothers’ ethnocultural background, years of education, and psychological distress, which were entered in the first block (ΔR 2 = .03, F (4,113) = 0.97, p = .426), mothers’ increased ACEs predicted lower maternal sensitivity (ΔR 2 = .05, F (1,112) = 6.47, p = .012, β = −.24, t = −2.54, p = .012).
Unexpectedly however, the second regression predicting the home environment indicated that after controlling for mothers’ ethnocultural background, years of education, psychological distress, and child age, which were entered in the first block (ΔR 2 = .27, F (5,96) = 7.19, p < .001), mothers’ ACEs did not significantly predict the home environment (ΔR 2 = .003, F (1,95) = .46, p = .501, β = −.06, t = −.68, p = .501).
Moderated mediation analyses
Finally, to examine the hypothesis that mothers’ ACEs would be indirectly associated with their reports of their children’s behavior problems via mothers’ psychological distress, and this indirect pathway would be moderated by mothers’ sensitivity and the home environment, we conducted two PROCESS analyses of moderated mediation (model 14). Analyses included the two dummy variables of ethnocultural background (Jewish/Arab, ultra-Orthodox/non-ultra-Orthodox), mothers’ years of education, and child gender as covariates. As shown in Figures 2 and 3, and in line with our hypothesis, the first analysis supported a moderated mediation effect (N = 80, b = −.62, 95% percentile CI [−1.73, −0.03]). The link between mothers’ ACEs and higher levels of children’s behavior problems was indirect through mothers’ psychological distress, but only for mothers with 1SD below the average sensitivity scores (b = .33, 95% percentile CI [0.02, 0.92]), not for mothers with average or above average sensitivity scores (b = .11, 95% percentile CI [−0.06, 0.39]; and b = −.08, 95% percentile CI [−0.38, 0.15] respectively). It should be noted that the direct link between mothers’ ACEs and child behavior problems was significant (see Figure 2), indicating a partial mediation effect of mothers’ psychological distress.
Figure 2. Indirect effects between mothers’ adverse childhood experiences (ACEs) and children’s behavior problems through mothers’ psychological distress, conditional on mothers’ sensitivity (N = 80). b = unstandardized beta coefficient of the path between variables. b’ = unstandardized beta coefficient of the direct path between maternal ACEs and children’s behavior problems. SE = standard error. Covariates include ethnocultural background, mothers’ education, and child gender. *p < .05. **p < .01.
Figure 3. Associations between mothers’ psychological distress and their reports of behavior problems in their children by maternal sensitivity (N = 80). The solid line represents a significant simple slope (p < .05). Dashed lines represent nonsignificant simple slopes (p > .05).
The second model examining the moderating role of the home environment in the indirect pathway between mothers’ ACEs and child behavior problems via mothers’ psychological distress suggested the moderating effect was not significant (N = 67, b = −.01, 95% percentile CI [−0.08, 0.01]).
Discussion
To the best of our knowledge, this was the first large-scale study in Israel to examine adults’ ACEs and their effects. The study focused on mothers who were referred to early intervention to improve their parenting. In our study, less than a third of the mothers reported they did not experience any ACEs. A third reported four or more ACEs, a score that is often used clinically as a cutoff point reflecting a high risk for poorer physical and mental health (Felitti et al., Reference Felitti, Anda, Nordenberg, Williamson, Spitz, Edwards, Koss and Marks1998; Hughes et al., Reference Hughes, Bellis, Hardcastle, Sethi, Butchart, Mikton, Jones and Dunne2017). The high rate of ACEs among mothers in our study is similar to that documented for Head Start parents in the US (Randell et al., Reference Randell, O’Malley and Dowd2015). This could be expected, as parents in both cases were considered at risk in terms of their ability to provide nurturing conditions for their child’s development. Yet, these rates are considerably higher than those typically reported in epidemiological and general population research in North America and Europe, i.e., 30%–60% with no ACEs and 7%–15% with four or more ACEs (Bellis et al., Reference Bellis, Hughes, Leckenby, Jones, Baban, Kachaeva, Povilaitis, Pudule, Qirjako, Ulukol, Raleva and Terzic2014; Racine et al., Reference Racine, Plamondon, Madigan, McDonald and Tough2018). It should be noted, however, that similarly high rates were found in the general population of females and males in Saudi Arabia (Almuneef et al., Reference Almuneef, Hollinshead, Saleheen, AlMadani, Derkash, AlBuhairan, Al-Eissa and Fluke2016) and in expectant mothers in middle-income countries (Brown et al., Reference Brown, Eisner, Walker, Tomlinson, Fearon, Dunne, Valdebenito, Hughes, Ward, Sikander, Osafo, Madrid, Baban, Van Thang, Fernando and Murray2021), although both studies utilized extended ACE self-reports. An epidemiological study among the general population in Israel is needed for a better understanding of the rates documented in our research.
ACEs and maternal sensitivity
Our research was among the first to examine the effects of ACEs on parenting outside North America and to extend research beyond mothers’ self-reports to include observations of mothers’ sensitivity during mother-child play interactions. Results indicate that maternal sensitivity plays a central role in the effects of ACEs on parents and their children. In our study, mothers’ exposure to multiple ACEs was related to lower maternal sensitivity towards the child. That is, mothers with increased ACEs tended to conduct less synchronous play interactions and were less attuned to their child’s signals and pace. They provided less appropriate structuring of their children’s attention and behavior and showed lower acceptance of their children’s behaviors that were not consistent with maternal wishes (Tarabulsy et al., Reference Tarabulsy, Provost, Bordeleau, Trudel-Fitzgerald, Moran, Pederson, Trabelsi, Lemelin and Pierce2009).
Numerous studies have suggested maternal sensitivity has enduring effects on various aspects of children’s development. These include brain morphology (Bernier et al., Reference Bernier, Dégeilh, Leblanc, Daneault, Bailey and Beauchamp2019), the formation of secure attachment (see a review in Zeegers et al., Reference Zeegers, Colonnesi, Stams and Meins2017), language development (Bornstein et al., Reference Bornstein, Putnick, Bohr, Abdelmaseh, Lee and Esposito2020), and socioemotional adjustment (Raby et al., Reference Raby, Roisman, Fraley and Simpson2015). Our study joins research efforts to identify mechanisms that shape maternal sensitivity and suggests the accumulation of ACEs may be a factor underlying this aspect of parenting.
Following the study of Bouvette-Turcot and colleagues (2020), we expected that ACEs’ effects on maternal sensitivity would be mediated by mothers’ psychological distress. However, we did not find support for this indirect link. Mothers’ ACEs were linked to poor maternal sensitivity above and beyond their psychological distress. These results emphasize the important role of ACEs in sensitive parenting. Future research examining other potential mechanisms that may underlie the impact of maternal ACEs on maternal sensitivity may help to guide prevention and early intervention efforts in mothers exposed to multiple ACEs. These mechanisms may include changes in mothers’ nervous and endocrine systems that are shaped by ACEs (see a review in Cooke et al., Reference Cooke, Connolly, Boisvert and Hayes2023), found by past studies, outside the realm of ACE research, to be related to insensitive parenting (see a review in Bos, Reference Bos2017). Other potential mediators are mothers’ knowledge of child development and their child-rearing attitudes. These factors are thought to be associated with quality of parenting (Bornstein et al., Reference Bornstein, Cote, Haynes, Hahn and Park2010), and a previous study found child-rearing attitudes were related to mothers’ ACEs (Sheffield Morris et al., Reference Sheffield Morris, Hays-Grudo, Zapata, Treat and Kerr2021). If knowledge of child development and child-rearing attitudes indeed mediate ACEs’ effects on maternal sensitivity, they may be relatively easily targeted in parental interventions.
ACEs, maternal sensitivity, and children’s behavior problems
Importantly, our findings suggest mothers’ sensitivity plays a central buffering role against the transmission of ACEs’ effects to the next generation. Consistent with results in North America (Cooke et al., Reference Cooke, Racine, Pador and Madigan2021; Zhang et al., Reference Zhang, Mersky and Lee2023), we found Israeli mothers’ exposure to multiple ACEs was associated with increased behavior problems in their children as early as toddlerhood, and this link was mediated by mothers’ higher psychological distress. Furthermore, similar to Wurster and colleagues (2020), we found the mediation was conditional on mothers’ sensitivity. Although our study did not have a longitudinal design, and causal interpretations are tentative, our results shed light on the possible process by which early adverse experiences may shape children’s wellbeing. Our study suggests that increased ACEs may lead to elevated maternal psychological distress, but mothers’ sensitivity may buffer the effects of their psychological distress on the emergence of behavior problems in their toddlers. That is, when mothers can, despite their distress, be available to their child, attuned to the child’s needs, and set clear and consistent limits, the child may not develop behavior problems. When mothers are distressed and less sensitive, the child may lack the necessary parental emotional support, scaffolding, and limit-setting and develop more behavior problems.
It should be noted that the direct effect of mothers’ ACEs on children’s behavior problems remained significant after considering the moderated mediation effect. This suggests other mechanisms beyond mothers’ psychological distress and their sensitivity may underlie this link. Future studies could examine the potential mediating role of mothers’ physiological dysregulation that may result from ACEs and be transmitted to the child even during pregnancy, leading, in turn, to increased child behavior problems (Bos, Reference Bos2017; Thomas-Argyriou et al., Reference Thomas-Argyriou, Letourneau, Dewey, Campbell and Giesbrecht2021). Another potential avenue of future research involves negative and incoherent mental representations of the parent-child relationship that may result from mothers’ adverse experiences with their parents (Crowell et al., Reference Crowell, Warner, Davis, Marraccini and Dearing2010). Such representations may lead to negative expectations of the child and color mothers’ interpretation of the child’s behavior, possibly contributing to a negative cycle that is perpetuated by the child’s behavior (Groh et al., Reference Groh, Fearon, van IJzendoorn, Bakermans-Kranenburg and Roisman2017).
ACEs and home environments
The third aspect of mother-child relationships we examined tapped a broader aspect of observed parenting: the home environment. We hypothesized that mothers’ ACEs would hamper their ability to provide their children with a daily routine and verbal, physical, and social stimuli that are varied, organized, and developmentally appropriate (Bradley, Reference Bradley2015), and this link would be mediated by mothers’ psychological distress. We also hypothesized that a better home environment would buffer against the effects of mothers’ ACEs on their children’s behavior problems. We found little support for these hypotheses. Correlation analyses indicated that mothers with increased ACEs provided a poorer home environment. Yet this link became nonsignificant when controlling for family background variables. In addition, mothers’ psychological distress was not associated with the home environment, and the home environment did not moderate the indirect link between mothers’ ACEs and their children’s higher levels of behavior problems.
We are aware of only one study that examined one aspect of ACEs, mothers’ experience of childhood maltreatment, in relation to the home environment, and it also failed to find significant associations (Ammerman et al., Reference Ammerman, Shenk, Teeters, Noll, Putnam and Van Ginkel2012). In light of the known effect of the home environment on children’s secure attachment, cognitive performance, and socioemotional development (for a review, see Bradley, Reference Bradley2015), and as our study found initial evidence of a bivariate association with mothers’ ACEs, we suggest more research should be conducted on this link with a larger sample. A larger sample may have increased power to capture more nuanced links that our study could not assess, such as the effects of specific aspects of ACEs (e.g., family dysfunction) on the different aspects of the home environment (e.g., organization and stimuli).
Strengths, limitations, and future directions
This study was unique in its sample, as it included two understudied ethnocultural groups: ultra-Orthodox Jews and Arab Muslims. Notably, there were no significant ethnocultural differences in ACE levels, maternal sensitivity, home environment, and children’s behavior problems. Furthermore, the association of mothers’ ACEs with lower maternal sensitivity and with increased child behavior problems, as well as the moderating mediating effect of mothers’ psychological distress in the link between mothers’ ACEs and more behavior problems of the child, remained significant when controlling for ethnocultural background. This strengthens the notion that ACEs may have universally harmful effects. Nevertheless, our lack of significant findings should be interpreted with caution. It could reflect lack of power due to the small size of our ethnocultural subsamples. Importantly, the subsamples were not large enough to examine the moderating role of ethnocultural background. This should be considered in future studies.
The lack of ethnocultural differences in ACE levels may suggest a similar prevalence of ACEs across ethnocultural groups. It may also suggest a similar level of mothers’ openness in reporting them. As noted earlier, researchers have raised concerns about the tendency of ultra-Orthodox Jews and Arab Muslims to conceal or ignore abuse, neglect, and family dysfunction (Nadan et al., Reference Nadan, Gemara, Keesing, Bamberger, Roer-Strier and Korbin2019). Recall that in our study, ACEs were collected by the families’ therapists. For the most part, therapists did not reside in the same town/village/neighborhood as the mothers. Yet they shared a similar ethnocultural background. Coupled with the fact that ACEs were collected as part of the intake, after an initial rapprochement was likely established with the mothers, this practice may have helped to reduce barriers and facilitate mothers’ reports on past adverse experiences. This may guide future practice with families from close-knit, traditional backgrounds.
Future studies may benefit from addressing the limitations of the study. These include a shared source correlation, namely mothers’ self-reports of both their own ACEs and their children’s behavior problems, as well as a focus on mothers without including fathers. In addition, mothers who participated in this study had a moderate level of at-risk parenting. Although the variability in our study’s measures was relatively high, range restriction might have led to attenuated estimates of the associations between variables. Future studies should include more heterogenous samples and a wider range of mothers, both mothers in the community who are considered at low risk in terms of their parenting and mothers who are considered at higher risk, such as mothers who are maltreating or neglecting their children or have a psychiatric disorder. Another limitation was the relatively small sample for which we conducted observations (due to resource limitations mentioned previously). This precluded our ability to explore the moderating role of ethnocultural background, nor could we study the effects of discrete types of ACEs of child maltreatment versus family dysfunction (Sayyah et al., Reference Sayyah, Merrick, Larson and Narayan2022) or explore pairs of ACEs that are thought to synergically increase the risk of poor mental health, such as sexual and physical abuse (Briggs et al., Reference Briggs, Amaya-Jackson, Putnam and Putnam2021). In a related context, researchers have called for extending the assessment of ACEs to other experiences, such as peer victimization (Sayyah et al., Reference Sayyah, Merrick, Larson and Narayan2022), neighborhood violence, and residential instability (Usacheva et al., Reference Usacheva, Choe, Liu, Timmer and Belsky2022). This line of research has yielded a three-dimensional model of childhood adversity of deprivation, threat, and unpredictability. Recent large-scale studies indicated that each of these dimensions uniquely affected different aspects of children’s development (Ellis et al., Reference Ellis, Sheridan, Belsky and McLaughlin2022; Narayan et al., Reference Narayan, Merrick, Lane and Larson2023; Usacheva et al., Reference Usacheva, Choe, Liu, Timmer and Belsky2022). This would be another important avenue for future research with larger samples. Finally, there is growing evidence that positive childhood experiences buffer or decrease the negative effects of ACEs on adults’ wellbeing (Narayan et al., Reference Narayan, Merrick, Lane and Larson2023). Future studies should consider whether they moderate the associations of ACEs with parenting and child behavior problems.
Implications
Our study offers several conclusions that can guide future research and practice. First, mothers were not referred to early intervention because of childhood trauma. Nevertheless, the study revealed a high level of multiple ACEs among these mothers and documented their toxic effects on the mother-infant/toddler relationship and toddlers’ socioemotional functioning. Notably, the most frequent types of ACEs reported in this study were emotional abuse and neglect. Both are more difficult to detect than other ACEs. The study therefore emphasizes the importance of including ACE assessments in research on early parenting, as well as in routine screening and as part of the intake process in clinical interventions, even when ACEs are not the reason for referral. It is also vital to increase the awareness of professionals who meet regularly with parents of infants and toddlers (e.g., genecologists, pediatricians, nurses at family health centers, educators who work with young adults or with infants and toddlers) and teach them how to ask about ACEs and how to respond when parents report being exposed to ACEs. Discussing ACEs with parents may be anxiety-provoking for researchers and professionals (Becker-Blease & Freyd, Reference Becker-Blease and Freyd2006), but research suggests that asking about ACEs is likely to arouse relatively little distress in clients and is perceived by them as highly important (Cromer et al., Reference Cromer, Freyd, Binder, DePrince and Becker-Blease2006). Finally, asking about ACEs must go hand in hand with providing parents with professional support or referring them to clinical interventions to prevent the harmful effects of ACEs on parenting and the socioemotional development of children.
Funding statement
This research was supported by the Israel Social Security Agency and the Yedidut Toronto Foundation. We would like to thank Mali Gut for her help in the study as part of her master’s thesis. We express our sincere gratitude to the families who participated in this research and their therapists who helped collect the data.
Competing interests
None.
Open access
