An unbreakable bond? Maternal adverse childhood experiences and mothers’ relationships with adult children

Abstract

Adverse childhood experiences (ACEs)—the constellation of experiences of abuse, neglect, and household dysfunction in childhood—affect the subsequent formation and maintenance of social relationships. Yet to date, there is limited knowledge on the potential consequences of maternal ACEs for the mother–adult child tie, one of the most important relationships across the life course. Insights from a life course perspective and research on the deleterious effects of ACEs suggest that mothers’ experiences of ACEs in childhood are so powerful that they may shape mother–adult child relationship quality decades later. To test this possibility, we draw on nationally representative, longitudinal cohort data from the National Longitudinal Survey of Youth 1979 (NLSY79) and the NLSY79 Child and Young Adult Supplement to examine associations between maternal ACEs and mother–adult child estrangement, contact, and closeness. Multi-level logistic and ordinal logistic regression models show that greater maternal ACEs exposure is associated with greater odds of estrangement from, less frequent contact with, and lower levels of closeness to the mother as reported by the adult child. Our findings highlight and extend research on intergenerational relationships by demonstrating that a mother’s ACEs exposure may be a fundamental determinant of her relationships with her adult children, documenting lengthier intergenerational consequences of ACEs than uncovered by prior research.

Introduction

Parent–adult child relationships are a key source of social support, providing emotional, instrumental, and financial benefits for both generations throughout the life course (Nomaguchi and Milkie 2020). Strong intergenerational relationships are central to social cohesion (Silverstein, Bengtson, and Lawton 1997), and their importance to parents and children has only grown in response to rising life expectancy and rates of marital instability (Swartz 2009). As a result, a wide range of scholarship considers the parent–child relationship to be one of the most enduring and important relationships in our lives, one characterized as “virtually indissoluble” (Pillemer et al. 2007, 786). This is perhaps especially the case for adult children’s relationships with their mothers. Mothers often take on the role of primary caregiver, and are expected to provide both instrumental and emotional support to children (Bianchi, Robinson, and Milkie 2006; Fingerman, Huo, and Birditt 2020). Mother–adult child relationships shape the well-being of both the mother and child, but not all ties are positive and life-enhancing (Nomaguchi and Milkie 2020; Pollmann-Schult 2014). Thus, understanding the characteristics associated with mother–adult child relationship dynamics is important for our holistic understanding of intergenerational health and well-being.

Although past research has documented numerous correlates of mother–adult child relationship dynamics, mothers’ experiences of early life adversity—also known as adverse childhood experiences (ACEs)—are underexplored but potentially foundational to mother–adult child relationship dynamics. Growing evidence indicates that exposure to ACEs can undermine healthy relationship functioning in adulthood (Wheeler, Barden, and Daire 2020; Umberson et al. 2014) via processes of emotional dysregulation, impairments in social competence and trust, and heightened hostility and vigilance to threat (Miller, Chen, and Parker 2011; Poole, Dobson, and Pusch 2018; Repetti, Taylor, and Seeman 2002). Although evidence on how single sources of early life adversity, like sexual abuse, affect subsequent social relationships is well established (Kearns-Bodkin and Leonard 2008; Walker, Holman, and Busby 2009), we know less about whether and how cumulative exposure to ACEs is associated with relationships with one’s children decades later. Integrating insights from sociology and family science on the determinants of intergenerational relationship dynamics with research on the consequences of ACEs for social relationships and well-being, we theorize that mother’s exposure to ACEs is a crucial predictor of key dimensions of the mother–adult child relationship.

In this study, we ask, how are maternal adverse childhood experiences (ACEs) associated with adult children’s estrangement from, contact with, and closeness to their mothers? To address this question, we use linked mother–child data from the National Longitudinal Survey of Youth 1979 to examine how six measures of maternal ACEs—physical abuse, emotional neglect, alcoholism within the home, mental illness in the home, parental separation, and parental death—cumulatively and independently are associated with relationship dynamics with adult children. We find that maternal ACEs are associated in a graded fashion with adult children’s reports of estrangement from mothers and that ACEs are associated with a lower likelihood of contact and emotionally close maternal–adult child relationships. We also find that two ACEs—emotional neglect and physical abuse—are independently and in combination associated with mother–adult child relationship dynamics above and beyond the influence of other ACEs. Our results indicate that mother–adult child relationships are strongly associated with the mother’s own experiences of childhood adversity, encouraging family scholarship to take a longer view in theorizing and empirically documenting characteristics associated with the parent–adult child relationship. The results also imply that the intergenerational reach of maternal ACEs is lengthier than prior research has uncovered, extending into the child’s adulthood. Taken together, our findings suggest that mitigating exposure to ACEs could be beneficial for relationships between mothers and their adult children and could result in more intact and emotionally close relationships.

Background

In this paper, we draw on life course theory to examine how events in the mother’s childhood are associated with key dimensions of the mother–adult child relationship. Two main tenets of life course theory are particularly relevant: timing and linked lives. First, the notion of timing in life course theory posits that time is both a sociohistorical and personal phenomenon important to understanding how “disruptive events” shape numerous outcomes across the life course (Elder, Johnson, and Crosnoe 2003). Specifically for this study, life course research suggests that events occurring during the sensitive period of childhood may have especially profound and enduring consequences throughout life (i.e., “the long arm of childhood”) (Hayward and Gorman 2004).

Second, a life course perspective highlights the interconnected nature of lives (i.e., “linked lives”; Settersten 2015), wherein events that occur in one person’s life indelibly influence close others. The “linked” nature of close social ties is especially evident in social relationships characterized by strong and enduring interdependence from childhood onward, such as in the case of mother and child (Kalmijn et al. 2019; Nomaguchi and Milkie 2020). For the mother–adult child relationship, this “linked” nature theoretically extends to the mother’s own childhood, with some evidence suggesting that a mother’s exposure to early life adversity might play a role in shaping her relationships with her own children (Kolomeyer et al. 2016; Miller, Chen, and Parker 2011).

Taken together, life course theory’s concepts of timing and linked lives suggest that understanding how mother–child relationships unfold over the life course requires attention to the mother’s early life factors. Although research on the mother–child relationship has considered the importance of the child’s early environment (e.g., research on the consequences of parental divorce; Amato 2000), we take a longer view of life course timing by considering how the mother’s early life environment patterns her relationships with her adult children across three key dimensions relevant to relationship quality—estrangement, contact, and closeness. We first define and describe the prevalence and known predictors of these aspects of the mother–adult child relationship. We then develop a conceptual model to theorize how a mother’s experiences with early life adversity might be associated with relationship dynamics between herself and her adult children in ways previously unexplored.

Parent–child relationships

Estrangement from, contact with, and closeness to parents

We focus on how mother–adult child relationships vary in three key dimensions: estrangement, contact, and closeness. Estrangement—which is understood as either no or low contact, or moderate contact but low relationship quality (Blake 2017; Gilligan, Suitor, and Pillemer 2015)—characterizes a small but important portion of parent–adult child relationships. Recent prevalence estimates for parent–child estrangement range from 6 to 26 percent (Reczek, Stacey, and Thomeer 2023), with mother–child estrangement being more rare than father–child estrangement (Blake 2017). Estrangement is noteworthy in the context of otherwise high levels of contact and closeness in the majority of parent–adult child relationships. Contact, which refers to both face-to-face communication and communicating via other means (e.g., texting), also provides an important gauge on parent–child relationship dynamics. Most young adults report seeing and talking to a mother regularly and frequently (Hartnett, Fingerman, and Birditt 2018), a finding corroborated by mother reports (Fingerman et al. 2016). Almost 70 percent of adult children report speaking to their mother once a week—and nearly 20 percent of adults at least daily (Lawton, Silverstein, and Bengtson 1994; Swartz 2009). Closeness, or feelings of emotional connection (Swartz 2009), likewise remains high for most mothers and adult children throughout life. Nearly 80 percent of adults consider their relationship with at least one parent to be “emotionally close” (Lawton, Silverstein, and Bengtson 1994; Swartz 2009). Most parents and adult children also report high levels of support, affection, and value similarity (Fingerman, Huo, and Birditt 2020; Van Gaalen and Dykstra 2006), often exchanging emotional, financial, and practical support into the child’s midlife years and beyond (Conti 2015; Silverstein, Gans, and Yang 2006).

In sum, most mother–adult child relationships remain intact with high contact and high relationship quality as indicated by closeness, although some are estranged, low in contact, and low in emotional closeness (Van Gaalen and Dykstra 2006). Understanding factors associated with estrangement, contact, and closeness separately is vital to gaining a full picture of parent–adult child relationships because intergenerational relationships vary across these three types of dynamics. For example, parent–child relationships may be characterized by frequent contact but low levels of closeness, or vice versa, and exposure to ACEs may independently predict contact and closeness. This multifaceted approach to examining mother–adult child relationships by considering estrangement, contact, and closeness is a key strength of our study, allowing us to better illuminate the complexity of intergenerational relationships.

Known predictors of estrangement from, contact with, and closeness to parents

Extant research has established that a wide range of relational factors throughout children’s lives strongly shape key dimensions of the parent–adult child relationship. Predictors of less frequent contact and less emotionally close relationships between parents and adult children include parental abuse and mistreatment, parental incarceration, an adult child’s romantic partner choice, both parents and adult children’s substance abuse, communication style differences, and parental divorce in childhood (Carr et al. 2015; Conti 2015; Padgett et al. 2012; Scharp and McLaren 2018; Scharp, Thomas, and Paxman 2015; Schoppe-Sullivan et al. 2023). While this work clearly shows that the child’s entire life course is relevant to understanding their relationship with their mothers in adulthood, associations between a mother’s experiences in childhood and adolescence and the maternal–adult child relationship remains largely unexamined. Strong evidence demonstrates that exposure to early life adversity shapes key dimensions of intimate relationships, like marriage (Miller, Chen, and Parker 2011; Umberson et al. 2014). We build on this research to consider whether a mother’s experiences of early childhood adversity are associated with her relationships with her adult children.

ACEs and intergenerational relationships: developing a conceptual model

Substantial evidence establishes the strong and enduring negative consequences of early childhood adversity for individual health and well-being across the life course (McEwen and McEwen 2017). Central to this research is the concept of “adverse childhood experiences” (ACEs), which captures multiple sources of potentially harmful experiences that occur in childhood (before age 18). The foundational ACEs study (Felitti et al. 1998), undertaken between 1995 and 1997, conceptualized ACEs as including abuse, neglect, and household dysfunction, such as domestic violence, substance abuse, and mental illness.¹ Though research had long established the deleterious effects of child abuse or trauma on adult well-being, the ACEs study demonstrated an additive effect in which exposure to multiple sources of childhood adversity showed a strong positive association with multiple health risk factors and harmful health outcomes in adulthood in a graded, stepwise fashion (Felitti et al. 1998). The study also demonstrated high rates of co-occurrence of ACEs: a non-trivial portion of those exposed to any one source of adversity also reported experiencing two or more additional ACEs. Thus, a focus on single sources of adversity obscures and likely underestimates the cumulative burden of ACEs throughout life. Thousands of studies (see Hughes et al. 2017 for an overview), mostly in medicine and public health, have since established ACE exposure as a powerful social determinant of later-life morbidity and mortality.

A much more recent body of research in the social sciences has begun to document the consequences of ACE exposure for a range of developmental processes and social outcomes, including the nature and quality of social relationships (Donnelly, Umberson, and Kroeger 2018; Ebbert et al. 2019; Umberson et al. 2014). As such, there are strong reasons for expecting childhood adversity to undermine the quality of interpersonal relationships in adulthood. Integrating research and theory from life-course epidemiology (Lynch and Smith 2005), socioemotional development (Pollak 2008; Repetti, Taylor, and Seeman 2002), and stress physiology (McEwen 1998), Miller, Chen, and Parker (2011) proposed that childhood stress can inhibit the ability to regulate emotions, undermine social competence, and increase mistrust of others, with negative consequences for relationship functioning.

Supporting this empirically, several studies indicate that children exposed to high levels of stress show signs of emotional dysregulation and impairments in social competence (Poole, Dobson, and Pusch 2018; Repetti, Taylor, and Seeman 2002), greater perceptions of threat (Miller, Chen, and Parker 2011), anger (Pollak and Sinha 2002), and hostility (Barefoot et al. 1991) in their social relationships, and lower levels of optimism (Lynch, Kaplan, and Shema 1997). Given that such factors are central to healthy relationship functioning, scholars have posited that childhood adversity may increase the risk of unstable and strained intimate ties. A growing body of empirical work supports this hypothesis. In one study, an index capturing exposure to eight ACEs predicted greater strain in intimate relationships in adulthood among Black men and women (Umberson et al. 2014). Moreover, Black men who experienced high levels of childhood adversity reported greater vigilance to threat in their intimate relationships in adulthood and described withdrawing emotionally and physically from relationship conflict (Umberson et al. 2016).

We develop a conceptual model to articulate why maternal ACEs might undermine the mother–adult child relationship (see fig. 1). We note that we do not formally test the mechanisms of this model but provide it as a theoretical substantiation of our overall theorizing on why maternal ACEs matters for the mother–adult child relationship. Theoretically, many of the same processes through which ACEs may undermine healthy functioning in intimate relationships, such as emotional dysregulation (Poole, Dobson, and Pusch 2018; Repetti, Taylor, and Seeman 2002), magnified perceptions of threat (Miller, Chen, and Parker 2011), and increased anger (Pollak and Sinha 2002) and hostility (Barefoot et al. 1991), may also predict strained intergenerational relationships with adult children. Empirically, the most established consequences of ACEs include multiple deleterious physical and mental health outcomes, including fair or poor self-rated health, psychological distress, and poor health behaviors (Hughes et al. 2017), which themselves predict parent–child estrangement (Reczek, Thomeer, and Bijou 2025). At the same time, growing evidence suggests that ACEs affect childbearing histories and union formation strategies (Williams and Finch 2019), potentially altering dynamics of intergenerational relationships. ACEs have also been linked to later-life poverty, educational attainment, and unemployment (Metzler et al. 2017), which are associated with parent–child relationship dynamics (Fingerman, Huo, and Birditt 2020). Finally, ACEs have been shown to influence harsh and negative parenting behaviors, such as slapping and yelling at children (Kolomeyer et al. 2016), which might stunt the formation of close, emotional bonds and undermine mother–child attachment (Bacon and Richardson 2001). In sum, as shown in figure 1, ACEs affect multiple maternal characteristics that are consequential for the emotional foundation and maintenance of the mother–adult child relationship.

Figure 1.

Conceptual model outlining relationship between maternal ACEs and key dimensions of the mother–adult child relationship. Note: Bulleted lists are not meant to be comprehensive but instead are examples of constructs.

A second pathway linking maternal ACEs exposure with mother–adult child relationship dynamics is via the influence of maternal characteristics like mental health and parenting behaviors that have been strongly shaped by their own ACE exposure, which, in turn, affect the child’s own development and life course experiences. For example, nascent research has established correlations between parents’ and children’s ACEs exposure (Schickedanz et al. 2021). Relatedly, children born to parents with greater ACEs exposure also report more behavioral problems in childhood (Schickedanz et al. 2018), suggesting that parental ACEs are consequential for children’s behavior and temperament. Further, children whose parents report heightened ACEs exposure tend to experience worse health than their counterparts whose parents reported no such exposure (Lê-Scherban et al. 2018). All of these characteristics may have implications for the mother–adult child relationship.

We build on past work that underscores the long reach of childhood adversity for intergenerational relationships (Kolomeyer et al. 2016; Miller, Chen, and Parker 2011) by examining whether mothers’ cumulative ACEs exposure is linked to relationships with their adult children using nationally representative, longitudinal cohort data. We expect that a mother’s experiences of early childhood adversity will be associated with a greater likelihood of estrangement from, less frequent contact with, and lower levels of emotional closeness to adult children. Although the theoretical and empirical reasons for examining the consequences of cumulative ACEs exposure for mother–adult child estrangement, contact, and closeness are convincing, a growing body of research is focused on identifying dimensions and groupings of ACEs that may be particularly consequential for some outcomes. For example, recent ACEs research differentiating items assessing household dysfunction from those assessing maltreatment suggests more negative effects of the latter for mental health in adolescence and early adulthood (Negriff 2020; Sayyah et al. 2022). Given how strongly maternal mental health is linked to maternal–child relationship quality (Foreman 1998), we expect that mother’s experiences of emotional neglect and physical abuse will be independently associated with key dimensions of the mother–adult child relationship.

Method

Data

We draw on two linked datasets. First, we use the National Longitudinal Survey of Youth 1979 (Bureau of Labor Statistics 2018; NLSY79), a nationally representative longitudinal cohort study. At the first interview, in 1979, 12,686 men and women were interviewed at ages 14–22. Respondents were interviewed annually until 1994 and biennially thereafter. Second, we draw on data from the Child and Young Adult (NLSY79-CYA) supplement, introduced in 1986, which contains a sample of 11,545 children who were born to mothers in the original NLSY79 sample. The CYA supplement followed children biennially throughout childhood, adolescence, and early adulthood. After children in the sample turned 30, they were interviewed every other wave, or every four years. The NLSY79 contains data on adversity in childhood, family background, parenting, and health, while the CYA supplement includes information on parent–child relationship quality and health, among other topics. For these reasons, the linked mother–child data from the NLSY79 and CYA supplement are the best source of data for examining our research questions.

ACEs index

We constructed our ACEs scale from reports of exposure to the following six forms of childhood adversity prior to the age of 18: (1) mental illness in the home (0 = no; 1 = yes); (2) alcoholism in the home (0 = no; 1 = yes); (3) physical harm or abuse (0 = never or once; 1 = more than once); (4) the absence of parental affection (0 = a great deal or quite a lot; 1 = a little or none at all), what we refer to as emotional neglect; (5) parental absence (0 = no; 1 = yes); and (6) the death of a biological or adoptive parent (0 = no; 1 = yes). These six items collectively capture distinct but related dimensions of ACEs: household dysfunction, abuse, and neglect (Felitti et al. 1998). The ACE index is constructed by summing the ACEs, and, consistent with prior research, binning and top coding the ACEs count into the following categories: 0, 1, 2, 3, and 4 or more (Hughes et al. 2017).

The first four ACEs—mental illness in the home, alcoholism in the home, emotional neglect, and physical abuse—were assessed from reports of retrospective exposure, starting in 2012. The final two ACEs—parental separation and parental death—were constructed prospectively from detailed measures of childhood living arrangements and dates of parental death as reported from 1979 onward. Mothers categorized as experiencing parental separation either reported (1) having resided with a biological or adoptive parent at birth but without that parent for a period exceeding four months prior to age 18 or (2) never having lived with any biological or adoptive parent at any point during childhood. The four-month period captures those who experienced prolonged separation from a parent (e.g., children who resided in foster care, children whose parents were incarcerated) rather than those who experienced ephemeral interruptions to typical family arrangements. We do not consider mothers who left home at age 17 to live independently to have experienced parental separation. Retrospective reporting of childhood adversity is common and has been shown to produce valid and reliable estimates of ACEs (Allen, Leadbeater, and Aber 1994; Dube et al. 2004; Warren, Lee, and Osypuk 2022).

Mother–adult child relationship dynamics

We examine three dimensions—estrangement, contact, and closeness—of the mother–adult child relationship, each reported by the adult child. Our first dependent variable is estrangement. Following Gilligan, Suitor, and Pillemer’s (2015) approach, we constructed our dichotomous estrangement indicator from both contact and closeness measures. Respondents were categorized as being estranged from their mother if they indicated (1) talking to or seeing her “not at all” or “about once a year or less” regardless of their assessment of closeness; or (2) talking to or seeing her “about once a year or less” or “several times a year” and rating their relationship with her as “not very close” or “fairly close.” Given the extreme positive skew of closeness (most children indicated being “extremely close” or “quite close” to mothers), it is reasonable to define as estranged those who were only “fairly close” and who had infrequent contact (Gilligan, Suitor, and Pillemer 2015). Respondents were categorized as having an intact (i.e., not estranged) relationship when they reported other configurations of more frequently talking to or seeing their mother, greater levels of closeness, or residing with the mother. Because we draw on contact and closeness variables, which can vary across waves, respondents can be categorized as estranged in one wave and as not estranged in other waves.

We also examine the components of estrangement, contact and closeness, independently. We assess two different dimensions of contact: talking to and seeing mothers. For the former, respondents were asked: “How often in the past year have you had contact with your mother, either by card, letter, e-mail message, or phone call?” For the latter, respondents were asked: “About how often do you see your mother?” Eight response categories captured frequency of contact for both measures, ranging from “not at all” to “every day.” We coded as low contact those who reported speaking with or seeing their mothers “about once a year or less” or “not at all”; moderate contact captured those who reported speaking with or seeing their mothers “two or three times a month,” “about once a month,” or “several times a year”; high contact included those who reported speaking with or seeing their mothers “every day,” “several times a week,” or “once a week,” in addition to those who reported residing with their mothers.² Our third dependent variable is closeness, which was measured by asking respondents, “How close do you feel to your mother? Would you say…” Respondents could answer “extremely close,” “quite close,” “fairly close,” or “not very close.” Our data include only adult children’s reports of contact and closeness with mothers, not mothers’ perceptions.³

Covariates

In all regression models, we control for a range of maternal background characteristics that prior research suggests are associated with maternal exposure to ACEs and relationship quality with adult children (Fingerman, Huo, and Birditt 2020). We adjust for mothers’ parents’ nativity status (one or both parents non-US born; both parents US born), highest grade completed by mother’s parents (in years), maternal birth year, whether no one in the home worked for pay at maternal age 14, mother’s family’s poverty status in 1978 (not in poverty; in poverty), maternal number of siblings (top-coded at 11), maternal nativity status (non-US born; US born), maternal rural residence at age 14 (suburban or urban; rural), maternal race (White; Black; Hispanic; or Other), and maternal first birth prior to the age of 18 (no; yes). We also adjust for child sex (male; female) and child age (in years). All variables are time-invariant, except for child age.

Our covariates are measured prior to or at maternal age 18 because we conceptualize them as potential confounders of the relationship between maternal ACEs and mother–adult child relationship dynamics. Our goal is to maintain proper time ordering in estimating associations between maternal ACEs (experienced by the mother prior to age 18) and mother–adult child relationship quality (reported by the child after age 18). As shown in figure 1, many maternal characteristics measured in adulthood that may affect mother–adult child relationship dynamics are themselves likely to be influenced by ACEs (Kolomeyer et al. 2016; Metzler et al. 2017), like parenting behaviors. Because those maternal adulthood characteristics may partially mediate the relationship between maternal ACE exposure and contact, closeness, and estrangement with adult children, controlling for them risks underestimating the true association between maternal ACEs and mother–adult child relationship dynamics, so we do not include them in our models. The time order of maternal adolescent childbearing in relation to ACEs exposure is less clear. We err on the side of minimizing potential bias due to confounding, presenting results that control for maternal-adolescent childbearing.

Analysis

We first present descriptive statistics for our analytic sample. We then show results from a series of multi-level mixed logistic and ordinal logistic regression models to examine mother–adult child estrangement, contact, and closeness as a function of maternal ACE counts, adjusting for maternal and child sociodemographic characteristics. We then present results from a series of models to investigate the potentially independent roles of singular ACEs. For these models, we estimated a series of logistic and ordinal logistic regression models predicting each outcome as a function of each ACE in separate regression models. We then included all six ACEs in the seventh model to examine which ACEs most strongly predicted each of our outcomes, net of other ACEs. Our final set of regression models predicted outcomes as a function of combinations of ACEs. We estimated multi-level mixed models to account for repeated observations within adult children and for the fact that children born to the same mother are likely to share observable and unobservable characteristics. Our multi-level models include 39,686 child waves, nested in 8,123 children, nested in 3,540 mothers. For ease of interpretation, we present our regression estimates in odds ratios (ORs). However, ORs are relative and do not present information in the natural metric of the dependent variable (Mize 2019), so we also use the suite of margins commands in Stata 18 (Long and Freese 2014) to estimate, by maternal ACEs, the predicted probability of reporting (1) estrangement from mothers; (2) frequencies of talking to and seeing mothers; and (3) levels of emotional closeness.

We employed multiple imputation with chained equations (m = 20) to account for missing data on the following characteristics of maternal respondents: (a) ACE index (n = 1,124), (b) her mother’s and father’s nativity status (n = 239), (c) her mother’s or father’s highest grade completed (n = 326), (d) whether no one in her home at age 14 worked for pay (n = 101), (e) rural residence at age 14 (n = 30), and (f) number of siblings at age 14 (n = 16). Following recommendations for “multiple imputation, then deletion” (Von Hippel 2007), cases with missing values on the dependent variable were included in the imputation equation and then removed from the analysis following imputation. Results are robust to using listwise deletion (shown in table A2 of the online appendix).

Results

Descriptive results

Table 1 describes each ACE and shows the percentage of adult children with mothers who reported each ACE. The most common ACE experienced was parental separation, with 23.5 percent of adult children’s mothers reporting a four-month or greater period of separation from a biological or adoptive parent. Other common experiences of childhood adversity were alcoholism in the mother’s home (21.5 percent) and emotional neglect (21.4 percent). By contrast, experiences of physical abuse (13.9 percent), mental illness in the home (9.3 percent), and parental death (5.9 percent) were less common.

Table 1.

Adverse childhood experiences (ACEs), ACE category descriptions, and percentages of respondents with mothers with each ACE.

ACE type	Description of ACE survey item	Percentage
Emotional neglect	Whether the respondent’s mother felt that her parents displayed no or little parental love and affection	21.40
Physical abuse	Whether the respondent’s mother’s mother or father hit, beat, kicked, or physically harmed her once or more than once, excluding spanking	13.88
Alcoholism in home	Whether the respondent’s mother resided with anyone who was a problem drinker or alcoholic	21.46
Mental illness in home	Whether the respondent’s mother resided with anyone who was depressed, mentally ill, or suicidal	9.28
Parental separation	Whether the respondent’s mother who was living with a biological or adoptive parent at birth was separated for a period greater than four months or never lived with a biological or adoptive parent	23.54
Parental death	Whether the respondent’s mother reported that her mother or father had died	5.89

Source: Data come from the NSLY79 and NLSY79 CYA Supplement.

Table 2 presents descriptive statistics for our analytic sample. Starting with the outcomes, 2.5 percent of children indicated a period of estrangement from their mothers across the study period. Most children also were categorized as having “high contact” with mothers both in terms of talking to mothers (89.7 percent) and seeing mothers (66.2 percent) once a week or more frequently. Relatively smaller portions of children had moderate contact and low contact. Face-to-face communication was less common than communicating in other ways (e.g., cell phone, e-mail). Adult children’s reports of closeness with mothers were slightly more varied. A majority of children indicated that they were extremely close to their mothers (52.1 percent), but a substantial portion said they were only quite close to mothers (26.8 percent). Smaller portions of children reported being fairly close (14.7 percent) and the smallest group of respondents indicated that they were not very close to their mothers (6.4 percent). Turning to maternal ACEs, 43.9 percent of children had mothers who reported no exposure to early life adversity. About 31.2 percent of adult children had mothers who indicated exposure to one ACE, and the proportion of children in each maternal ACE category decreases monotonically as maternal ACE exposure increases.

Table 2.

Descriptive characteristics of sample.

	Percentage or mean	SD
Estrangement
Not estranged	97.50
Estranged	2.50
Contact: talking to mothers
Low contact	2.01
Moderate contact	8.25
High contact	89.74
Contact: seeing mothers
Low contact	6.90
Moderate contact	26.94
High contact	66.16
Closeness
Not very close	6.41
Fairly close	14.68
Quite close	26.83
Extremely close	52.08
ACE index
Zero ACEs	43.92
1 ACE	31.15
2 ACEs	14.26
3 ACEs	7.34
4 or more ACEs	3.32
Mother’s parents’ nativity status
One parent or both parents non-US born	13.40
Both or all parents US born	86.60
Mother’s parents’ highest education (years)	10.94	3.46
No one in mother’s household worked for pay at age 14	13.60
Mother’s family in poverty in 1978	30.68
Mother’s number of siblings	4.31	2.67
Mother born in USA	92.18
Mother’s rural residence at age 14	20.40
Mother’s race ethnicity
White	41.02
Black	34.91
Hispanic	22.16
Other	1.91
Mother’s birth year	1960.35	2.21
Mother had teen birth	9.22
Age of child	25.12	5.17
Child’s sex
Male	48.75
Female	51.25

Note: Estimates come from the NLSY79 and NLSY79CYA. Eight thousand one hundred twenty-three children nested in 3,540 mothers.

Cumulative ACEs exposure

We present multi-level regression results in table 3. In Panel A, we show the results from logistic regression models predicting estrangement as a function of maternal ACEs and maternal and child sociodemographic confounding variables. Results reveal a positive relationship between mothers’ exposure to early childhood adversity and adult children’s reports of estranged relationships for all categories of ACEs exposure. Children whose mothers reported four or more ACEs are 4.4 times more likely to experience estrangement from mothers relative to children whose mothers reported no exposure to childhood adversity (P < .01). However, even children whose mothers reported experiencing only one ACE have higher odds of estrangement relative to children whose mothers experienced no ACEs (OR = 2.3; P < .001), suggesting that the presence of even a single ACE is associated with an increased likelihood of mother–adult child relationship dissolution. Pairwise comparisons between ACEs exposure groups provide evidence for a graded association. Children whose mothers reported two and three ACEs have greater odds of estrangement from mothers than children whose mothers experienced one ACE.

Table 3.

Odds ratios from multi-level logistic and ordinal logistic regression models predicting estrangement from, contact with, and closeness to mothers.

	Logistic	Ordinal logistic
	Panel A	Panel B	Panel C	Panel D
Variables	Estrangement from mothers	Talking to mothers	Seeing mothers	Closeness to mothers
Maternal ACEs
0 ACEs	Ref.	Ref.	Ref.	Ref.
1 ACE	2.25***	0.56***	0.74**	0.76**
	(0.483)	(0.063)	(0.069)	(0.066)
2 ACEs	4.00***a	0.41***a	0.45***a	0.57***
	(1.080)	(0.057)	(0.051)	(0.062)
3 ACEs	7.11***a	0.22***ab	0.34***a	0.39***
	(2.223)	(0.037)	(0.052)	(0.056)
4 or more ACEs	4.35**	0.26***a	0.37***a	0.41***
	(2.003)	(0.065)	(0.084)	(0.087)
Mother’s parents both US born	0.63	1.24	1.44*	1.59**
	(0.213)	(0.231)	(0.222)	(0.230)
Mother’s parents’ highest education (years)	1.00	0.99	0.92***	0.98
	(0.032)	(0.016)	(0.013)	(0.012)
No one in mother’s home worked for pay at age 14	1.46	0.91	0.95	0.98
	(0.378)	(0.129)	(0.114)	(0.109)
Mother’s family poverty status in 1978	1.16	0.86	0.95	1.08
	(0.271)	(0.108)	(0.097)	(0.108)
Mother’s number of siblings	0.94	1.00	1.01	1.01
	(0.034)	(0.019)	(0.016)	(0.015)
Mother born in USA	3.02*	0.54**	0.64*	0.67*
	(1.340)	(0.126)	(0.123)	(0.118)
Mother’s rural residence at age 14	0.91	1.10	1.07	1.08
	(0.197)	(0.126)	(0.101)	(0.093)
Mother’s race ethnicity
Non-Hispanic White	Ref.	Ref.	Ref.	Ref.
Non-Hispanic Black	0.65*	1.31*	1.93***	1.87***
	(0.142)	(0.150)	(0.183)	(0.164)
Hispanic	1.02	1.01	1.22	1.37**
	(0.276)	(0.147)	(0.148)	(0.152)
Non-Hispanic Other	1.55	0.61	0.98	0.80
	(0.822)	(0.180)	(0.249)	(0.186)
Mother’s birth year	1.10*	0.97	0.96*	0.96**
	(0.043)	(0.020)	(0.016)	(0.015)
Whether mother had a teenage birth	2.10***	0.53***	0.99	0.71***
	(0.435)	(0.061)	(0.098)	(0.072)
Age of child (years)	1.10***	0.92***	0.86***	0.97***
	(0.009)	(0.004)	(0.003)	(0.002)
Child is female	0.66**	2.68***	1.45***	1.07
	(0.089)	(0.192)	(0.078)	(0.057)
Variance components
Variance in mothers	5.51	2.58	2.50	1.79
	(0.61)	(0.18)	(0.13)	(0.11)
Variance in children	3.09	2.13	2.34	3.15
	(0.41)	(0.15)	(0.10)	(0.11)
Cutpoints for ordinal logistic regression
Cutpoint 1		−73.57	−84.97	−93.98
		(40.24)	(33.32)	(30.39)
Cutpoint 2		−71.11	−81.79	−91.84
		(40.24)	(33.32)	(30.39)
Cutpoint 3				−89.73
				(30.39)
Observations (child waves)	39,686	39,686	39,686	39,686

Notes: Standard errors in parentheses. Thirty-nine thousand six hundred eighty-six child waves, nested in 8,123 children, nested in 3,540 mothers. ^*P < .05; ^**P < .01; ^***P < .001.

^a P < .05 compared to one ACE.

^b P < .05 compared to two ACEs.

^c P < .05 compared to three ACEs.

Estrangement from a parent is composed of dimensions of both contact (talking to and seeing) and closeness. To gain clarity on the individual measures that constitute estrangement, we estimated ordinal logistic regression models predicting measures of contact and closeness reported by the child in adulthood in panels B, C, and D of table 3, respectively. We find that mothers’ exposure to childhood adversity significantly predicts these dimensions of the intergenerational relationship. Children whose mothers experienced more ACEs generally report less frequent contact with and lower feelings of emotional closeness to mothers at each ACE’s category of exposure. For example, the odds of being in the next more-frequent-contact category for talking to mothers are 74.1 percent lower (P < .001; Panel B) and for seeing mothers are 63.0 percent lower (P < .001; Panel C) for children whose mothers experienced four or more ACEs compared to children whose mothers experienced none. Comparing categories of ACEs indicates a graded association of maternal ACEs on adult children’s reported contact with mothers. Children whose mothers experienced two, three, or four or more ACEs report less frequent contact with mothers than did children whose mothers experienced only one form of childhood adversity.

Results from Panel D in table 3 show that the consequences of maternal ACEs extend to children’s emotional closeness to mothers. Across all levels of ACE exposure, the story is clear: children whose mothers were exposed to one (OR = 0.76; P < .01), two (OR = 0.57; P < .001), three (OR = 0.39; P < .001), or four or more (OR = 0.41; P < .001) ACEs report lower levels of closeness to mothers compared to children whose mothers had no ACE exposure. Unlike in the case of estrangement and contact, we find no evidence for a graded association between ACEs and mother–adult child closeness. Our results, taken together, demonstrate the powerful—and consistent—consequences of adversity in the mother’s childhood for shaping dimensions of the intergenerational relationship between mothers and their adult children.

We contextualize our results by plotting the predicted probabilities of reporting estrangement from mothers (fig. 2) as well as contact with mothers (Panels A and B of fig. 3) and closeness to mothers (fig. 4). Figure 2 illustrates that although estrangement from mothers is rare, it is patterned by maternal ACEs. The probability of reporting estrangement is only about 1.6 percent for children of mothers with no ACEs but more than 5 percent for children of mothers with three ACEs. Turning to figure 3, we find similar patterns in frequency of talking to mothers (Panel A) and seeing mothers (Panel B). As maternal ACE exposure increases, the probability of reporting low or moderate contact increases whereas the probability of reporting high contact decreases. Figure 4 shows the results for closeness. There is more variability in reported closeness to mothers than in reported estrangement or contact. We find that as maternal ACEs increase, children are more likely to report their relationships as not very close, fairly close, and quite close, and less likely to report them as extremely close. The likelihood of reporting a relationship as extremely close is about 57.5 percent for children of mothers with no ACEs but just above 42 percent for children of mothers with four or more ACEs.

Figure 2.

Predicted probabilities of estrangement from mothers as a function of maternal ACEs. Note: Models adjust for mothers’ parents’ nativity status, highest grade completed of mother’s parents, maternal birth year, whether no one in the home worked for pay at maternal age 14, mother’s family’s poverty status at age 14, maternal number of siblings, maternal nativity status, maternal rural residence at age 14, maternal race, maternal first birth prior to age of 18, child age, and child sex.

Figure 3.

Predicted probabilities of contact with mothers as a function of maternal ACEs. Note: Models adjust for mothers’ parents’ nativity status, highest grade completed of mother’s parents, maternal birth year, whether no one in the home worked for pay at maternal age 14, mother’s family’s poverty status at age 14, maternal number of siblings, maternal nativity status, maternal rural residence at age 14, maternal race, maternal first birth prior to age of 18, child age, and child sex.

Figure 4.

Predicted probabilities of closeness to mothers as a function of maternal ACEs. Note: Models adjust for mothers’ parents’ nativity status, highest grade completed of mother’s parents, maternal birth year, whether no one in the home worked for pay at maternal age 14, mother’s family’s poverty status at age 14, maternal number of siblings, maternal nativity status, maternal rural residence at age 14, maternal race, maternal first birth prior to age of 18, child age, and child sex.

Single sources and combinations of ACEs

We next considered whether individual ACEs were associated with estrangement from, contact with, and closeness to mothers. We present ORs and standard errors in table 4 and full regression results in tables A3–A6 of the online appendix. Results reveal that each of the same five ACEs—physical abuse, emotional neglect, alcoholism in the home, mental illness in the home, and parental separation—are individually associated with estrangement from, infrequent contact with, and diminished closeness to mothers. Put another way, children whose mothers reported experiencing any one of those five ACEs are more likely to become estranged from mothers and less likely to have high contact with and high levels of closeness to mothers compared to children whose mothers did not report that ACE exposure. A mother’s experience of parental death, however, was not independently associated with any of the mother–adult child relationship dynamics.⁴ When we enter all six ACEs into one model, we find that emotional neglect and physical abuse are independently associated with a greater likelihood of adult children’s estrangement from, reduced contact with, and lesser closeness to mothers, net of exposure to other ACEs. Put differently, mothers’ experiences of emotional neglect and physical abuse are associated with key dimensions of the mother–adult child relationship above and beyond the role of any single ACE captured in our scale. Additionally, when controlling for other ACEs, a mother’s experience of parental separation is individually associated with adult children’s reports of estrangement from and reduced contact (talking to and seeing) with mothers, and maternal experiences of mental illness in the home is independently associated with adult children’s reports of diminished closeness to mothers.

Table 4.

Odds ratios from multilevel logistic and ordinal logistic regression models predicting estrangement from, contact with, and closeness to mothers as a function of individual adverse childhood experiences.

Variables	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7
Panel A: Estrangement
Maternal ACEs
Emotional neglect	2.99***						2.09**
	(0.62)						(0.46)
Physical abuse		3.34***					2.04**
		(0.77)					(0.51)
Alcoholism in home			1.92**				1.22
			(0.42)				(0.27)
Mental illness in home				2.33**			1.52
				(0.63)			(0.42)
Parental separation					1.98**		1.87**
					(0.39)		(0.37)
Parental death						1.18	1.29
						(0.42)	(0.46)
Panel B: Talking to mothers
Maternal ACEs
Emotional neglect	0.39***						0.51***
	(0.05)						(0.06)
Physical abuse		0.36***					0.57***
		(0.05)					(0.08)
Alcoholism in home			0.56***				0.81
			(0.06)				(0.10)
Mental illness in home				0.51***			0.75
				(0.08)			(0.12)
Parental separation					0.61***		0.66***
					(0.07)		(0.07)
Parental death						0.98	0.95
						(0.19)	(0.19)
Panel C: Seeing mothers
Maternal ACEs
Emotional neglect	0.48***						0.60***
	(0.05)						(0.06)
Physical abuse		0.46***					0.66**
		(0.05)					(0.08)
Alcoholism in home			0.63***				0.83
			(0.06)				(0.08)
Mental illness in home				0.59***			0.79
				(0.08)			(0.11)
Parental separation					0.72***		0.75**
					(0.06)		(0.13)
Parental death						0.78	0.78
						(0.13)	(0.13)
Panel D: Closeness
Maternal ACEs
Emotional neglect	0.47***						0.54***
	(0.04)						(0.05)
Physical abuse		0.54***					0.78*
		(0.06)					(0.09)
Alcoholism in home			0.73**				0.93
			(0.07)				(0.09)
Mental illness in home				0.59***			0.74*
				(0.07)			(0.09)
Parental separation					0.83*		0.86
					(0.07)		(0.07)
Parental death						0.87	0.91
						(0.13)	(0.13)
Control variables	Y	Y	Y	Y	Y	Y	Y

Notes: Standard errors in parentheses. Thirty-nine thousand six hundred eighty-six child waves, nested in 8,123 children, nested in 3,540 mothers. ^*P < .05; ^**P < .01; ^***P < .001.

Given growing evidence that combinations of ACEs may be especially consequential (Negriff 2020; Sayyah et al. 2022), we conducted additional analyses presented in table 5 that consider how the following categories of maternal ACE exposure are associated with mother–adult child relationship dynamics: (1) no combinations of ACEs; (2) alcoholism and mental illness; (3) physical abuse and emotional neglect; and (4) alcoholism, mental illness, physical abuse, and emotional neglect (both combinations). We caution, however, that because these analyses are likely underpowered, they should be viewed as preliminary until confirmed by future research with a larger sample.⁵ Results in table 5 show that certain combinations of ACE exposure for mothers are differentially associated with children’s reports of estrangement from, contact with, and closeness to mothers. For example, the combination of physical abuse and emotional neglect more strongly predicts adult children’s reports of maternal estrangement from, lower contact with, and lower closeness to mothers than does the combined experience of alcoholism and mental illness in the childhood home. Though the combination of alcoholism and mental illness in the home is not significantly associated with estrangement and contact, it is negatively associated with maternal closeness to a similar degree as the combination of physical abuse and emotional neglect. The category of both combinations (alcoholism and mental illness and emotional neglect and physical abuse) is independently associated with less frequent contact with and diminished closeness to mothers, and its association with seeing mothers was significantly stronger than was the combination of alcoholism and mental illness in the childhood home. Small cell sizes preclude us from estimating associations between all other possible ACE combinations and our outcomes. In sum, we find evidence that certain combinations of ACEs that assess maltreatment—in particular, emotional neglect and physical abuse—appear to be particularly important for mother–adult child relationship dynamics.

Table 5.

Odds ratios from multilevel logistic and ordinal logistic regression models predicting estrangement from, contact with, and closeness to mothers as a function of combinations of adverse childhood experiences.

	Logistic	Ordinal logistic
	Panel A	Panel B	Panel C	Panel D
Variables	Estrangement from mothers	Talking to mothers	Seeing mothers	Closeness to mothers
Combinations of ACEs
No combination of ACEs	Ref.	Ref.	Ref.	Ref.
Alcoholism and mental illness	1.06	0.67	0.81	0.67*
	(0.527)	(0.169)	(0.176)	(0.135)
Physical abuse and emotional neglect	3.67***a	0.33***a	0.46***a	0.43***
	(1.147)	(0.060)	(0.072)	(0.067)
Alcoholism, mental illness, physical abuse, and emotional neglect	2.96	0.32***	0.36**a	0.44**
	(1.874)	(0.105)	(0.119)	(0.122)
Mother’s parents both US born	0.65	1.22	1.39*	1.56**
	(0.222)	(0.226)	(0.216)	(0.227)
Mother’s parents’ highest education (years)	1.00	0.99	0.92***	0.98
	(0.032)	(0.016)	(0.012)	(0.012)
No one in mother’s home worked for pay at age 14	1.69*	0.81	0.88	0.91
	(0.434)	(0.116)	(0.107)	(0.100)
Mother’s family poverty status in 1978	1.17	0.86	0.95	1.07
	(0.268)	(0.102)	(0.101)	(0.102)
Mother’s number of siblings	0.93	1.00	1.01	1.02
	(0.034)	(0.019)	(0.016)	(0.015)
Mother born in USA	2.82*	0.55**	0.66*	0.68*
	(1.256)	(0.128)	(0.127)	(0.120)
Mother’s rural residence at age 14	0.90	1.12	1.08	1.09
	(0.196)	(0.129)	(0.103)	(0.094)
Mother’s race ethnicity
Non-Hispanic White	Ref.	Ref.	Ref.	Ref.
Non-Hispanic Black	0.61*	1.35**	1.98***	1.89***
	(0.135)	(0.155)	(0.190)	(0.165)
Hispanic	0.99	1.05	1.23	1.39**
	(0.270)	(0.155)	(0.151)	(0.156)
Non-Hispanic Other	1.60	0.61	0.97	0.80
	(0.851)	(0.182)	(0.248)	(0.187)
Mother’s birth year	1.10*	0.97	0.96*	0.95**
	(0.043)	(0.020)	(0.016)	(0.015)
Whether mother had a teenage birth	2.15***	0.52***	0.98	0.70***
	(0.444)	(0.060)	(0.097)	(0.071)
Age of child (years)	1.10***	0.92***	0.86***	0.97***
	(0.009)	(0.004)	(0.003)	(0.002)
Child is female	0.65**	2.70***	1.46***	1.07
	(0.088)	(0.193)	(0.079)	(0.057)
Variance components
Variance in mothers	5.74	2.69	2.57	1.81
	(0.64)	(0.19)	(0.13)	(0.11)
Variance in children	3.11	2.13	2.35	3.15
	(0.41)	(0.15)	0.10)	(0.11)
Cutpoints for ordinal logistic regression
Cutpoint 1		−75.52	−86.26	−95.25
		(40.58)	(33.58)	(30.46)
Cutpoint 2		−73.06	−83.08	−93.11
		(40.58)	(33.58)	(30.46)
Cutpoint 3				−91.00
				(30.46)
Observations (child waves)	39,686	39,686	39,686	39,686

Notes: Standard errors in parentheses. Thirty-nine thousand six hundred eighty-six child waves, nested in 8,123 children, nested in 3,540 mothers. †P < .10; ^*P < .05; ^**P < .01; ^***P < .001.

^a P < .05 compared to alcoholism and mental illness.

^b P < .05 compared to physical abuse and emotional neglect.

Discussion

Mother–child relationships are among the most enduring and consequential social relationships (Nomaguchi and Milkie 2020; Swartz 2009). Given the health consequences of poor quality ties, it is important to understand what factors predict particularly problematic relationship dynamics. Taking a life course perspective and using nationally representative, longitudinal cohort data, our study examines how maternal ACEs are associated with the long-term dynamics of one of the most important relationships in the life course: the relationship between mothers and their adult children.

Drawing on our conceptual model shown in figure 1, our study results produce two key findings. First, we found that maternal ACEs were predictive of adult children’s reports of a greater likelihood of estrangement from, infrequent contact with, and lower feelings of emotional closeness to mothers. This is the case even after adjusting for a robust battery of maternal and child sociodemographic and socioeconomic characteristics that influence both maternal ACE exposure and mothers’ relationships with adult children. Our coefficients are large in magnitude, certainly much larger than other predictors of mother–adult child relationship dynamics included in our models, suggesting that ACEs are of key importance in intergenerational relationship dynamics. Notably, the probability of estrangement by maternal ACEs exposure demonstrates that the consequences of childhood adversity for the impermanence of the mother–child relationship are significantly and substantively meaningful. The predicted probabilities of contact and closeness show just how powerfully maternal ACE exposure stratifies each dimension of the mother–adult child relationship. A key strength of our study is that we leverage reports of ACEs from mothers and reports of relationships from children, minimizing concerns of reporting bias from mothers.

Second, our findings indicate that mothers’ ACEs generally exhibited a “dose–response” association in patterning relationship dynamics with adult children, corroborating and adding to a growing body of research on the graded effects of ACEs (Felitti et al. 1998). Prior studies have noted that the likelihood of negative outcomes increases with each additional ACE (Felitti et al. 1998; Felitti 2002). Although each maternal ACE except parental death was independently associated with children’s perceptions of estrangement from, infrequent contact with, and lesser feelings of closeness to mothers, children whose mothers reported two or more ACEs generally fared worse in terms of relationship dynamics than children whose mothers reported only one ACE. Therefore, maternal ACEs are associated with estrangement, contact, and closeness with adult children in a graded way.

In addition to observing graded associations between maternal ACEs exposure and intergenerational relationship dynamics, we also found evidence that maternal exposure to childhood abuse and neglect was especially consequential. A limitation of the additive ACEs scale is that it assumes identical effect sizes of each of the individual ACEs included in the scale (see Negriff 2020). Yet a small but growing body of research indicates that childhood abuse and maltreatment ACE items are more consequential than those that assess household dysfunction for a range of outcomes in adolescence and early adulthood, including mental health (Chapman et al. 2004; Negriff 2020) and maternal prenatal depressive symptoms (Atzl et al. 2019). Our results confirm and extend such evidence for intergenerational relationships between mothers and their adult children. Net of the other ACEs in our model, physical abuse and emotional neglect were independently associated with a greater likelihood of estrangement from, reduced contact with, and diminished closeness to mothers. Though explanations for this pattern are beyond the scope of the present study, maternal mental health—strongly shaped by childhood maltreatment and also predictive of lower-quality relationships—may be a key mediator. It is also possible that the combination of physical abuse and emotional neglect may have even stronger synergistic effects on the outcomes we examine (i.e., that the combination of both ACEs exceed the sum of the individual associations of each ACE), a possibility that future research should explore (see Briggs et al. 2021).

Our findings make central contributions to two literatures that are often studied separately: (1) characteristics associated with intergenerational relationships across the life course and (2) how ACEs influence the formation and maintenance of social relationships. We address each in turn. Our conceptual model (fig. 1) suggests that a mother’s early life course experiences are consequential for mother–adult child relationship dynamics, and we find support for our guiding framework. Fundamental dimensions of the mother–adult child relationship—estrangement, contact, and closeness—are possibly shaped by experiences in the mother’s early life course, typically before the child is even born. Contact frequency and relationship quality in intergenerational relationships are influenced by events that happen when children are young, such as parental divorce, and by events that happen when children are adults, such as substance abuse, discordant political views, and disagreements in life choices (Amato 2000; Carr et al. 2015; Padgett et al. 2012; Scharp and McLaren 2018; Scharp, Thomas, and Paxman 2015). We drew on a life course perspective and empirical research demonstrating the profound effects of ACEs on the formation and maintenance of social relationships to document an underexplored characteristic associated with mother–adult child relationship dynamics: mothers’ experiences of adverse childhood experiences. Findings from our study should encourage future scholars studying the determinants of the parent–child relationship to take a longer view that recognizes the importance of parents’ own developmental history, including exposure to early life adversity.

Our study further establishes that the intergenerational consequences of ACEs extend later into the life course than previously observed in prior research. Past studies documenting the long arm of childhood adversity show that parental ACEs are generally associated with children’s physical, mental, and emotional health in childhood (Lê-Scherban et al. 2018; Schickedanz et al. 2018). To our knowledge, ours is the first study to document intergenerational consequences of parental ACEs that extend across generations into a child’s adulthood. The consequences of ACEs seemingly have ripple effects that extend a longer arm into the following generation than prior research has documented. The findings highlight that parents’ own childhood is a crucial stage in the life course, with adversity experienced during this time having potentially harmful downstream effects on relationships between parents and their adult children much later in life. The poor intergenerational relationship dynamics we observed for adult children with mothers who had the greatest ACEs exposure may continue into later life, with potential implications for caregiving availability across generations (Fingerman, Huo, and Birditt 2020).

Limitations and future research

A major contribution of our analysis was to provide an empirical test of associations between maternal ACEs and mother–adult child relationship dynamics, shown in fig. 1. Though empirically establishing the importance of specific mechanisms was beyond the scope of the present study, doing so should be a priority for future research. In fact, many of the previously identified more proximate predictors of maternal–child relationship quality (e.g., parental divorce, substance abuse; Amato 2000; Gilligan, Suitor, and Pillemer 2015; Kaufman and Uhlenberg 1998; Schoppe-Sullivan et al. 2023) may in fact be more fundamentally influenced by maternal ACEs. Now that we have identified the importance of maternal ACEs as an important predictor of mother–adult child relationship dynamics, future research should begin to determine the extent to which these factors mediate the relationship between maternal ACEs and mother–adult child estrangement, contact, and closeness.

The background of our conceptual model also highlights the importance of structural sources of adversity (e.g., racism, income inequality). In identifying maternal ACEs as an important characteristic associated with intergenerational relationship dynamics, it is important to recognize that ACEs themselves are distributed unevenly throughout the population in ways that reflect structural disadvantage (Slopen et al. 2016). Although the focus of our study is on how ACEs are associated with the mother–adult child relationship and not on who is most at risk for ACEs exposure, we recognize that structural sources of adversity may shape both exposure to abuse, neglect, and household dysfunction in the ACEs scale and aspects of the intergenerational relationship. Additional research should also consider the role of positive childhood experiences (PCEs), which include peer support at school, participation in organized activities, and having positive relationships with adult mentors (Bethell et al. 2019). PCEs are associated with better mental health in adulthood and several other positive outcomes, net of ACEs (Bethell et al. 2019). PCEs might mitigate or buffer against the deleterious consequences of ACEs, and future research should examine this possibility.

Some limitations are also worth mentioning. We were unable to examine how paternal ACEs are associated with father–adult child estrangement, contact, and closeness. The NSLY79 CYA interviewed children born only to women in the original sample, not those born to men, prohibiting us from considering relationships between paternal exposure to ACEs and father–adult child relationship dynamics. Future work should explore whether children whose fathers report greater ACEs exposure similarly report lower relationship quality with fathers. Limitations in the traditional ACEs scale have been the subject of growing attention, including the recall bias inherent in retrospective reports. The best evidence suggests that false negatives outnumber false positives in recalling and reporting early life adversity (Hardt and Rutter 2004), which would lead to more conservative estimates in our study. Similarly, lack of information in the NLSY79 on profoundly harmful ACEs, such as sexual abuse, may further contribute to more conservative estimates.

Conclusion

Our study is novel in that it demonstrates how events that happen before a child is even born—in this case, maternal ACEs—are associated with dimensions of the mother–adult child relationship decades later. Merging insights from family sociology and research on how ACEs undermine the formation and maintenance of social relationships, we theorized that ACEs are a relatively overlooked but potentially foundational determinant of mother–adult child relationship dynamics. Children whose mothers reported greater exposure to early life adversity were more likely to be estranged from, experience less frequent contact with, and rate their relationships as less close with mothers in a stepwise, graded fashion. In particular, physical abuse and emotional neglect, individually and in combination, appear to be especially important for the mother–adult child relationship. Our study illuminates that important dynamics of the intergenerational relationship between mother and adult child are being shaped by events that happen to the mother early on in her life course and that the intergenerational consequences of ACEs on children extend further than prior research has documented, vis-à-vis their associations with adult children’s relationships with mothers. Minimizing children’s exposure to ACEs or mitigating the profoundly harmful consequences of such exposure would likely improve parent–child relationships in future generations. Childhood matters, not just for oneself but for one’s children as well.

About the authors

Lawrence Stacey is Assistant Professor of Sociology at Vanderbilt University, where he is also affiliated with the LGBTQ+ Policy Lab and the Center for Research on Inequality and Health. His research revolves around gender, sexuality, health, families, and early life adversity. Recent work of his appears in Demography, Social Science & Medicine, Journal of Marriage and Family, The Journal of Sex Research, and Sexualities, among other outlets.

Rin Reczek is Professor of Sociology and an affiliate at the Institute for Population Research at The Ohio State University. In addition to over 80 articles and book chapters in the areas of family, health, gender, and sexuality, Rin has published the 2022 book Families We Keep: LGBTQ Adults and their Enduring Bonds with Parents (NYU Press) and the forthcoming book Families We Lose: Family Estrangement and the Democratization of Kinship (NYU Press).

Kristi Williams is Professor of Sociology and Associate Dean for Faculty Affairs at The Ohio State University and a Senior Scholar at the Council on Contemporary Families. Her scholarship examines how the changing intersections of union formation and fertility shape health and well-being across the life course and across generations. Her most recent research explores the effects of adverse childhood experiences on life course developmental processes including marriage and parenthood, with consequences for midlife health.

Jake Hays is a Research Fellow at the Stone Center for Inequality Dynamics at the University of Michigan’s Survey Research Center. His research focuses on family demography, social inequality, health, and the life course. His recent publications include “Nonresident Parent Wealth among Children” (forthcoming at Demography) and “Absolute Wealth Mobility in the United States During the COVID-19 Pandemic” (Population Research and Policy Review).

Mieke Beth Thomeer is Professor of Sociology at the University of Alabama at Birmingham. Her research focuses on how families matter for health throughout the life course, with a specific focus on processes related to gender, race/ethnicity, and sexuality. Her research on childbearing biographies and mid- and later-life health is supported by the National Institute on Aging. She is Deputy Editor for the Journal of Marriage and Family.

Funding

This work was supported by the National Institute on Aging (NIA) (R01AG069251; Thomeer & Reczek, Principal investigators) and the Eunice Kennedy Shriver National Institute of Child Health and Human Development grant awarded to the Institute for Population Research at The Ohio State University (P2CHD058484) from the National Institutes of Health.

Conflicts of interest

None declared.

Data availability

The data underlying this article are provided by the Bureau of Labor Statistics and are publicly available via the NLS Investigator: https://www.nlsinfo.org/investigator/pages/login.