Attachment as a Moderator in the Relation Between Child Maltreatment and Symptoms of Depression

Abstract

Experiencing child maltreatment is a risk factor for later psychopathology; however, not all survivors of child maltreatment go on to develop mental health diagnoses. There are likely important risk factors that act as moderators interacting with child maltreatment to contribute to the development of psychopathology. The present study examined the attachment dimensions of anxiety and avoidance as potential moderators in the association between child maltreatment and depressive symptomatology in a sample of college students. An attachment style high in anxiety or avoidance was expected to exacerbate the effect of child maltreatment on depressive symptomatology across both primary and secondary caregiver attachment relationships. This study was conducted at a private university in the northeastern United States in a sample of college students (N = 203; 52% male; M_age = 19.85, SD_age = 2.19). Participants completed online measures of attachment, current mood symptoms, and demographic information. Two moderation models were tested, one for attachment to primary caregivers and one for attachment to secondary caregivers. Anxious attachment to primary caregivers moderated the relationship between child maltreatment and depressive symptoms (B = −0.16, p < .01, R² =.44). However, moderation was not significant in the secondary caregiver attachment relationship. Maladaptive attachment styles, as well as child maltreatment itself, may result in disruptions in the development of positive internal working models of the self and others and adaptive emotion regulation. In cases of child maltreatment, interventions focused on the parent–child attachment relationship may have long-lasting effects and implications for the child’s future mental health. This research highlights important areas of intervention in cases of child maltreatment as well as important differences in the anxiety and avoidance dimensions of attachment.

Keywords

child abuse attachment mental health and violence family issues and mediators

Child maltreatment (CM) is prevalent in the United States, where it is estimated that one in eight children will experience maltreatment by the time they are 18 years old (Wildeman et al., 2014). The United States Department of Health and Human Services (U.S. DHHS) defines CM as an action (e.g., physical abuse, sexual abuse, emotional abuse) or an omission (e.g., physical or emotional neglect), which results in death, harm, or puts a child at risk of harm (U.S. DHHS, 2020). An estimated total of 4.3 million allegations of CM were made in 2018 in the United States and of reports that were investigated, an estimated 678,000 victims were identified (U.S. DHHS, 2020). CM is a developmental risk factor that can lead to various forms of psychopathology throughout the lifespan (see Jaffee, 2017 for a review). In 2018, parents were most often identified as the perpetrator of CM (U.S. DHHS, 2020), thus, the relationship between child and parent is an essential area of intervention in cases of CM (Valentino, 2017). Children’s attachment relationship to their caregiver has a long-lasting impact as it shapes their relationships throughout their life (Jones et al., 2018). Whereas a secure attachment style has been identified as a protective factor against adverse outcomes of CM (Meng et al., 2018), attachment styles characterized by anxiety and avoidance have been associated with maladaptive outcomes such as depression (Conradi et al., 2018). Research on CM and attachment primarily examine these constructs from infancy through adolescence, whereas less is known about how adults’, specifically college students’, attachment relationships to their primary and secondary caregivers may continue to mitigate or exacerbate the effects of experiencing CM on depression.

College students are particularly vulnerable to mental health issues, due in part to the multitude of stressors associated with the change in lifestyle that college necessitates (Ibrahim et al., 2013). Ibrahim et al. (2013) conducted a systematic review of 24 studies (nine of which were conducted in the United States) and found that the weighted mean prevalence of depression in college students was 30.6%. A study based in the United States found that 33% of college students at a Midwestern university had a mental health problem and that depression had the second-highest prevalence (13%–15%), with eating disorders being the most prevalent (Zivin et al., 2009). A 10-year study examining trends in depression and suicidal ideation in a sample of 155,026 college students found that 26.9% endorsed depressive symptoms in the past 2 weeks (Lipson et al., 2019). Furthermore, it was found that the prevalence of screening positive for depressive symptoms had steadily increased over the years, from 24.8% in 2009 to 29.9% in 2017 (Lipson et al., 2019). College students are particularly vulnerable to depression; examining risk and protective factors for this population may greatly improve important outcomes for this population such as academic success and social relationships (Zivin et al., 2009).

CM and Depression

CM has been associated with greater odds of experiencing depression across multiple studies; two meta-analyses examining the relationship between CM and developing depression found that individuals who experienced CM were 2.03 to 3.73 times more likely to develop depression (Jaffee, 2017; Li et al., 2016; Nelson et al., 2017). Depressive disorders can occur at any age, even as early as childhood and adolescence, and early onset of a depressive disorder increases the risk of comorbid personality and substance use disorders (American Psychiatric Association, 2013). Across 10 studies, Nelson et al. (2017) found that individuals with a maltreatment history began experiencing depression about 4 years earlier than individuals who did not experience CM. Furthermore, a maltreatment history has been associated with recurrent and persistent depressive episodes as well as poor response to treatment and remission (Nanni et al., 2012; Nelson et al., 2017). Depressive disorders permeate many aspects (e.g., educational, economic, social) of a person’s life and a history of CM makes treatment and recovery even more difficult (Nanni et al., 2012; Zivin et al., 2009). Various factors have also been associated with resilience against the harmful effects of CM, including family (i.e., familial support, connectedness, early family experiences), parent–child relationship (i.e., parenting practices and attachment), and individual (i.e., self-esteem, self-efficacy, adaptive coping, and feelings of control) factors (Afifi & MacMillan, 2011; Cicchetti & Banny, 2014; Collishaw et al., 2007; Meng et al., 2018). Thus, elucidating what factors contribute to, and protect against, the development of depression after CM is essential to early intervention efforts.

Whereas many factors have been associated with exacerbating or mitigating adverse outcomes after experiencing CM, it is clear that caregivers play a crucial role in either protecting against, or further increasing risk for adverse outcomes after CM (Valentino, 2017). Parents are overwhelmingly identified as the perpetrators of CM; in 2018, over 90% of victims were maltreated by their parent(s), and further, only 22.9% of victims were removed from their home (U.S. DHHS, 2020). This underscores the importance of the parent–child relationship in mitigating risk for future adverse outcomes (Valentino, 2017). In fact, research examining attachment as a target of intervention with maltreated children has found favorable outcomes (e.g., reduced disorganized attachment, increased secure attachment; Stronach et al., 2013; Valentino, 2017). In the present study, current attachment relationships to caregivers were examined as a moderator of the relation between CM and depressive symptoms in adulthood.

Attachment

Attachment theories suggest that attachment styles are developed through interactions between the infant and caregiver; these interactions are thought to determine whether the infant feels secure or insecure in novel situations (Ainsworth & Bell, 1970; Bowlby, 1988). Secure attachment is thought to be established in infancy by a reliable and responsive caregiver (Toth et al., 2013). Secure attachments are facilitated by available, helpful, caring, and loving caregiving that attends to the child’s physical and emotional needs (Bowlby, 1988).

Insecure attachment is characterized by a child’s uncertainty about whether their caregiver is dependable, consistent, and available to meet their needs and is thought to be a result of a caregiver who is sometimes or always unavailable for the child (Bowlby, 1988). Anxious-ambivalent insecure attachment has been characterized by uncertainty of whether a caregiver is dependable, whereas anxious–avoidant insecure attachment has been indicated by a belief that a caregiver will ignore them in times of need (Bowlby, 1988). Caregivers who maltreat their children may be inconsistent in their caregiving practices, at times they may be caring and at other times abusive, leaving children unsure whether their parent is safe to approach; children who experience this type of caregiving may demonstrate disorganized attachment (Bowlby, 1988). Children with this attachment style act in bizarre manners, such as freezing or engaging in stereotypies, when reunited with parents in paradigms such as the Strange Situation (Bowlby, 1988).

Disorganized attachment and other insecure attachment styles have been found in individuals who have experienced maltreatment across age-groups, from infancy to adulthood (Bifulco et al., 2006; Bowlby, 1988; Cicchetti & Banny, 2014; Oshri et al., 2015; Toth et al., 2013). Oshri et al. (2015) found that sexual and emotional abuse were positively correlated with anxious and avoidant attachment styles in adolescents. In a study of high-risk women, Bifulco et al. (2006) found that 92% of women reporting the most insecure attachment styles reported experiencing CM. Furthermore, the percentage of women who reported experiencing CM decreased as attachment styles approached secure attachment (Bifulco et al., 2006). Individuals who experience CM may be at greater risk for having insecure attachment styles, even as adults.

The earliest attachment relationships, developed between child and caregiver, are thought to influence later attachment to peers and romantic partners throughout adolescence and adulthood (Bowlby, 1988; Cicchetti & Banny, 2014); in fact, research has shown that attachment styles demonstrate stability from infancy through adolescence and into adulthood (Fraley et al., 2011; Hamilton, 2000; Waters et al., 2000). Models of adult attachment, based on internal working models of self and others, provide a framework for the application of early attachment styles to later adult attachment (Bartholomew & Horowitz, 1991; Gillath et al., 2016; Hazan & Shaver, 1987). In particular, anxiety in recent adult attachment models relates to feelings of unworthiness, as well as a fear of being abandoned or rejected, whereas avoidance refers to an aversion to closeness or dependency on someone else (Gillath et al., 2016). The present study utilizes this model to examine adult attachment to caregivers. Although research has demonstrated that children who experience maltreatment are likely to exhibit disorganized attachment, the reliable measurement of disorganized attachment in adults is still under development (see Beeney et al., 2017).

Attachment and Depression

Attachment relationships with caregivers may be related to the risk for depression in a multitude of ways, including shaping attachment styles to others, cognitions about self and others (e.g., internal working models), as well as emotion regulation (Bowlby, 1988; Spruit et al., 2020). Insecure attachment relationships can lead children to develop negative internal working models of themselves (e.g., “I deserved what happened to me”) and others (e.g., “I can’t trust others”; Bowlby, 1988). These negative internal working models can result in beliefs that they are unworthy of love and affection and that others are untrustworthy and uncaring (Bowlby, 1988). In this way, early attachment relationships define how children interact with the external world (Bowlby, 1988), and they can make it difficult for an individual to develop secure attachments to others throughout life, which may prevent them from developing important social relationships (i.e., romantic relationships, peer relationships). These negative internal working models can be viewed as general cognitive schemas that increase vulnerability to depression (Spruit et al., 2020). In addition, in times of stress, individuals with secure attachments may be soothed by the fact that they have a secure attachment and positive internal working models (i.e., the knowledge that others will be responsive when a problem arises; Malik et al., 2015). These thoughts can help individuals manage their emotions effectively.

Attachment is multifaceted and can be indicative of multiple pathways to vulnerability for depression. Specifically, insecure attachment relationships with caregivers may confer risk for depression through maladaptive cognitive schemas about the self and others, insecure attachment in other relationships, and poor emotion regulation (Bowlby, 1988; Malik et al., 2015; Spruit et al., 2020). Anxious and avoidant dimensions of attachment have been associated with more severe depression and more time experiencing symptoms over a 7-year period (Conradi et al., 2018). In two meta-analyses, children, adolescents, and adults with insecure attachments demonstrated higher rates of depression (Dagan et al., 2018; Spruit et al., 2020). Many studies have examined the relationship between attachment styles to caregivers in infancy through adolescence; however, few studies have examined adults’ attachment styles to their caregivers from childhood (e.g., parents) and whether these attachment styles continue to affect mental health in adulthood.

Present Study

Individuals who experience CM are two to three times more likely to develop depression (Jaffee, 2017; Li et al., 2016; Nelson et al., 2017) and have been found to demonstrate insecure attachment styles, characterized by anxiety and avoidance (Bifulco et al., 2006; Conradi et al., 2018; Dagan et al., 2018; Oshri et al., 2015; Spruit et al., 2020; Toth et al., 2013). Furthermore, attachment styles characterized by anxiety and avoidance have been found to be related to depression and poorer outcomes of the course of depression (Conradi et al., 2018; Dagan et al., 2018; Spruit et al., 2020). Despite strong relations between attachment to caregivers, CM, and depressive symptoms, it is unknown whether CM and adult attachment relationships to caregivers interact to be associated with depressive symptoms in adults. Specifically, it is unknown whether the impact of CM on depressive symptoms may be conditional upon levels of anxious and avoidant attachment to caregivers. High levels of anxious and avoidant attachment styles were expected to exacerbate the impact of CM on depressive symptoms, whereas low levels of attachment anxiety and avoidance were expected to mitigate the effects of CM on depressive symptoms. There is a lack of research examining the continuing influence of attachment with caregivers into adulthood and past research has often focused on one caregiver. Specifically, maltreatment literature highlights the mother–child relationship as an important target of intervention (Valentino, 2017); however, research has suggested that there are differences in the way that mothers and fathers interact with their children and, therefore, they may differentially affect aspects of children’s development (Bretherton, 2010). To obtain a comprehensive view of how attachment to caregivers may influence an individual’s trajectory, both primary and secondary caregiver attachment relationships were examined in the present study. No differential hypotheses were made with regard to primary versus secondary caregiver. Results may outline under what conditions targeting attachment relationships may exacerbate or mitigate the effect of CM on depressive symptoms.

Method

Participants

Participants (N = 203) self-selected to participate in the online study. Participants were students at a college in the northeastern United States. Although 215 participants self-selected to participate in the study, 12 participants were missing data on all variables. The age of participants ranged from 18 to 36 (M = 19.85, SD = 2.19), males comprised 52% of the sample (n = 102), and 63% of participants were White (see Table 1 for additional demographic information). Participants had to be at least 18 years old to complete the study; there were no other exclusionary or inclusionary criteria for participation.

Table 1.

Demographic Characteristics of the Sample and Reported Caregivers.

Race and ethnicity	Count	%
Black	22	11.17
Asian	28	14.21
White	124	62.94
Hispanic/Latino	8	4.06
American Indian/Alaskan Native	2	1.02
Multiracial	13	6.60
Sex
Male	102	51.78
Female	95	48.22
Hearing status
Hearing	184	93.4
Deaf	6	3.05
Hard of hearing	7	3.55
Number of mental health treatments
None	145	73.61
One treatment	36	18.27
Two treatments	16	8.12
Annual household income of family of origin
0–49k	51	25.89
50–99k	59	29.95
100–149k	46	23.35
150–199k	23	11.68
200k and above	18	9.13
Primary caregiver
Biological mother	167	84.77
Biological father	18	9.13
Grandmother	6	3.05
Other	6	3.05
Secondary caregiver
Biological mother	16	8.12
Biological father	142	72.08
Grandmother	14	7.11
Grandfather	6	3.05
Stepmother	3	1.52
Stepfather	5	2.54
Other	11	5.58

Note. Only n = 197 participants completed the demographic information out of the total sample (N = 203). Percentages are out of n = 197 participants.

Materials

Demographic survey

Participants completed a demographic survey including questions regarding their mental health diagnoses (excluding depression), mental health treatments (therapy, medication, and/or other), and biological relatives’ mental health. Participants were also asked about their hearing status, as there is a large population of Deaf and hard-of-hearing individuals at this particular institution. Participants were asked about their approximate average annual household income of their family as an indicator of socioeconomic status (SES). They were also asked to identify their primary and secondary caregivers in childhood. A primary caregiver was defined as

“the individual who took on the most responsibility for seeing that you were cared for (i.e., fed you, made sure you were ready for school, put you to bed, took you to the doctor).”

Secondary caregiver was defined as

the individual who may have been less involved in seeing that you were cared for. This may be the caregiver who worked more or was absent more often. Alternatively, this may be a person who cared for you when your primary caregiver was unavailable (i.e., a grandparent).

Depressive symptoms

The Beck Depression Inventory–II (BDI-II; Beck et al., 1996) was used to evaluate current depressive symptoms. Participants reported how they had been feeling for the past 2 weeks including the present day. The measure was comprised of 21 questions that assessed various symptoms of depression; each symptom had four statements regarding severity rated on a scale of 0 (e.g., “I am no more irritable than usual”) to 3 (e.g., “I am irritable all the time”), with higher scores indicating the presence of a depressive symptom. The total possible range of scores was from 0 to 63 with higher scores indicative of more severe depressive symptoms (Wang & Gorenstein, 2013). The BDI-II has been widely used and translated into many languages and has been reported to have good psychometric properties (Wang & Gorenstein, 2013). Cronbach’s alpha in the present study was high (α = .93; see Table 2).

Table 2.

Correlations, Means, Standard Deviations, Ranges, and Psychometric Properties of Variables.

Variable	1	2	3	4	5	6	7	8	9	10
BDI-II score	—
Sex	.19**	—
Hearing status	.29**	.09	—
Count of diagnoses	.37**	.21**	.09	—
Count of treatments	.38**	.26**	.16*	.70**	—
CTQ score	.51**	.12	.25**	.27**	.26**	—
Anxious attachment to primary caregiver	.51**	.11	.23**	.24**	.25**	.66**	—
Avoidant attachment to primary caregiver	.32**	−.09	.08	.07	.08	.62**	.61**	—
Anxious attachment to secondary caregiver	.37**	.06	.15*	.21**	.16*	.55**	.56**	.31**	—
Avoidant attachment to secondary caregiver	.24**	.09	.04	.12	.08	.45**	.25**	.43**	.61**	—
M	14.53	—	—	0.42	0.35	38.09	2.34	2.79	2.49	3.38
SD	11.71	—	—	0.68	0.63	13.57	1.09	1.47	1.12	1.62
Cronbach’s α	.93	—	—	—	—	.93	.92	.97	.89	.97
Range	0–55	—	—	0–3	0–2	25–97	1.0–6.2	1.0–6.7	1.0–6.3	1.0–7.0

BDI-II = Beck Depression Inventory–II; CTQ = Childhood Trauma Questionnaire.

p <.05. **p <.01.

Child maltreatment

The Childhood Trauma Questionnaire–Short Form (CTQ), a 28-item Likert-type scale self-report questionnaire, was used to assess experiences of childhood trauma (Bernstein et al., 2003). The instrument consisted of 25 items assessing childhood experiences of physical abuse, physical neglect, emotional abuse, emotional neglect, and sexual abuse, and 3 items assessed minimization of abuse experiences. Physical abuse was defined as a physical assault on a child that caused or could have caused physical harm (Bernstein et al., 2003). Physical neglect was a lack of appropriate care of a child, including basic needs such as adequate nutrition, clothing, safety, and medical care. Emotional abuse was characterized by an adult telling a child negative things about themself (e.g., demeaning or humiliating a child). Emotional neglect was characterized by ignoring a child’s emotional and psychological needs. Sexual abuse was defined as sexual behavior or contact occurring between a child and an older person (Bernstein et al., 2003). Each statement was answered on a Likert-type scale from 1 (never true) to 5 (very often true) with some items being reverse scored. These scores were summed to obtain an overall score with more severe CM, in terms of frequency, indicated with higher scores (Bernstein et al., 2003). The CTQ has demonstrated adequate psychometric properties (see Bernstein et al., 2003). The Cronbach’s alpha for this measure was high (α = .93; see Table 2).

Adult attachment

The Experiences in Close Relationships–Revised (ECR-R), a 36-item self-report measure, was used as an indicator of adult attachment (Fraley et al., 2000). It can be modified to measure several different kinds of attachment relationships including attachment to caregivers. Statements were rated on a 7-point Likert-type scale from 1 (strongly disagree) to 7 (strongly agree; e.g., “I do not often worry about being abandoned”). Eighteen items measured anxious attachment and the remaining 18 measured avoidant attachment (Fraley et al., 2000). The measure was scored by taking the mean of the anxious and avoidant items, so that two scores were obtained, one for anxious attachment and one for avoidant attachment (Fraley, 2012). The ECR-R has been widely used and its reliability is high; Cronbach’s alphas have been reported to range near .90 (Ravitz et al., 2010). Sibley et al. (2005) found that the ECR-R demonstrated convergent and discriminant validity when scores on the ECR-R predicted variance in anxiety and avoidance experiences in social interactions, as measured by diary entries. In the present study, the ECR-R was used to assess participants’ current attachment relationship to the individuals who were their primary and secondary caregivers in childhood as defined in the demographic survey; four scores were obtained for each participant (e.g., primary caregiver anxious attachment and primary caregiver avoidant attachment). Cronbach’s alphas in the present study were high and ranged from .89 to .97 (see Table 2).

Procedure

Courses at the university required that students engaged in research or an alternative research-related activity (e.g., read an article and write a summary) as a course requirement. If students chose to participate in research, they selected from the various studies that were currently being conducted at the university. Thus, participants self-selected to participate in the present study and were given course credit in exchange for their participation. After signing up for the study, they were redirected to the survey hosted on Qualtrics (https://www.qualtrics.com). After consent was obtained, the survey measures were presented in a randomized order. In addition, the order of items within each measure was randomized to overcome the concern of local item dependence (Royal, 2016). Local item dependence occurs when an individual’s response to one item on a survey biases their answer to another item on the survey (e.g., due to proximity of the two items; Royal, 2016). One way to overcome local item dependence is by randomizing items within a survey (Royal, 2016; Wilson et al., 2017).

Analyses

Covariates and order effects

Several demographic variables were examined as potential covariates as the prevalence of CM differs among races, sexes, and SESs. In a study examining rates of confirmed CM in the United States, Black children and female children were at a higher risk of having a confirmed case of CM than any other race or sex (Wildeman et al., 2014). Low SES has been found to be a risk factor for child abuse and neglect (Mulder et al., 2018; Stith et al., 2009). Rates of CM among individuals who identify as Deaf or hard-of-hearing are higher than those of hearing individuals (Schenkel et al., 2014). In addition, the rate and severity of depression differ among races (González et al., 2010).

Mental health conditions were of interest, specifically, the number of diagnoses one had, the number of treatments one was receiving (medication, therapy, and/or other), and family mental health. Ethnicity, sex, family’s SES (as indexed by annual household income), hearing status, number of mental health diagnoses, and number of mental health treatments were examined as potential covariates. Step AIC and best subsets regression analyses of all these potential covariates indicated the inclusion of sex, ethnicity, number of treatments, diagnosis total, and hearing status. These covariates were included in the inferential analyses. No effect of the order in which the measures were presented was found ps > .05. Correlations, means, and standard deviations of the variables included in the models are found in Table 2.

Moderation analyses

Path analyses were conducted in MPlus (version 8.3; Muthén & Muthén, 1998–2017) to evaluate saturated moderation models using maximum likelihood estimation with robust standard errors due to some variables that were not normally distributed. Ten cases were missing data on at least two or more variables. Outliers were first modified at three standard deviations above or below the mean for eight cases (Kline, 2011). All covariate and predictor variables in the model were mean-centered and the interaction terms were created using the mean-centered variables; four interaction terms were created (e.g., CTQ × Avoidant attachment with primary caregiver, CTQ × Anxious attachment with primary caregiver). The target population of the study was young adults; however, there were six participants who fell outside of this age range. Analyses were run with and without these participants and the significance of results did not change, therefore, these individuals are included in the final analyses.

Results

Descriptive Statistics

Depressive symptoms were correlated with all variables in the model except race. CM, anxious attachment to primary caregiver, and anxious attachment to secondary caregiver were associated with all variables in the model except for sex. Avoidant attachment to both primary and secondary caregivers was only associated with depressive symptoms, CM, and the other attachment variables (see Table 2). For individuals who identified their primary and secondary caregivers (n = 197), biological mothers comprised 85% of reported primary caregivers and biological fathers comprised 72% of reported secondary caregivers (see Table 1).

Attachment to Primary Caregiver

The first-order model for primary caregiver indicated that CM, anxious attachment to primary caregiver, and hearing status were significant in the model, such that higher CM, higher anxious attachment, and identifying as Deaf or hard-of-hearing were associated with greater depressive symptoms (see Table 3). The moderation model for primary caregivers indicated only anxious attachment to primary caregiver as a significant moderator of the relation between CM and depressive symptoms B = −0.16 (p = .002, 95% CI = [−0.25, −0.06]); avoidant attachment to primary caregiver was not a significant moderator.

Table 3.

Primary Caregiver First-Order Model Results.

Outcome	Variable	R ²	B	p	95% CI
BDI-II		.41		.00***
	Sex		1.20	.35	−1.30	3.71
	Hearing status		4.17	.04*	0.26	8.09
	Number of diagnoses		2.60	.09	−0.40	5.61
	Number of treatments		2.27	.18	−1.08	5.60
	Race		−0.18	.65	−0.94	0.59
	Anxious attachment		2.65	.02*	0.47	4.84
	Avoidant attachment		−0.07	.92	−1.54	1.39
	CTQ score		0.23	.01*	0.05	0.40

Note. All predictor and covariate variables in the model were centered except for the interaction term. CI = confidence intervals; BDI-II = Beck Depression Inventory–II; CTQ = Childhood Trauma Questionnaire.

p < .05. **p <.01. ***p <.001.

As CM increased, the effect of anxious attachment to primary caregiver decreased; likewise, as anxious attachment to primary caregiver increased, the effect of CM decreased. Meaning, at high levels of CM, anxious attachment to primary caregiver did not differentiate depressive symptoms. Similarly, CM did not differentiate scores on depressive symptoms at high levels of anxious attachment to primary caregiver. Simple slopes indicated that slopes were significant at low (one standard deviation below the mean) and at mean values of the interaction term (see Figure 1). At low levels of CM, mean anxious attachment and low anxious attachment significantly differentiated severity of depressive symptoms. Low CM interacted with mean anxious attachment to be associated with higher levels of depressive symptoms. In addition, low CM interacted with low levels of anxious attachment to predict even lower levels of depressive symptoms than mean levels of anxious attachment. Hearing status was the only other variable associated with BDI-II score, such that identifying as Deaf or hard-of-hearing was associated with higher levels of depressive symptoms. This model accounted for 43.5% of the variance in depressive symptoms (p < .001; see Table 4).

Figure 1.

Graph of simple slopes depicting interaction of anxious attachment to primary caregiver and Childhood Trauma Questionnaire score. Bold lines represent the slope, non-bold lines represent 95% confidence intervals for the slope.

Table 4.

Primary Caregiver Moderation Model Results.

Outcome	Variable	R ²	B	p	95% CI
BDI-II		.44		.00***
	Sex		0.90	.47	−1.51	3.26
	Hearing status		4.35	.04*	0.29	8.41
	Number of diagnoses		2.56	.09	−0.36	5.48
	Number of treatments		2.33	.17	−1.03	5.69
	Race		−0.22	.56	−0.96	0.52
	Anxious attachment		3.16	.00**	0.91	5.41
	Avoidant attachment		−0.66	.36	−2.07	0.76
	CTQ score		0.36	.00**	0.13	0.59
	Anxious attachment × CTQ		−0.16	.00**	−0.25	−.06
	Avoidant attachment × CTQ		0.03	.46	−0.05	0.11
Simple slopes
	Low anxious attachment		0.53	.00***	0.27	0.79
	Mean anxious attachment		0.36	.00**	0.13	0.59
	High anxious attachment		0.19	.12	−0.05	0.43

p < .05. **p <.01. ***p <.001.

Attachment to Secondary Caregiver

In the first-order model, only CM and hearing status were associated with depressive symptoms. Higher levels of CM and identifying as Deaf or hard-of-hearing were associated with increased levels of depressive symptoms (see Table 5). Neither the anxious nor avoidant attachment dimension with one’s secondary caregiver was a moderator of the relation between CM and depressive symptoms. CM and hearing status were significantly associated with BDI-II scores, such that higher CM, B = 0.34 (p <.001, 95% CI = [0.17, 0.52]), and identifying as Deaf or hard-of-hearing, B = 4.63 (p =.03, 95% CI = [0.50, 8.75]), were associated with higher levels of depressive symptoms. This model predicted 38.4% of the variance in depressive symptoms (p <.001; see Table 6).

Table 5.

Secondary Caregiver First-Order Model Results.

Outcome	Variable	R ²	B	p	95% CI
BDI-II		.38		.00***
	Sex		1.38	.31	−1.27	4.03
	Hearing status		4.54	.03*	0.54	8.54
	Number of diagnoses		2.60	.11	−0.55	5.74
	Number of treatments		2.66	.13	−0.79	6.10
	Race		−0.44	.26	−1.19	0.32
	Anxious attachment		1.22	.10	−0.24	2.69
	Avoidant attachment		−0.21	.73	−1.40	0.98
	CTQ score		0.32	.00***	0.18	0.46

p < .05. **p <.01. ***p <.001.

Table 6.

Secondary Caregiver Moderation Model Results.

Outcome	Variable	R ²	B	p	95% CI
BDI-II		.38		.00***
	Sex		1.36	.31	−1.27	3.99
	Hearing status		4.63	.03*	0.50	8.75
	Number of diagnoses		2.67	.11	−0.61	5.94
	Number of treatments		2.60	.16	−1.01	6.21
	Race		−0.45	.25	−1.21	0.32
	Anxious attachment		1.33	.07	−0.13	2.80
	Avoidant attachment		−0.31	.60	−1.47	0.86
	CTQ score		0.34	.00***	0.17	0.52
	Anxious attachment × CTQ		−0.04	.29	−0.12	0.04
	Avoidant attachment × CTQ		0.01	.75	−0.06	0.09

p < .05. **p <.01. ***p <.001.

Discussion

It was expected that both anxious and avoidant attachment dimensions in adults’ relationships with their primary and secondary caregivers would influence the relation between CM and depressive symptoms. This hypothesis was partially supported in that anxious attachment to primary caregivers was a moderator of the relation between CM and depressive symptomatology. Individuals who experienced low levels of CM and had low levels of anxious attachment to primary caregivers reported lower depressive symptomatology than individuals experiencing low levels of CM and mean or high levels of anxious attachment to primary caregivers. At high levels of anxious attachment, CM did not exhibit an effect on depressive symptoms; similarly, at high levels of CM, anxious attachment to primary caregiver did not exhibit an effect on depressive symptoms. No main or moderating effects were found for avoidant attachment with primary caregivers or either attachment dimension to secondary caregivers. It is well established that CM is strongly related to depression (Jaffee, 2017); however, the role of attachment to caregivers, especially in adulthood, has not been thoroughly explored.

Given that CM is most often perpetrated by caregivers, and that a high percentage of victims of abuse remain in the home (U.S. DHHS, 2020), it is important to examine the outcomes for these individuals as they develop and enter adulthood. Due to lifestyle changes that occur when attending college, students are particularly vulnerable to mental health issues (Ibrahim et al., 2013). In the present study, it is notable that, for individuals with high levels of attachment anxiety to primary caregivers, depressive symptoms did not increase, as would be anticipated, with increased CM. This indicates that at high levels of anxious attachment to primary caregivers, CM may not confer additional risk for depressive symptoms. Attachment styles to caregivers may affect an individual’s internal working models, shaping how they see both the self and others, emotion regulation, and their attachment styles to other important figures in their lives (Bowlby, 1988; Spruit et al., 2020). In turn, these factors may increase risk for depressive symptoms. High anxious attachment to primary caregivers may be a significant risk factor for depression to the extent that other risk factors may have little impact on increasing the risk for depressive symptoms. The present study emphasizes the importance of examining anxious attachment in the primary caregiver relationship as a target of intervention regardless of whether an individual experienced CM.

Whereas anxious attachment demonstrated a relation to depressive symptoms, avoidant attachment to either caregiver had no main or moderating effect. These findings provide additional evidence that anxious and avoidant attachment dimensions may be related to psychopathology in different ways. Anxious attachment refers to a fear of abandonment, which may be the consequence of a belief that oneself is unworthy of love (Bartholomew & Horowitz, 1991; Gillath et al., 2016). Avoidance refers to a fear of closeness and dependency on others, which may stem from a belief that others are untrustworthy and unavailable (Bartholomew & Horowitz, 1991; Gillath et al., 2016). In the context of the present study, fears relating to unworthiness and abandonment regarding one’s primary caregiver affected the relation between low levels of CM and depression, and at high levels, was associated with high depressive symptoms regardless of levels of CM. Concerns about closeness and dependence did not affect this relationship in the present study.

There are important differences in the way that individuals high in anxious or avoidant attachment styles react to threat, and these emotional regulatory differences may affect one’s vulnerability to depression (Ein-Dor et al., 2010). As individuals who are high in avoidance want to avoid closeness and dependence, they attempt to confront stressors independently, without the help of others (Ein-Dor et al., 2010). One strategy these individuals may employ is trying to reduce the seriousness of threat and deactivate emotions in response to threat. Alternatively, individuals high in anxious attachment tend to rely on others for help when they feel threatened, thus they may engage in hyperactivation of emotions in response to threat, in the hopes of getting others to help them. This emotional overarousal leads to the use of ineffective emotion-based coping strategies such as catastrophizing, and a focus on distressing information (Ein-Dor et al., 2010). Those utilizing emotion-based coping strategies focus on adverse emotions which in turn intensify the experience of these negative emotions and can lead to the development of psychopathology (Hofmann et al., 2012; Pascuzzo et al., 2015). Past research has found that emotion-focused coping mediated the relation between anxious attachment in romantic relationships and psychopathology (Pascuzzo et al., 2015). An anxious attachment style may be characterized by emotion-based coping strategies, which can cause individuals to become preoccupied with negative emotions and can lead to the development of depressive symptomatology.

Given the information that those who are high in avoidant attachment do not react emotionally to threat, and tend to downplay the seriousness of threat, it might be expected that avoidant attachment would moderate the relation between CM and depression in the opposite direction (i.e., high avoidance attenuates the effect of CM on depression). One reason this might not have been the case is that the present study did not assess individuals’ current primary attachment relationship; as one ages, attachment may transfer from parents to romantic partners (Feeney, 2004). This may explain why avoidant attachment to caregivers did not significantly moderate the relation between CM and depression. The present study may have found different results if participants were asked about their current most significant attachment relationship. Individuals with avoidant attachments to parents may have a different attachment figure to whom their attachment is more salient in how they respond to threat. If this attachment relationship was assessed, it is possible that avoidant attachment may have moderated the relation between avoidant attachment and depressive symptoms. It is important to note that individuals can be high in both anxious and avoidant attachment dimensions, as these constructs are not mutually exclusive. However, in the analyses, these constructs were treated as separate variables to provide a more accurate portrayal of the complexity of attachment. Through analyzing the construct of attachment in this way, it is clear that anxious and avoidant styles may have different effects on the development of psychopathology and symptomatology.

Attachment to secondary caregivers did not significantly affect depressive symptoms or the association between CM and depressive symptoms. Models have been developed that attempt to explain the importance of different attachment figures on child outcomes (Bretherton, 2010). One of these models, attachment hierarchy, suggests that the primary attachment figure is preferred when a child is distressed and that this primary figure has more influence on child outcomes compared with other attachment figures (Al-Yagon et al., 2016; van IJendoorn et al., 1992). The present study may provide support for the attachment hierarchy theory, as attachment anxiety to primary caregivers, but not secondary caregivers, was found to be a moderator in the relationship between CM and depressive symptoms.

Other studies have also found support for the attachment hierarchy model; Umemura et al. (2013) found that distressed toddlers sought their primary caregiver, defined as the caregiver who spent more time with them and engaged in more caregiving activities, regardless of their history of security with this caregiver (e.g., even if the attachment to their primary caregiver was not secure). This result found by Umemura et al. (2013) suggests that it may not be the security of the relationship that affects child preference, but the status of being the primary caregiver. Early attachment styles are developed through the parent-child relationship (Ainsworth & Bell, 1970; Bowlby, 1988); it may be that the parent who spends more time caring for the child may have a more significant influence on attachment-related outcomes that increase vulnerability to depressive symptoms (e.g., internal working models, emotion regulation, and attachment style to others). Caregivers who are less involved in caring for a child, such as secondary caregivers, may have less of an impact on these outcomes.

Strengths and Limitations

The present study contributed to the existing literature in several important ways. This study is unique in its examination of both primary and secondary caregiver attachment relationships. Research often focuses on primary caregivers and romantic attachment is typically the focus of research with adults. Although many individuals’ primary attachment relationship may be transferring from parents to romantic partners in this period, research indicates that attachment styles developed in childhood typically remain stable into adulthood (Fraley et al., 2011; Hamilton, 2000; Waters et al., 2000). This study showed that primary caregiver attachment was influential, even in a sample made up primarily of young adults (97% of the sample) who may have an attachment relationship to other figures. Young adults face significant challenges and changes in life; independence, autonomy, identity formation, personal and professional issues are just some of the stage salient tasks this population has to navigate (Toth & Cicchetti, 2013). Many young adults move away from caregivers and the familiarity of their home environment to get an education or a job. Young adults may also establish significant relationships with friends and romantic partners. A history of CM and poor attachment to caregivers may increase the difficulty of success with these tasks, as well as increase one’s vulnerability to developing psychopathology. Results from the current sample may generalize to other samples of young adults attending college.

The present sample was not a clinical sample, which can be considered another advantage of this study (Corwin & Keeshin, 2011). The results of this study reflect the symptomatology of young adults who have experienced CM but may not have developed psychopathology or be diagnosed. This is an important population that may sometimes be ignored due to their subclinical symptoms or disorders. Notably, the mean BDI-II score in the current sample was 14.53 (SD = 11.71), which is indicative of mild depression (Wang & Gorenstein, 2013). A large proportion of the present sample (44.39%) scored in the mild to severe depression range on the BDI-II (Wang & Gorenstein, 2013). Previous studies measuring depression in college students with the BDI-II have found lower scores (M = 11.03, SD = 8.17, N = 414; Storch et al., 2004; M = 9.27, SD = 8.07, N = 7,369; Whisman et al., 2012). The depression scores in the current sample may be a result of the self-selection of the participants due to the overt focus on childhood adversity and mental health depicted in study materials. In the present study, rates of CM were comparable to community samples. In community samples (n = 12,432–12,915) the mean CTQ score was 38.78 (SD = 14.98; MacDonald et al., 2016). The present sample had a mean CTQ score of 38.09 (SD =13.57). This speaks to the pervasiveness of CM and its deleterious effects across demographic characteristics such as race, ethnicity, and sex. It is also notable that identifying as Deaf and hard-of-hearing (n = 13; 6.6% of the sample) was significantly associated with depressive symptoms across models. Previous research has found that rates of anxiety or depression were significantly higher in adults identifying as Deaf (Kushalnagar et al., 2019). College students who identify as Deaf or hard-of-hearing may be at increased risk for depressive symptoms.

Limitations of the present study include that it was cross-sectional in nature, thus conclusions about causality cannot be drawn. Results should be interpreted carefully, as it is possible that depressive symptoms may influence respondents’ answers on other measures. The directionality of the results should also be interpreted with caution as the self-reports were retrospective in nature. Ideally, research questions such as these would be studied in a longitudinal design. As mentioned previously, participants self-selected to participate in this study based on a description of the study, which may have resulted in sampling bias. The study was conducted online thus participants were not taking the survey in the same environment. Furthermore, it is possible that participants may have answered carelessly due to the online survey environment. However, measures did include reverse-scored items and alphas were high among measures, suggesting that internal consistency was high. Researchers using surveys should take steps to address potential careless responding.

Clinical Implications and Future Directions

A secure attachment relationship between caregiver and child can help a child develop positive internal working models of the self and others, emotion regulation, and increase feelings of security in relationships (Spruit et al., 2020). Attachment relationships, especially those to mothers, have been identified as an important intervention target after experiences of CM (Valentino, 2017). Notably, 85% of the participants in the present study identified their mother as their primary caregiver. In 2018, mothers, alone or with a partner, were the most common perpetrators of CM in the United States (U.S. DHHS, 2020). Thus, many interventions targeting the mother–child relationship in families with histories of CM have been developed and evaluated (Valentino, 2017). These interventions have demonstrated improvements in children’s attachment styles, such that children’s attachment becomes more secure (Valentino, 2017). The present study underscores the importance of individuals’ attachment style to primary caregivers and provides further evidence that an anxious attachment style in this relationship continues to be associated with adverse outcomes into adulthood. Historically, most attachment research has focused on the mother–child attachment relationship; recently more focus has been spent on examining both mother and father relationships (Bretherton, 2010). Due to a large research emphasis on mothers, as well the fact that in the United States, mothers are more likely to be the primary caregiver (Umemura et al., 2013), the effects of primary caregivers and the impact of mothers have been confounded. Future research should consider examining attachment utilizing a primary/secondary caregiver distinction instead of a mother/father distinction.

Additional research is needed to examine the adaptive facets of attachment styles that are typically thought of as maladaptive. Attachment anxiety and avoidance are typically conceptualized as hindrances to psychological well-being. However, evidence to the contrary exists, suggesting that for adults, attachment anxiety and avoidance may actually lead to adaptive functioning in some domains (Ein-Dor et al., 2010). Future research examining disorganized and fearful attachment types in adults will be important in developing a more thorough picture of the complexity of attachment throughout the lifespan and how CM may influence adult attachment.

Future research should examine adult attachment in samples that are more racially and ethnically diverse and in samples of young adults who may not attend college to explore the generalizability of the results. Further longitudinal research elucidating risk and protective factors in the relation between CM and mental health outcomes is necessary to develop interventions. Complex models that include additional mediating, and moderating variables, such as emotion regulation and additional mental health diagnoses, would be beneficial in creating a more comprehensive picture of the intricacies of CM and its outcomes. Longitudinal studies examining the effects of CM into adulthood are needed to establish causality and further explore the long-lasting effects of experiencing CM. Research and intervention on the environment in which CM is perpetrated are also needed for prevention of the occurrence of CM. CM is a heavily studied topic in the field of psychology; however, it remains an unsolved problem in society.

Conclusion

CM and attachment styles have long-lasting effects on depressive symptoms into adulthood. Consistent with prior research, the present study found that individuals who experienced more frequent CM reported higher levels of depressive symptoms. High anxious attachment to primary caregivers was associated with increased depressive symptoms; however, contrary to hypotheses, it did not interact with CM. Only low and mean levels of anxious attachment to primary caregivers interacted with a low frequency of CM and these interactions, in turn, were associated with low depressive symptoms. Contrary to expectations, anxiety and avoidance in the attachment relationship to secondary caregivers and avoidant attachment to primary caregivers were not related to depressive symptoms. Results suggest that anxious attachment to primary caregivers has continued influence on mental health outcomes into adulthood and at high levels of anxious attachment, CM does not confer additional risk for depressive symptoms. In adults, CM and anxious attachment to primary caregiver still continue to affect mental health.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Sarah Nowalis

Author Biographies

Sarah Nowalis, MS, is a project coordinator in the Social Development Lab at the State University of New York at Buffalo. She is interested in how trauma, stress, and environmental factors influence the development of aggression and externalizing behaviors in children and adolescents.

Stephanie A. Godleski, PhD, is an assistant professor in the Department of Psychology at the Rochester Institute of Technology. The primary focus of her research is pathways of risk and resilience in predicting social competence across the lifespan. Specifically, she has focused on protective (e.g., parenting) and maladaptive (e.g., prenatal exposure) processes that predict and maintain negative social-emotional developmental and health outcomes.

Lindsay S. Schenkel, PhD, is an associate professor in the Department of Psychology at Rochester Institute of Technology. Her research focuses on risk factors, such as early trauma, in the development of later mood and psychotic disorders. She is also involved in the development of preventive interventions for at-risk youth and young adults that incorporate family therapy, cognitive, and social-cognitive treatments.

References

Afifi

T. O.

MacMillan

H. L.

(2011). Resilience following child maltreatment: A review of protective factors. Canadian Journal of Psychiatry, 56(5), 266–272. https://doi.org/10.1177/070674371105600505

Ainsworth

Bell

(1970). Attachment, exploration, and separation: Illustrated by the behavior of one-year-olds in a strange situation. Child Development, 41(1), 49–67. https://doi.org/10.2307/1127388

Al-Yagon

Kopelman-Rubin

Klomek

A. B.

Mikulincer

(2016). Four-model approach to adolescent-parent attachment relationships and adolescents’ loneliness, school belonging, and teacher appraisal. Personal Relationships, 23, 141–158. https://doi.org/10.1111/pere.12116

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders: DSM-5 (5th ed.).

Bartholomew

Horowitz

L. M.

(1991). Attachment styles among young adults: A test of four-category model. Journal of Personality and Social Psychology, 61(2), 226–244. https://doi.org/10.1037/0022-3514.61.2.226

Beck

A. T.

Steer

R. A.

Brown

G. K.

(1996). Manual for the Beck Depression Inventory-II. Psychological Corp.

Beeney

J. E.

Wright

A. G. C.

Stepp

S. D.

Hallquist

M. N.

Lazarus

S. A.

Beeney

J. R. S.

Scott

L. N.

Pilkonis

P. A.

(2017). Disorganized attachment and personality functioning in adults: A latent class analysis. Personality Disorders: Theory, Research, and Treatment, 8(3), 206–216. https://doi.org/10.1037/per0000184

Bernstein

D. P.

Stein

J. A.

Newcomb

M. D.

Walker

Pogge

Ahluvalia

Stokes

Handelsman

Medrano

Desmond

Zule

(2003). Development and validation of a brief screening version of the Childhood Trauma Questionnaire. Child Abuse and Neglect, 27(2), 169–190. https://doi.org/10.1016/S0145-2134(02)00541-0

Bifulco

Kwon

Jacobs

Moran

P. M.

Bunn

Beer

(2006). Adult attachment style as mediator between childhood neglect/abuse and adult depression and anxiety. Social Psychiatry and Psychiatric Epidemiology, 41(10), 796–805. https://doi.org/10.1007/s00127-006-0101-z

10.

Bowlby

(1988). A secure base: Parent-child attachment and healthy human development. Basic Books.

11.

Bretherton

(2010). Fathers in attachment theory and research: A review. Early Child Development and Care, 180(1–2), 9–23. https://doi.org/10.1080/03004430903414661

12.

Cicchetti

Banny

(2014). A developmental psychopathology perspective on child maltreatment. In Lewis

Rudolph

K. D.

(Eds.), Handbook of developmental psychopathology (pp. 723–741). Springer.

13.

Collishaw

Pickles

Messer

Rutter

Shearer

Maughan

(2007). Resilience to adult psychopathology following childhood maltreatment: Evidence from a community based sample. Child Abuse and Neglect, 31(3), 211–229. https://doi.org/10.1016/j.chiabu.2007.02.004

14.

Conradi

H. J.

Kamphuis

J. H.

de Jonge

(2018). Adult attachment predicts the seven-year course of recurrent depression in primary care. Journal of Affective Disorders, 225, 160–166. https://doi.org/10.1016/j.jad.2017.08.009

15.

Corwin

D. L.

Keeshin

B. R.

(2011). Estimating present and future damages following child maltreatment. Child and Adolescent Psychiatric Clinics of North America, 20(3), 505–518. https://doi.org/10.1016/j.chc.2011.03.005

16.

Dagan

Facompré

C. R.

Bernard

(2018). Adult attachment representations and depressive symptoms: A meta-analysis. Journal of Affective Disorders, 236, 274–290. https://doi.org/10.1016/j.jad.2018.04.091

17.

Ein-Dor

Mikulincer

Doron

Shaver

P. R.

(2010). The attachment paradox: How can so many of us (the insecure ones) have no adaptive advantages? Perspectives on Psychological Science, 5(2), 123–141. https://doi.org/10.1177/1745691610362349

18.

Feeney

J. A.

(2004). Transfer of attachment from parents to romantic partners: Effects of individual and relationship variables. Journal of Family Studies, 10, 220–238. https://doi.org/10.5172/jfs.327.10.2.220

19.

Fraley

R. C.

(2012, June 4). Information on the experiences in Close Relationships-Revised (ECR-R) Adult Attachment Questionnaire. http://labs.psychology.illinois.edu/~rcfraley/measures/ecrr.htm

20.

Fraley

R. C.

Vicary

A. M.

Brumbaugh

C. C.

Roisman

G. I.

(2011). Patterns of stability in adult attachment: An empirical test of two models of continuity and change. Journal of Personality and Social Psychology, 101(5), 974–992. https://doi.org/10.1037/a0024150

21.

Fraley

R. C.

Waller

N. G.

Brennan

K. A.

(2000). An item-response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78, 350–365. https://doi.org/10.1037/0022-3514.78.2.350

22.

Gillath

Karantzas

G. C.

Fraley

R. C.

(2016). Adult attachment: A concise introduction to theory and research. Academic Press.

23.

González

H. M.

Tarraf

Whitfield

K. E.

Vega

W. A.

(2010). The epidemiology of major depression and ethnicity in the United States. Journal of Psychiatric Research, 44(15), 1043–1051. https://doi.org/10.1016/j.jpsychires.2010.03.017

24.

Hamilton

C. E.

(2000). Continuity and discontinuity of attachment from infancy through adolescence. Child Development, 71(3), 690–694. https://doi.org/10.1111/1467-8624.00177

25.

Hazan

Shaver

(1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511–524. https://doi.org/10.1037/0022-3514.52.3.511

26.

Hofmann

S. G.

Sawyer

A. T.

Fang

Asnaani

(2012). Emotion dysregulation model of mood and anxiety disorders. Depression and Anxiety, 29, 409–416. https://doi.org/10.1002/da.21888

27.

Ibrahim

A. K.

Kelly

S. J.

Adams

C. E.

Glazebrook

(2013). A systematic review of studies of depression prevalence in university students. Journal of Psychiatric Research, 47, 391–400. https://doi.org/10.1016/j.jpsychires.2012.11.015

28.

Jaffee

S. R.

(2017). Child maltreatment and risk for psychopathology in childhood and adulthood. Annual Review of Clinical Psychology, 13, 525–551. https://doi.org/10.1146/annurev-clinpsy-032816-045005

29.

Jones

J. D.

Fraley

R. C.

Ehrlich

K. B.

Stern

J. A.

Lejuez

C. W.

Shaver

P. R.

Cassidy

(2018). Stability of attachment style in adolescence: An empirical test of alternative developmental processes. Child Development, 89(3), 871–880. https://doi.org/10.1111/cdev.12775

30.

Kline

R. B.

(2011). Methodology in the social sciences: Principles and practice of structural equation modeling (3rd ed.). Guilford Press.

31.

Kushalnagar

Reesman

Holcomb

Ryan

(2019). Prevalence of anxiety or depression diagnosis in deaf adults. The Journal of Deaf Studies and Deaf Education, 24(4), 378–385. https://doi.org/10.1093/deafed/enz017

32.

D’Arcy

Meng

(2016). Maltreatment in childhood substantially increases the risk of adult depression and anxiety in prospective cohort studies: Systematic review, meta-analysis, and proportional attributable fractions. Psychological Medicine, 46(4), 717–730. https://doi.org/10.1017/S0033291715002743

33.

Lipson

S. K.

Lattie

E. G.

Eisenberg

(2019). Increased rates of mental health service utilization by U.S. college students: 10-year population-level trends (2007–2017). Psychiatric Services, 70(1), 60–63. https://doi.org/10.1176/appi.ps.201800332

34.

MacDonald

Thomas

M. L.

Sciolla

A. F.

Schneider

Pappas

Bleijenberg

Bohus

Bekh

Carpenter

Carr

Dannlowski

Dorahy

Fahlke

Finzi-Dottan

Karu

Gerdner

Glaesmer

Grabe

H. J.

Heins

. . . Wingenfeld

(2016). Minimization of childhood maltreatment is common and consequential: Results from a large multinational sample using the Childhood Trauma Questionnaire. PLOS ONE, 11(1), Article e0146058. https://doi.org/10.1371/journal.pone.0146058

35.

Malik

Wells

Wittkowski

(2015). Emotion regulation as a mediator in the relationship between attachment and depressive symptomatology: A systematic review. Journal of Affective Disorders, 172, 428–444. https://doi.org/10.1016/j.jad.2014.10.007

36.

Meng

Fleury

Xiang

D’Arcy

(2018). Resilience and protective factors among people with a history of child maltreatment: A systematic review. Social Psychiatry and Psychiatric Epidemiology, 53, 453–475. https://doi.org/10.1007/s00127-018-1485-2

37.

Mulder

T. M.

Kuiper

K. C.

van der Put

C. E.

Stams

G. J. M.

Assink

(2018). Risk factors for child neglect: A meta-analytic review. Child Abuse and Neglect, 77, 198–210. https://doi.org/10.1016/j.chiabu.2018.01.006

38.

Muthén

L. K.

Muthén

B. O.

(1998–2017). Mplus user’s guide (8th ed.).

39.

Nanni

Uher

Danese

(2012). Childhood maltreatment predicts unfavorable course of illness and treatment outcome in depression: A meta-analysis. American Journal of Psychiatry, 169(2), 141–151. https://doi.org/10.1176/appi.ajp.2011.11020335

40.

Nelson

Klumparendt

Doebler

Ehring

(2017). Childhood maltreatment and characteristics of adult depression: Meta-analysis. The British Journal of Psychiatry, 201, 96–104. https://doi.org/10.1192/bjp.bp.115.180752

41.

Oshri

Sutton

T. E.

Clay-Warner

Miller

J. D.

(2015). Child maltreatment types and risk behaviors: Associations with attachment style and emotion regulation dimensions. Personality and Individual Differences, 73, 127–133. https://doi.org/10.1016/j.paid.2014.09.015

42.

Pascuzzo

Moss

Cyr

(2015). Attachment and emotion regulation strategies in predicting adult psychopathology. SAGE Open, 5, 1–15. https://doi.org/10.1177/2158244015604695

43.

Ravitz

Maunder

Hunter

Sthankiya

Lancee

(2010). Adult attachment: A 25-year review. Journal of Psychosomatic Research, 69, 419–432. https://doi.org/10.1016/j.jpsychores.2009.08.006

44.

Royal

K. D.

(2016). The impact of item sequence order on local item dependence: An item response theory perspective. Survey Practice, 9(4), 1–7.

45.

Schenkel

L. S.

Rothman-Marshall

Schlehofer

D. A.

Towne

T. L.

Burnash

D. L.

Priddy

B. M.

(2014). Child maltreatment and trauma exposure among deaf and hard of hearing young adults. Child Abuse and Neglect, 38(10), 1581–1589. https://doi.org/10.1016/j.chiabu.2014.04.010

46.

Sibley

C. G.

Fischer

Liu

J. H.

(2005). Reliability and validity of the revised experiences in close relationships (ECR-R) self-report measure of adult romantic attachment. Personality and Social Psychology Bulletin, 31(11), 1524–1536. https://doi.org/10.1177/0146167205276865

47.

Spruit

Goos

Weenink

Rodenburg

Niemeyer

Stams

Colonnesi

(2020). The relation between attachment and depression in children and adolescents: A multilevel meta-analysis. Clinical Child and Family Psychology Review, 23, 54–69. https://doi.org/10.1007/s10567-019-00299-9

48.

Stith

M. S.

Liu

T. L.

Davies

Boykin

E. L.

Alder

M. C.

Harris

J. F.

Som

McPherson

Dees

J. E. M. E. G

. (2009). Risk factors in child maltreatment: A meta-analytic review of the literature. Aggression and Violent Behavior, 14(1), 13–29. https://doi.org/10.1016/j.avb.2006.03.006

49.

Storch

E. A.

Roberti

J. W.

Roth

D. A.

(2004). Factor structure, concurrent validity, and internal consistency of the Beck Depression Inventory—Second Edition in a sample of college students. Depression and Anxiety, 19(3), 187–189. https://doi.org/10.1002/da.20002

50.

Stronach

E. P.

Toth

S. L.

Rogosch

Cicchetti

(2013). Preventative interventions and sustained attachment security in maltreated children. Development and Psychopathology, 25(4 Pt. 1), 919–930. https://doi.org/10.1017/S0954579413000278

51.

Toth

S. L.

Cicchetti

(2013). A developmental psychopathology perspective on child maltreatment. Child Maltreatment, 18(3), 135–139. https://doi.org/10.1177/1077559513500380

52.

Toth

S. L.

Gravener-Davis

J. A.

Guild

D. J.

Cicchetti

(2013). Relational interventions for child maltreatment: Past, present and future perspectives. Developmental Psychopathology, 25(4 Pt. 2), 1601–1617. https://doi.org/10.1017/S0954579413000795

53.

Umemura

Jacobvitz

Messina

Hazen

(2013). Do toddlers prefer the primary caregiver or the parent with whom they feel more secure? The role of toddler emotion. Infant Behavior and Development, 36, 102–114. https://doi.org/10.1016/j.infbeh.2012.10.003

54.

U.S. Department of Health & Human Services, Administration for Children and Families, Administration on Children, Youth and Families, & Children’s Bureau. (2020). Child maltreatment 2018. https://www.acf.hhs.gov/cb/research-data-technology/statistics-research/child-maltreatment

55.

Valentino

(2017). Relational interventions for maltreated children. Child Development, 88, 359–367. https://doi.org/10.1111/cdev.12735

56.

van IJzendoorn

M. H.

Sagi

Lambermon

M. W. E.

(1992). The multiple caretaker paradox: Data from Holland and Israel. New Directions for Child and Adolescent Development, 57, 5–24 https://doi.org/10.1002/cd.23219925703

57.

Wang

Gorenstein

(2013). Psychometric properties of the Beck Depression Inventory-II: A comprehensive review. Revista Brasileira de Psiquiatria, 35(4), 416–431. https://doi.org/10.1590/1516-4446-2012-1048

58.

Waters

Merrick

Treboux

Crowell

Albersheim

(2000). Attachment security in infancy and early adulthood: A twenty-year longitudinal study. Child Development, 71(3), 684–689. https://doi.org/10.1111/1467-8624.00176

59.

Whisman

M. A.

Judd

C. M.

Whiteford

N. T.

Gelhorn

H. L.

(2012). Measurement invariance of the Beck Depression Inventory—Second Edition (BDI-II) across gender, race, and ethnicity in college students. Assessment, 20(4), 419–428. https://doi.org/10.1177/1073191112460273

60.

Wildeman

Emanuel

Leventhal

J. M.

Putnam-Hornstein

Waldfogel

Lee

(2014). The Prevalence of confirmed maltreatment among US children, 2004-2011. Journal of the American Medical Association Pediatrics, 168(8), 706–713. https://doi.org/10.1001/jamapediatrics.2014.410

61.

Wilson

E. V.

Srite

Loiacono

(2017). A call for item-ordering transparency in online IS survey administration. [Conference session]. 22nd Americas Conference on Information Systems, Boston, MA, United States.

62.

Zivin

Eisenberg

Gollust

S. E.

Golberstein

(2009). Persistence of mental health problems and needs in a college student population. Journal of Affective Disorders, 117(3), 180–185. https://doi.org/10.1016/j.jad.2009.01.001