The Devastating Clinical Consequences of Child Abuse and Neglect: Increased Disease Vulnerability and Poor Treatment Response in Mood Disorders

It is estimated that one in four children will experience child abuse or neglect at some point in their lifetime, and one in seven children have experienced abuse over the past year. In 2016, 676,000 children were reported to child protective services in the United States and identified as victims of child abuse or neglect (1). However, it is widely accepted that statistics on such reports represent a significant underestimate of the prevalence of childhood maltreatment, because the majority of abuse and neglect goes unreported. This is especially true for certain types of childhood maltreatment (notably emotional abuse and neglect), which may never come to clinical attention but have devastating consequences on health independently of physical abuse and neglect or sexual abuse. Although rates of children being reported to child protective services have remained relatively consistent over recent decades (Figure 1), our understanding of the devastating medical and clinical consequences of childhood maltreatment has grown, and childhood maltreatment is now well established as a major risk factor for adult psychopathology. In this review, we seek to summarize the burgeoning literature on childhood maltreatment, specifically focusing on the link between childhood maltreatment and mood disorders (depression and bipolar disorder). The data converge to point toward future directions for education, prevention, and treatment to decrease the consequences of childhood maltreatment, especially in regard to mood disorders.

The link between childhood maltreatment and risk for mood disorders and differences in disease course following illness onset has been well documented (2–8). Multiple studies have demonstrated greater rates of childhood maltreatment in patients with major depression and bipolar disorder (9–11). Indeed, a recent meta-analysis revealed that 46% of individuals with depression report childhood maltreatment (12). Patients with bipolar disorder also report high levels of childhood maltreatment (13, 14), with estimates as high as 57% (15). Childhood maltreatment is associated with an increased risk and earlier onset of unipolar depression, with syndromal depression occurring on average 4 years earlier in individuals with a history of childhood maltreatment compared with those without such a history (12). Childhood maltreatment is also associated with a more pernicious disease course, including a greater number of lifetime depressive episodes and greater depression severity, with the majority of studies showing more recurrence and greater persistence of depressive episodes (16–18). For example, Wiersma et al. (19), in an analysis of 1,230 adults with major depressive disorder drawn from the Netherlands Study of Depression and Anxiety, found that childhood maltreatment (measured with the Childhood Trauma Interview) was associated with chronicity of depression, defined as being depressed for ≥24 months over the past 4 years, independent of comorbid anxiety disorders, severity of depressive symptoms, or age at onset. Increased risk for suicide attempts and comorbidities, including increased rates of anxiety disorders, PTSD, and substance use disorders, are reported in individuals with depression who experience childhood maltreatment. Individuals with major depressive disorder and atypical features report significantly more traumatic life events (including physical abuse, sexual abuse, and other forms of trauma) both before and after their first depressive episode, independently of sex, age at onset, or duration of depression (20). Additionally, childhood maltreatment has consistently been shown to be associated with poor treatment outcome (after psychotherapy, pharmacotherapy, and combined treatment) in depression, as assessed by lack of remission or response or longer time to remission (12, 18, 21, 22).

Although the studies cited above describe a link between childhood maltreatment and a more pernicious depression course, most studies have been cross-sectional, and the possibility of recall bias and mood effects (owing to the retrospective investigation of childhood maltreatment in individuals who are currently depressed) cannot be ruled out. However, studies over the past few years comparing retrospective and prospective measurement of childhood maltreatment suggest consistency between retrospective reports and prospective designs (23, 24), although a recent meta-analysis (25) suggested poor agreement between these measures, with better agreement observed when retrospective measures were based on interviews and in studies with smaller samples. Longitudinal and prospective studies are emerging that have further confirmed and extended our understanding of the devastating consequences of childhood maltreatment on illness course (5, 7). Ellis et al. (26) recently reported that childhood maltreatment increased risk for more severe trajectories of depressive symptoms during a 7-year longitudinal study in 243 adolescents in the Orygen Adolescent Development Study. Gilman et al. (27) reported that childhood maltreatment increased the risk for recurrent depressive episodes and suicidal ideation by 20%−30% during a 3-year follow-up of 2,497 participants diagnosed with major depressive disorder in the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). Additionally, Widom et al. (7), in a study that followed a cohort of 676 children with documented childhood maltreatment and compared risk for major depression in adulthood between them and a cohort of 520 children matched on age, race, sex, and family social class who were not exposed to childhood maltreatment, found a clear association between childhood maltreatment and both increased risk for depression and earlier onset of the disorder.

Although more research has been reported investigating the link between childhood maltreatment and disease onset and course in unipolar depression, more recent evidence supports the link between childhood maltreatment and disease onset and course in bipolar disorder (28). Childhood maltreatment is associated with increased disease vulnerability and earlier age at onset of bipolar disorder (29). Jansen et al. (30) sought to determine whether childhood maltreatment mediated the effect of family history on diagnosis of a mood disorder. The findings indicated that one-third of the effect of family history on risk for mood disorders was mediated by childhood maltreatment. As with depression, studies on bipolar disorder with a prospective or longitudinal approach are few, but they are informative. Using data from the NESARC (N=33,375), Gilman et al. (31) found that childhood physical and sexual abuse were associated with increased risk for first-onset and recurrent mania independently of recent life stress. An association between childhood maltreatment and prodromal symptoms has also been reported in bipolar disorder (32), suggesting that childhood maltreatment may contribute to disease vulnerability before onset of the first manic episode. Childhood maltreatment in the context of bipolar disorder is also associated with a more pernicious disease course, including greater frequency and severity of mood episodes (both depressive and manic), greater severity of psychosis symptoms, and greater risk for comorbidities (i.e., anxiety disorders, PTSD, substance use disorders), rapid cycling, inpatient hospitalizations, and suicide attempts (28, 33–41). Studies are beginning to emerge investigating treatment response in bipolar disorder following childhood maltreatment. Such studies remain few, but they suggest that childhood maltreatment is associated with a poor response to benzodiazepines (42) and anticonvulsants (41) in bipolar disorder. The concatenation of findings in depression and bipolar disorder are concordant in that childhood maltreatment increases risk for, and early onset of, first mood episode and episode recurrence. Childhood maltreatment affects disease trajectories, including in its association with more insidious mood episodes, poor treatment response, a greater risk for comorbidities, and a greater risk for suicide ideation, attempts, and completion. The link between childhood maltreatment and increased prevalence of suicide-related behaviors is of particular importance given the high rate of suicide ideation, attempts, and completion in depression and bipolar disorder. Despite many prevention strategies (e.g., education and outreach and clinical studies to identify risk factors for impending suicide attempts in individuals with mood disorders), suicide rates have not decreased but in fact have increased in the United States. The link between childhood maltreatment and suicide-related behavior has been reviewed by several groups (21, 33, 43–47). Dube et al. (48) reported that adverse childhood experiences, including childhood maltreatment, increased the risk for suicide attempts twofold to fivefold in 17,337 adults in the now classic Adverse Childhood Experiences Study. Gomez et al. (49) reported that physical or sexual abuse increased the odds of suicide ideation, planning, and attempts among the 9,272 adolescents in the U.S. National Comorbidity Survey Adolescent Supplement. Miller et al. (50) examined the relationship between childhood maltreatment and prospective suicidal ideation in a cohort of 682 youths followed over a 3-year period. Emotional maltreatment predicted suicidal ideation, independently of previous suicidal ideation and depressive symptom severity. Childhood maltreatment is also associated with earlier age at first suicide attempt (51). Additionally, an association between childhood maltreatment and suicide risk in 449 individuals age 60 or older was recently reported from the Multidimensional Study of the Elderly, in the Family Health Strategy in Porto Alegre, Brazil (52). The effect was independent of depressive symptom severity. These findings suggest that childhood maltreatment increases risk for suicide-related behavior across the lifespan. More work is warranted in investigating the biological mechanisms that may mediate the association between childhood maltreatment and suicide-related behaviors.

Although childhood maltreatment at any age can result in long-lasting consequences (53), there is evidence that the timing, duration, and severity of maltreatment mediate the risk for later psychopathology (54). Childhood maltreatment that occurs earlier in life and continues for a longer duration is associated with the worst outcomes (55). This is supported by preclinical models (rodent and nonhuman primate) that investigated maternal separation (56, 57), a paradigm more similar to neglect in humans. One study in rodents found that maternal separation during the early postnatal period (days 2–15) but not the later postnatal period (days 7–20) is associated with anxious and depressive-like behaviors in adulthood (57). Although this postnatal period coincides with in utero development in humans, there is evidence that in utero insults in the form of stress can have consequences similar to early-life trauma (58, 59), supporting the translational validity of these models. Clinical studies also support the importance of timing of childhood maltreatment in moderating risk for psychopathology. Cowell et al. (60) investigated the timing and duration of childhood maltreatment in 223 maltreated children between the ages of 3 and 9 and found that children who were maltreated during infancy and those who experienced chronic maltreatment had poorer inhibitory control and working memory. Dunn et al. (61) investigated the relationship between timing of childhood maltreatment and depression and suicidal ideation in early adulthood among 15,701 participants in the National Longitudinal Study of Adolescent Health, and found that exposure to early maltreatment, especially during the preschool years (between ages 3 and 5), was most strongly associated with depression. Additionally, sexual abuse occurring during early childhood, compared with adolescence, was reported to be more strongly associated with suicidal ideation (61). While these studies suggest that childhood maltreatment that occurs earlier in development may further increase risk for developing mood disorders and associated behaviors in adulthood, it is important to emphasize that evidence suggests that exposure to maltreatment during later childhood and adolescence also independently increases risk for mood disorders. Emotional abuse and neglect, especially if it occurs between ages 8 and 9, increases depressive symptoms (62). Emotional abuse during adolescence also increases risk for depression (63).

More work is emerging investigating the negative consequences of bullying. A study of 1,420 participants (ages 9–16) revealed that victims of bullying showed an increased prevalence of generalized anxiety disorder, depression, and suicide-related behavior (64). A recent study of more than 5,000 children that comprised a longitudinal data set (the Avon Longitudinal Study of Parents and Children in England and the Great Smoky Mountains Study in the United States) (65) found an increased risk for mental health problems, including anxiety, depression, and self-harm, in individuals who experienced bullying, but not other maltreatment. Additionally, an association between childhood bullying by peers and risk for suicide-related behaviors (ideation, planning, attempting, and onset of plan among ideators), independent of childhood maltreatment by adults, was reported in a sample of U.S. Army soldiers (66).

Some studies suggest that differential periods of sensitivity to different subtypes of maltreatment are distinctly associated with an increased risk for mood disorders. Recently, a stronger relationship was reported between adult depression and early childhood sexual abuse (occurring at age 5 or earlier) and later childhood physical abuse (occurring at age 13 or later), compared with maltreatment that occurred during other developmental periods (67). Harpur et al. (68) reported that early childhood maltreatment (between birth and age 4) predicted more anxiety symptoms, and maltreatment that occurred in late childhood or early adolescence (between ages 10 and 12) predicted more depressive symptoms in adolescence. Taken together, these studies suggest that maltreatment at any age and across different contexts (physical and emotional, familial- and peer-induced) often result in long-lasting and severe consequences and that there may be specific sensitive periods in development when exposure to distinct types of maltreatment may differentially increase risk for affective disorders in adulthood. To date, the majority of research investigating the impact of childhood maltreatment timing on illness risk and course in mood disorders has focused on depression. One study (69) reported that early sexual or physical abuse (before age 11) in 225 early psychosis patients (6.7% with a bipolar disorder diagnosis) coincided with lower scores on the Global Assessment of Functioning Scale and the Social and Occupational Functioning Assessment Scale during a 3-year follow-up period, whereas late sexual or physical abuse (between ages 12 and 15) did not. More work investigating timing of maltreatment and associated clinical outcomes is warranted.

Several groups have sought to determine the impact of single types of childhood maltreatment on mood disorders. Although all types of childhood maltreatment (physical, emotional, and sexual) increase disease vulnerability and risk for more severe illness course in mood disorders, including increased risk for suicide (52), there may be some distinctions between individual subtypes and associated outcomes (70). An association between sexual abuse and lifetime risk for anxiety disorders, depression, and suicide attempts independent of other types of maltreatment has been reported (2, 71, 72). In bipolar disorder, physical abuse and sexual abuse independently increase risk for illness vulnerability and more severe course (13). One study of 446 youths (ages 7 to 17) found that physical abuse was independently associated with a longer duration of illness in bipolar disorder, a greater prevalence of comorbid PTSD and psychosis, and a greater prevalence of family history of a mood disorder when compared with sexual abuse, which was only associated with a greater prevalence of PTSD (13). Recent life stress in adulthood was found to increase risk for first-onset mania in individuals with a history of childhood physical maltreatment, but not individuals with a history of sexual maltreatment (31). However, it should be noted that early-life sexual abuse in the study was a strong risk factor for mania even in the absence of recent life stress.

Neglect is the least studied form of early-life adversity, and emerging data suggest differential consequences following neglect as compared with abuse (73). Similarly, long-lasting consequences following emotional maltreatment, independently of other forms of maltreatment, have also been reported (47, 74, 75). In a 2015 meta-analysis, emotional abuse showed the strongest association with depression, followed by neglect and sexual abuse (76), a finding supported by another recent meta-analysis (77). Spertus et al. (78) reported that emotional abuse and neglect predicted depressive symptoms even after controlling for physical and sexual abuse, further suggesting emotional abuse and neglect to be independently related to illness severity in depression. Parental “verbal aggression” was found to increase risk for depression and anxiety in adolescents, with risk suggested to be greater following verbal aggression compared with physical abuse (79). Khan et al. (63) recently reported that nonverbal emotional abuse in males and peer emotional abuse in females are important predictors of lifetime history of major depression and are more predictive than number of types of maltreatment experienced. Another recent meta-analysis (12) reported that in individuals with depression, emotional neglect was the most common reported form of childhood maltreatment, and emotional abuse was most closely related to symptom severity. High prevalence of emotional maltreatment is also reported in bipolar disorder (approximately 40%), with emotional maltreatment associated with disease vulnerability and more severe illness course, including rapid cycling, comorbid anxiety or stress disorders, suicide attempts or ideation, and cannabis use (80).

Although studies on subtypes of maltreatment are only now burgeoning, they are concordant in implicating emotional maltreatment, in addition to physical and sexual maltreatment, in increasing risk for, and differences in disease course of, mood disorders. Emotional maltreatment and neglect are clearly the least studied of all forms of childhood adversity. This is in part because they are often overlooked and least likely to come to clinical attention, as compared with physical and sexual abuse, which can, of course, result in physical injury. Because emotional maltreatment and neglect are likely the most prevalent forms of childhood maltreatment in psychiatric populations (81), and given findings suggesting that independent of other forms of maltreatment, emotional maltreatment has long-lasting consequences that increase risk for mood disorders and illness outcome (74, 75), more research on the role of emotional maltreatment and neglect are urgently needed.

Although the findings described above suggest the hypothesis that different subtypes of early-life adversity may independently increase risk for mood disorders and that some subtypes may be more closely related to specific differences in illness course and severity, it is clear that subtypes of abuse and neglect, as a rule, do not occur in isolation but instead occur together in the same individuals. For example, individuals experiencing physical or sexual abuse likely also experience emotional maltreatment. Some studies have investigated the impact of multiple types of childhood maltreatment. A recent meta-analysis reported that 19% of individuals with major depression report more than one form of childhood maltreatment and, while all childhood maltreatment subtypes have been shown to increase the risk of depression, experiencing multiple forms of childhood maltreatment further elevates this risk (12). The Adverse Childhood Experiences study provided evidence of an additive effect of eight early-life stress events (including abuse but also other early-life stressors, such as divorce, domestic violence, household substance abuse, and parental loss) on adult psychopathology. Specifically, individuals with four or more early-life stress events had significantly increased risk for depression, anxiety, suicide attempts, substance use disorders, and other detrimental outcomes (82, 83). An additive or cumulative effect of early-life stress on increased risk for mood, anxiety, and substance use disorders has also been reported by others (5, 6). Multiple adverse childhood experiences (maltreatment plus other forms of stressful events) also result in higher rates of comorbidities (7, 82). Likewise, a dose-response relationship between number of types of childhood maltreatment and illness severity in bipolar disorder has been suggested, including increased risk for comorbid anxiety disorders and substance use disorders (84).

As depicted in Figure 2, several putative biological mechanisms by which childhood maltreatment may increase the risk for mood disorders and disease progression have been described (21, 85). These include, but are not limited to, inflammation and other immune system perturbations, alterations in the hypothalamic-pituitary-adrenal (HPA) axis, and genetic and epigenetic processes as well as structural and functional brain imaging changes. These studies provide insight into modifiable targets and provide direction to improve both treatment and prevention strategies.

A sizable percentage of patients with mood disorders have a history of childhood maltreatment. While the devastating consequences of childhood maltreatment cannot be disavowed, several limitations in research should be noted. Research groups often assess childhood maltreatment differently, and this can result in a measurement bias. Demographic characteristics and differences in assessments (age and sex ratio of participants; clinical versus nonclinical populations being studied; observer-rated versus self-rated depression measures) are all suggested to contribute to differences in prevalence of childhood maltreatment and relation with illness severity (12). For example, studies using the Childhood Trauma Questionnaire report higher rates of emotional abuse compared with studies using other measures to investigate childhood maltreatment (12). Further study is warranted investigating the neurobiological mechanisms, underlying genetics, familial factors, and modifiable targets that may drive development of mood disorders following childhood maltreatment. A promising area is network-based approaches to understand this link (224). Additionally, consequences following different types of maltreatment require further investigation, as different forms of childhood maltreatment may be associated with distinct neural consequences, and a better understanding of these relations is critical for the development of more effective interventions and prevention strategies. For example, Heim et al. (225) reported that victims of sexual abuse exhibit more alterations in the somatosensory area, whereas victims of emotional abuse exhibit differences in areas mediating emotional processing and self-awareness, including the anterior cingulate and parahippocampal gyrus. More work is needed to investigate whether there are sensitive periods in development when maltreatment has more robust consequences on neurobiology. Humphreys et al. (226) recently reported that hippocampal volume differences were associated with stress severity during early childhood (≤5 years of age), but there was no association between hippocampal volumes and stress occurring during later childhood. Studies investigating interactions between childhood maltreatment and genetic variation or familial risk for mood disorders could identify mechanisms underlying risk and resiliency in the absence of some study-related confounders (e.g., medication).

Longitudinal studies are critically needed to distinguish what behaviors and mechanisms (genetic and neurobiological) may contribute to risk and whether alterations in behaviors or neurobiology are secondary to mood disorder onset. It is important to emphasize that sex differences likely contribute to outcomes following childhood maltreatment (227). These include females, compared with males, having a higher risk for internalizing disorders (depression and anxiety) (228, 229), greater deficits in neural systems underlying emotional regulation (187, 230), and being more susceptible to stress-induced changes in the HPA axis (231) following maltreatment. Males, compared with females, may be more vulnerable to developing externalizing disorders (conduct disorders and substance use disorders) (232). However, few studies have investigated sex differences following childhood maltreatment. More research on sex differences is critically needed, including on the underlying neurobiology. As previously reviewed (21), early-life adversity is associated with increased vulnerability to several major medical disorders, including coronary artery disease and myocardial infarction, cerebrovascular disease and stroke, type 2 diabetes, asthma, and certain forms of cancer. More work is needed on medical morbidities that may increase risk for early mortality following early-life adversity. Additionally, more research is needed on disparities that contribute to, and minority communities that show, elevated rates of early-life adversity. As discussed above, rates of early-life adversity are higher among individuals with developmental disabilities (138–140). Rates of trauma are also higher in youths in the lesbian, gay, bisexual, transgender, and questioning (LGBTQ) community (233). Few studies have been published in this area. Youths in the LGBTQ community show higher rates of mood disorders, anxiety, suicide, and alcohol and drug use (234). In a recent study, Rhoades et al. (235) investigated the relationship between parental rejection, homelessness, and mental health outcomes in LGBTQ youths. Parental rejection was associated with higher rates of homelessness, with experience of homelessness associated with greater feelings of hopelessness, PTSD and depressive symptoms, and greater prevalence of past suicide attempts and more individuals saying they are likely to attempt suicide in the future. More work is critically needed in vulnerable populations, including work focused on interventions that may improve mental health outcomes, for example, interventions that focus on the family unit and interpersonal relationships to foster support and educational interventions, which may decrease peer victimization and cyberbullying (236, 237).

In summary, studies converge on and consistently support the finding that childhood maltreatment increases disease vulnerability for mood disorders, as well as a more pernicious disease course. A reduction in the prevalence of childhood maltreatment would have a substantial impact on decreasing disease burden (238). Studies suggesting modifiable targets are only just beginning to emerge and point to behavioral and environmental factors that could be focused on for early interventions.

Dr. Lippard’s research is supported by NIH grant K01AA027573. Dr. Nemeroff’s research is supported by NIH grants MH117293 and AA-024933.