Childhood maltreatment and violence: Mediation through psychiatric morbidity

Abstract

Childhood maltreatment is associated with multiple adverse outcomes in adulthood including poor mental health and violence. We investigated direct and indirect pathways from childhood maltreatment to adult violence perpetration and the explanatory role of psychiatric morbidity. Analyses were based on a population survey of 2,928 young men 21–34 years in Great Britain in 2011, with boost surveys of black and minority ethnic groups and lower social grades. Respondents completed questionnaires measuring psychiatric diagnoses using standardized screening instruments, including antisocial personality disorder (ASPD), drug and alcohol dependence and psychosis. Maltreatment exposures included childhood physical abuse, neglect, witnessing domestic violence and being bullied. Adult violence outcomes included: any violence, violence toward strangers and intimate partners (IPV), victim injury and minor violence. Witnessing domestic violence showed the strongest risk for adult violence (AOR 2.70, 95% CI 2.00, 3.65) through a direct pathway, with psychotic symptoms and ASPD as partial mediators. Childhood physical abuse was associated with IPV (AOR 2.33, 95% CI 1.25, 4.35), mediated by ASPD and alcohol dependence. Neglect was associated with violence toward strangers (AOR 1.73, 95% CI 1.03, 2.91), mediated by ASPD. Prevention of violence in adulthood following childhood physical abuse and neglect requires treatment interventions for associated alcohol dependence, psychosis, and ASPD. However, witnessing family violence in childhood had strongest and direct effects on the pathway to adult violence, with important implications for primary prevention. In this context, prevention strategies should prioritize and focus on early childhood exposure to violence in the family home.

Introduction

Childhood maltreatment – including physical abuse, sexual abuse, witnessing domestic violence and neglect – has a highly adverse impact on physical and mental health and is strongly associated with both internalizing and externalizing behaviors later in life (Asgeirsdottir et al., 2011, Brown et al., 2009, Gilbert et al., 2009, Goodwin and Stein, 2004, Hart and Rubia, 2012, Turner et al., 2006). Psychopathology attributed to maltreatment is often severe and extensive, and has been specifically associated with mood and anxiety disorders (Branje et al., 2009, Horwitz et al., 2001, Turner et al., 2006), personality disturbance (Johnson et al., 1999, Luntz and Widom, 1994), substance misuse (Asgeirsdottir et al., 2011, Harrison et al., 1997, Nelson et al., 2006) and psychosis in adulthood (Bebbington et al., 2011, Morgan and Fisher, 2007). It is also linked to adolescent delinquency (Stouthamer-Loeber, Loeber, Homish, & Wei, 2001) and adult criminal offending, including perpetration of violence (Fagan, 2005, Gilbert et al., 2009, Maas et al., 2008). The “cycle of violence” hypothesis postulates that children exposed to abuse and neglect will develop abusive tendencies themselves in adulthood (Widom, 1989b). Nevertheless, recent investigation was unable to confirm this hypothesis (Forsman & Langstrom, 2012). One reason is that pathways by which childhood maltreatment is linked to adult psychopathology and adult violence perpetration remain unclear. Literature reviews have also cast doubt and shows methodological inconsistencies across studies (Maas et al., 2008, Widom, 1989a). To enable better-targeted and more effective interventions for reducing the harmful impact of childhood maltreatment, it is important to establish a better understanding of whether and how maltreated children may become violent adults.

A key question is whether childhood maltreatment leads directly to later violence, for example through learning and socializing mechanisms, or whether psychiatric morbidity mediates this association. Genetic susceptibility to aggression may partly explain intergenerational transmission of violence (Caspi et al., 2002), whilst social factors may also be important. Social learning theory suggests that children learn through observing how others’ behavior is rewarded and punished – and imitate rewarded behavior (Bandura, 1965). By exposing children to repeated violence, adults model violence as an effective strategy for attaining their needs and regulating their own emotions. Children learn to view violence as a justified response, incorporate violence into their schema for responding to conflict, and readily attribute hostile intentions to others (Huesmann & Kirwil, 2007). Additionally, repeated exposure to violence can lead to desensitization whereby over time the emotional arousal triggered by exposure diminishes (Molitor and Hirsch, 1994, Paik and Comstock, 1994). This may be magnified if maltreating adults ignore or punish children's initial distressed response, encouraging a reduced emotional response to violence – whether as victim or perpetrator.

Alternatively, maltreatment may lead to violence via psychopathology. Maltreatment has damaging effects on the development of key brain regions involved in information processing, social cognition and emotion regulation (Nemeroff & Binder, 2014), predisposing children to the development of adult psychopathology. Psychological and social consequences of maltreatment which place children at risk of developing psychopathology include insecure attachment, low self-esteem, emotional dysregulation, a fractured sense of self, difficulties trusting others, peer rejection and poor educational attainment (Cichetti & Toth, 2005). Maltreatment often occurs in the context of other risk factors for mental illness, including poverty, disrupted family relationships, and parental psychopathology (Hecht & Hansen, 2001). In turn, psychopathology may in some cases increase the risk of violence. Substance misuse, psychotic disorders and antisocial personality disorder may particularly predispose individuals to violence (Coid et al., 2006, Coid et al., 2013a), either through their association with low impulse control, impaired affect regulation, paranoid ideation and narcissistic cognitions (Nestor, 2002).

Few studies have investigated the explanatory and mediating role of psychiatric morbidity on associations between early maltreatment and violence, showing mixed results. White and Widom (2003) examined mediation on intimate partner violence through antisocial personality disorder (ASPD) and alcohol misuse. They found that while ASPD completely mediated the link between child abuse and neglect and violence, there was no mediation from alcohol amongst men. Another study failed to find significant mediation, however their conclusions were likely affected because they used a general composite measure of mental health problems (Millett, Kohl, Jonson-Reid, Drake, & Petra, 2013), which lacked specificity in terms of psychopathology. Meanwhile, a more recent study conducted amongst Chinese males reported partial mediation through ASPD/BPD traits on the path between abuse in childhood and intra-familial aggression. Therefore, to date, only two studies have found that antisocial personality disorder (ASPD) fully or partially mediates the link between child abuse and neglect and later intimate partner violence among men (Liu et al., 2012, White and Widom, 2003). These studies are all limited in terms of the amount of information collected regarding psychopathology (Liu et al., 2012), and on the types of adult violence outcomes.

In the present study, we investigated a wider range of childhood exposures and psychopathology in adulthood than in previous studies to test for multiple mediators whilst controlling for coexisting psychiatric morbidity. Besides studying childhood experiences of physical, sexual abuse and neglect, we also included witnessing domestic violence and being the victim of bullying which may be associated with an elevated risk of adult violent behavior (Brockenbrough, Dewey, & Loper, 2002), but have often been omitted from existing explanatory studies. In addition to antisocial personality disorder, we also examined substance dependence and psychotic disorder as potential mediators, since these are associated both with childhood maltreatment and adult violence (Douglas et al., 2009, Pickard and Fazel, 2013), but have been overlooked in previous studies.

We aimed, firstly, to investigate whether direct associations existed between different types of early maltreatment and violence perpetration, and whether there was an incremental degree of association from exposure to multiple maltreatment types. Secondly, to identify potential clinical mediators, we investigated associations between mental health categories with childhood maltreatment and violent outcomes in adulthood. Finally, we quantified the magnitude of any explanatory role we identified in these associations.