Does adolescent's exposure to parental intimate partner conflict and violence predict psychological distress and substance use in young adulthood? A longitudinal study

Abstract

Little is known about the extent to which parental conflict and violence differentially impact on offspring mental health and substance use. Using data from a longitudinal birth cohort study this paper examines: whether offspring exposure to parental intimate partner violence (involving physical violence which may include conflicts and/or disagreements) or parental intimate partner conflict (conflicting interactions and disagreements only) are associated with offspring depression, anxiety and substance use in early adulthood (at age 21); and whether these associations are independent of maternal background, depression and anxiety and substance use. Data (n = 2,126 women and children) were taken from a large-scale Australian birth-cohort study, the Mater University of Queensland Study of Pregnancy (MUSP). IPC and IPV were measured at the 14-year follow-up. Offspring mental health outcomes – depression, anxiety and substance use were assessed at the 21-year follow-up using the Composite International Diagnostic Interview (CIDI). Offspring of women experiencing IPV at the 14-year follow-up were more likely to manifest anxiety, nicotine, alcohol and cannabis disorders by the 21-year follow-up. These associations remained after adjustment for maternal anxiety, depression, and other potential confounders. Unlike males who experience anxiety disorders after exposure to IPV, females experience depressive and alcohol use disorders. IPV predicts offspring increased levels of substance abuse and dependence in young adulthood. Gender differences suggest differential impact.

Introduction

Does adolescent's exposure to parental intimate partner conflict and violence predict psychological distress and substance use in young adulthood? A longitudinal study on parental intimate partner conflict (IPC) and violence (IPV) is not uncommon. UNICEF has estimated that between 133 million and 275 million children around the world witness frequent parental intimate partner conflict and or violence each year (Pinheiro, 2006). IPC generally involves arguments and/or disagreements with verbal expressions which contribute visible evidence of these disagreements to offspring. Routine communications and interactions are generally affected, for example one partner may leave the home for a period of time. IPV, by contrast, involves observable physical violence such as one partner striking the other. Such violence is often accompanied by interactions which include conflict and disagreements. A study from a nationally representative sample in the United States estimated that about 1 in 4 women and 1 in 7 men had experienced IPV in their lifetime (Breiding, Black, & Ryan, 2008). An earlier Australian population study found that 23% of women had ever experienced IPV (Mulroney, 2003), with perhaps 60% of these women having children in their care, and that some 66% of these children had witnessed the violence (Linacre, 2007). Although a recent study conducted among convenience samples of university students in 15 nations found that males and females were almost equally likely to perpetrate intimate partner violence, especially in North America (Straus & Michel-Smith, 2013), other studies suggest that the prevalence of male perpetrated IPV is much higher and more severe than the prevalence and severity of female perpetrated violence (Feder and MacMillan, 2012, Halford et al., 2010, MacMillan and Wathen, 2014, Tjaden and Thoennes, 2000). Our study focuses on the consequences for offspring of parental IPV against the mother using maternal reports of male-partner violence toward them.

In the context of increasing concern about the consequences of parental IPV on offspring (Paterson, Taylor, Schluter, & Iusitini, 2013), a number of questions remain to be resolved. Firstly do IPV and non-violent Inter-partner Conflict (IPC) have similar consequences for offspring outcomes? Secondly, to what extent is the association between IPV, IPC and offspring mental health a result of confounding? A variety of social and demographic factors may confound the association between IPV, IPC and offspring mental health and behavior, including teenage pregnancy, poverty/socioeconomic disadvantage, maternal mental health and alcohol use by caregivers (Fergusson et al., 2005, Gilbert et al., 2012). Finally, relatively few previous studies have been able to resolve alternative cause–effect sequences. Thus, it is plausible that early onset of child mental health and behavior problems may lead to IPV or IPC rather than the reverse being the case. There is a need to test a causal sequence using a prospective cohort design (Kitzmann, Gaylord, Holt, & Kenny, 2003).

It is not known whether there are critical or sensitive periods for the child's exposure to IPV. Child behavior manifests marked changes particularly around the period of pubertal development (Najman et al., 2008) and post pubertal offspring health may be affected by the witnessing of marital conflict and/or violence during this stage of development. Focusing on young adult mental health and behavior outcomes in a causal model is also likely to provide a better prediction of adult life course trajectories than assessing pre-pubertal outcomes. In addition, large-scale population-based prospective studies using diagnostic measures of outcome are rare (Fergusson, Boden, & Horwood, 2008) and no previous such studies have addressed the question of whether adolescent exposure to IPV predicts young adult outcomes. The current study will compare the impact of offspring exposure to parental non-violent intimate partner conflict (IPC) and intimate partner violence (IPV) in adolescence (child age 14 years); as these predict offspring mental health and substance use at 21 years of age using DSM-IV measures of outcome.

Growing-up in a home characterized by ongoing non-violent conflict, whether or not it is accompanied by divorce, is associated with greater psychological distress and lower levels of well-being in young adulthood (Amato and Sobolewski, 2001, Turner and Kopiec, 2006). These outcomes have been largely reported in cross sectional designs (Kim, Jackson, Conrad, & Hunter, 2008). We have been unable to find specific studies assessing the possibility of differential outcomes when the child is exposed to IPV compared to children experiencing IPC.

IPV is arguably a severe form of IPC (Martinez-Torteya, Bogat, von Eye, & Levendosky, 2009). There is some evidence to suggest that IPV impacts on offspring may be manifested in aggression, substance use, emotional withdrawal, attention problems, and psychiatric symptoms (Harris et al., 2007, Nayak et al., 2012, Turner and Kopiec, 2006, van der Kolk, 2005). Moreover, these negative consequences may be carried into offspring adulthood and affect the offspring's broader social environment (Harris et al., 2007, Lieberman et al., 2011).

There is a dearth of longitudinal studies of the effects of IPV on adolescent offspring. Among longitudinal studies that were conducted on adolescents – one longitudinal U.S. national survey among adolescents ages 12–17 (N = 3,614 at wave 1) examined whether exposure to IPV at wave 1 was associated with posttraumatic symptoms, delinquency, depression and binge drinking in wave 2 and 3 conducted about 1 and 2 years later respectively. Exposure to IPV at wave 1 was associated with offspring depression, delinquent acts and binge drinking at wave 3 (Cisler et al., 2012). Similarly, another study conducted by the same research group using the latter dataset found associations between children experiencing physical abuse, sexual assault, witnessing inter-parental conflict or community violence at wave 1 and substance use at wave 2 some 15 months later (Begle et al., 2011). However, the analyses aggregated all types of victimization together. Despite these findings it is generally the case that there are few studies using longitudinal design assessing the impact of IPV on adolescents, and to our best knowledge, none that have followed adolescents into young adulthood. In addition, while most studies which have examined the effect of IPV on children and adolescents used self-report or mother report scales (usually Child Behavior Check-list; CBCL, see for example, Howell, 2011, Lamers-Winkelman et al., 2012), the present study uses structured clinical interviews based on DSM criteria to assess the young adults depression, anxiety and substance use.

The impact of parental IPV and IPC on offspring may differ by gender. While many studies have controlled for gender, very few studies have examined gender differences directly. Studies that have assessed gender differences do not report on gender differences or have not found significant differences (Kitzmann et al., 2003, Lamers-Winkelman et al., 2012, Wolfe et al., 2003). Two recent studies have examined gender differences in the context of child exposure to family violence. One was based on longitudinal data from the Project on Human Development in Chicago Neighborhoods including 1,421 adolescents aged 16–22 (wave III of this study) who were 10–16 years when family conflict was assessed (wave I) (Skeer et al., 2011). The main dependent variable was substance use and dependence. For males, living in families with high levels of conflict was not a risk factor for either substance use or symptoms of anxiety or depression while for females, living in families with high conflict was associated with conduct problems and substance use and dependence (Skeer et al., 2011). In contrast, a U.S. nationally representative sample of 3,614 adolescents aged 12–17 years which examined gender differences in outcomes following the witnessing of IPV found males were more affected than females (Begle et al., 2011). There remains a need to explore gender differences in offspring outcomes of IPC and IPV directly.

Exposure to domestic violence is associated with an increased rate of symptoms of anxiety and depression for the victim of violence (Fergusson et al., 2005, Gao et al., 2010, Lindhorst and Beadnell, 2011, Straus and Mickey, 2012). The severity of the negative consequences of IPV on children may depend upon the extent to which the mother shows symptoms of depression and anxiety and whether she succeeds in being supportive to her child (Howell, 2011, Renner and Boel-Studt, 2013). For example, Holmes recent study using secondary data analysis from the National Survey of Child and Adolescent Well-Being (NSCAW) found that inter-parental violence was associated with maternal poor mental health, i.e., major depressive episodes, heavy alcohol use, and substance use in the past year, which in turn was significantly associated with child's (ages 3–8 years) aggressive behavior (Holmes, 2013). Regardless of IPC or IPV, prenatal smoking is associated with offspring externalizing and internalizing problems at childhood and adolescence even after controlling for numerous covariates (Ashford, van Lier, Timmermans, Cuijpers, & Koot, 2008). Thus, while there might be a direct association between offspring exposure to parental IPV and/or IPC and their emotional and behavioral distress (Yoo & Huang, 2012), the impact of IPV or IPC on offspring may be in part mediated by maternal distress (Renner & Boel-Studt, 2013), prenatal smoking (Wakschlag, Pickett, Kasza, & Loeber, 2006), maternal substance use (Skeer et al., 2009, Skeer et al., 2011), family structure (single versus intact families) and low income (Turner, Finkelhor, Hamby, & Shattuck, 2013). These potential confounders should therefore be controlled when examining the effects of experiencing IPC or IPV in adolescence, on the offspring psychological and SUDs in early adulthood.

In summary, little is known about whether IPV (mainly domestic violence) and IPC have similar consequences on offspring mental health problems and substance (licit and illicit) using behaviors. No previous studies have examined the possibility that the consequences of IPV may differ from those associated with IPC. Much of the available research is cross sectional rather than longitudinal and does not adequately control for a range of potential confounders. Finally there is a need to determine whether there are gender differences in the impact of IPC and IPV on offspring.

• Is offspring exposure to parental IPV or IPC in adolescence (age 14) associated with offspring depression, anxiety and substance use in early adulthood (at age 21)?

• Are associations between offspring exposure to parental IPV or IPC in adolescence and subsequent anxiety, depression and substance use independent of maternal background, depression and anxiety and substance use?

• Are the consequences of adolescent exposure to parental IPV greater than the effects of exposure to IPC?

• Are the associations between offspring exposure to IPV or IPC in adolescence and subsequent anxiety, depression or substance use in early adulthood differ by gender?