Elsevier

Aggression and Violent Behavior

Volume 24, September–October 2015, Pages 261-272
Aggression and Violent Behavior

Longitudinal predictors of domestic violence perpetration and victimization: A systematic review

https://doi.org/10.1016/j.avb.2015.06.001Get rights and content

Highlights

  • We review prospective longitudinal studies of child and adolescent predictors of DV.

  • We identify five common predictors of DV perpetration and victimization for men and women.

  • Most consistent predictors of DV are child abuse and family of origin risks.

  • Additional predictors are child/adolescent behavior problems and sociodemographic risk.

  • Policy, intervention and research need to adopt a whole of life approach to addressing DV.

Abstract

Domestic violence (DV) is a serious and complex social issue which is associated with significant costs to both those individuals who are directly affected and the wider community. Preventative approaches with vulnerable population groups represent an important component of any integrated response to DV and should be informed by an understanding of those factors that influence violence developmentally. This paper reports the findings of a systematic review of longitudinal studies that have prospectively investigated childhood and/or adolescent predictors of DV perpetration and/or victimization among adult men and women in intimate relationships. We identified 25 original studies that met the inclusion criteria, all of which investigated predictors of domestic physical abuse. Few studies prospectively examined psychological, sexual and verbal abuse. Child and adolescent abuse, family of origin risks, child and adolescent behavioral problems, adolescent peer risks, and sociodemographic risks were all identified as significant predictors of DV perpetration and victimization. It is concluded that early childhood and adolescent factors are consistent predictors in the development of DV perpetration and victimization and that prevention and early intervention approaches targeting these factors are likely to prove the most effective.

Introduction

Domestic violence (DV) is a serious and complex social issue which is associated with significant health, economic and social costs to individuals and the wider community (Mitchell, 2011). DV is a term commonly used to refer to violent acts between adult intimate partners, and can include physical, sexual, emotional and/or psychological abuse (Mitchell, 2011). A significant body of DV literature has amassed over the past 30 years, and yet there have been few attempts to review the body of literature that has investigated developmental influences on DV using longitudinal research designs. Longitudinal studies provide a key method for identifying the risk and protective factors that predict DV perpetration and victimization by clarifying the temporal ordering and strength of potentially modifiable influences. Prospective longitudinal studies across different periods in the life-course can identify key developmental patterns and influences and in this way inform opportunities for DV prevention through the life-course (Hemphill et al., 2009).

A clear understanding of the patterns of DV and predictors of perpetration and victimization among both men and women is critical to inform preventative strategies (Hemphill et al., 2009, McCambridge et al., 2011). A number of longitudinal studies have now been reported internationally documenting risk and protective factors for DV across childhood, adolescence and adulthood in a range of populations. These studies enable the onset of DV initiation to be identified and linked to prior developmental experiences. Such longitudinal data enables the temporal ordering of developmental antecedents to be separated from the onset of DV incidents, a precondition for causal relationships to be examined. Studies that rely on retrospective reports of childhood adversity make it difficult to draw valid conclusions regarding early life influences of DV experienced in adulthood owing to the well-recognized biases and fallibility of human memory. For example previous research indicates that recall of child abuse results in a substantial rate of false negatives, measurement error, and bias that could elevate type I errors, with higher well-being linked to retrospective forgetting and lower well-being tied to greater retrospective reporting (Hardt & Rutter, 2004). Prospective reports avoid some of the validity threats associated with retrospective reporting, especially when studying extended periods of the life-course (Schwartz and Sudman, 1994, Stone et al., 2007, Tourangeau et al., 2000). To the authors' knowledge, there has been no prior systematic review of prospective longitudinal studies which allow the researcher to assess developmental trends and establish the sequencing and patterns of life-course influences and behaviors that contribute to DV in adulthood. Therefore, the authors undertook a systematic review of prospective longitudinal studies of childhood and adolescent predictors of DV perpetration and victimization among men and women and evaluated the strength of this evidence.

Previous reviews have investigated DV risk factors across childhood, adolescence, and adulthood from cross-sectional and longitudinal studies (e.g., Capaldi et al., 2012, Cattaneo and Goodman, 2005, Schumacher et al., 2001 Stith, Smith, Penn, Ward, & Tritt, 2004). For example, Schumacher et al. (2001) reviewed 72 studies of male perpetrated physical abuse and found moderate to strong risk factors (rs  0.30) included low SES, low education level, experience of child abuse, witnessing parental violence in childhood, anger/hostility, depression, and alcohol and other drug abuse. Similarly, in a meta-analysis of 85 studies of male perpetrated domestic physical abuse Stith et al. (2004) found moderate to strong risk factors for male perpetration (rs  0.20) included a history of physical abuse perpetration (r = 0.24), marital dissatisfaction (r = 0.30), alcohol abuse (r = 0.24), drug use (r = 0.31), a violence-condoning attitude (r = 0.30), traditional sex role ideology (r = 0.30), anger/hostility (r = 0.26) and depression (r = 0.23). Three risk or co-occurring factors for female victimization were identified: female violence toward male partners (r = 0.41), depression (r = 0.28) and fear of partner violence (r = 0.27), although these factors are likely to be consequences of previous victimization experience. A significant limitation of Schumacher et al.'s and Stith et al.'s studies was failure to report the design of included studies. Further, the childhood risk factors (experience of child abuse, witnessing parental violence) in Schumacher et al.'s review appear to be based on retrospective recall.

A subsequent critical review of 62 studies by Cattaneo and Goodman (2005) reported that the length of time partners had lived together and a history of abuse within the relationship predicted rates of ongoing physical abuse perpetration across samples of men facing criminal or civil action or in treatment for DV. This review focused on patterns of DV reabuse and did not consider developmental risks of DV.

Capaldi et al. (2012) recently undertook a systematic review of over 200 studies of DV among adults and adolescents and found several demographic and contextual risk factors, including low SES, younger age, minority group membership, stress, types of friends and social support, drug use, and relationship conflict and satisfaction. They also described a small number of more recent prospective studies that spanned from childhood into adulthood. These studies indicated a varied set of potential risk factors, such as experience of child abuse, witnessing parental violence, association with aggressive peers in adolescence, and early antisocial behavior. Capaldi et al. (2012) noted that the more recent emergence of studies with longitudinal designs opened opportunities to examine developmental hypotheses.

None of the prior reviews approached DV specifically from a developmental perspective or focused on prospective longitudinal evidence. Therefore, in the current review, the authors take a life-course approach to identify risk factors (stable longitudinal predictors of outcomes) assessed prospectively in childhood and adolescence that contribute to subsequent DV experienced in adulthood.

Different types of violence are more likely to be captured by different study designs (Graham-Kevan and Archer, 2003, Johnson, 2005, Johnson, 2006, Johnson and Leone, 2005, Kelly and Johnson, 2008). For example, the term “intimate terrorism” has been used to capture situations in which a perpetrator is both violent and controlling (i.e., where the violence is one part of a general strategy of power and control). In contrast, “situational couple violence” is a term used to describe situations that involve occasional outbursts of violence, but where violence is not a part of a general pattern of control (Johnson, 2005, Johnson, 2006). These two types of domestic violence are considered to be qualitatively different, with characteristics that make each one more accessible by specific sampling methods. For example, samples drawn from law enforcement agencies, hospitals, or shelters successfully capture intimate terrorism, largely perpetrated by men (Biroscak et al., 2006, Johnson, 2005, Johnson, 2006), whereas situational couple violence is more commonly identified in large-scale surveys of the general population and generally show some gender symmetry in perpetration (Johnson, 2005, Johnson, 2006, Kimmel, 2002). While general population studies may claim random or representative samples, in this instance the bias of even minor non-response has critical implications. For example, intimate terrorists may be extremely unlikely to agree to participate in such a survey, and their partners are often isolated and likely to fear reprisals if they answer such questions (Johnson, 2005). However, because victims of intimate terrorism are attacked more frequently, are more likely to be injured, and are more likely to seek help from police and medical facilities (Johnson & Leone, 2005), it makes sense that samples recruited from law enforcement agencies, emergency rooms, and shelters tap this difficult to reach population. For the purposes of the present review, we acknowledge that retaining individuals who experience intimate terrorism in longitudinal studies is difficult and hence such studies likely underestimate these groups.

Section snippets

Method

A systematic review of the scientific literature was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher, Schulz, Altman, & Group, for the CONSORT, 2001) to identify relevant prospective longitudinal studies investigating child and adolescent predictors of DV perpetration and/or victimization experienced in the context of adult romantic relationships. Inclusion and exclusion criteria of included papers were discussed and

Results

A summary of measures of DV predictors and outcomes and significant findings of each study is presented in Table 2.

Discussion

A careful search for prospective longitudinal studies examining child and adolescent predictors of adult DV perpetration and victimization using specified inclusion criteria identified only 25 studies. Across these studies, five domains were consistently identified as predictive of both perpetration and victimization. Child and adolescent abuse experiences and family of origin risks were the most commonly investigated and consistent predictors, followed by child and adolescent behavioral and

Conclusions

The selection criteria identified only 25 prospective longitudinal studies examining childhood and adolescent developmental predictors of DV perpetration and victimization. Such studies depended heavily on the CTS and on population-based surveys. Significant predictors of DV captured in these studies included child abuse, family of origin risks, child and adolescent behavior problems, adolescent substance use, adolescent peer risks and, less consistently, sociodemographic risk. These findings

References (67)

  • R. Braaf et al.

    The gender debate in domestic violence: The role of data

    Issues Paper 25

    (2013)
  • D.M. *Capaldi et al.

    Aggression toward female partners by at-risk young men: The contribution of male adolescent friendships

    Developmental Psychology

    (2001)
  • D.M. Capaldi et al.

    A systematic review of risk factors for intimate partner violence

    Partner Abuse

    (2012)
  • L.B. Cattaneo et al.

    Risk factors for reabuse in intimate partner violence: A cross-disciplinary critical review

    Trauma, Violence & Abuse

    (2005)
  • M. *Cui et al.

    The continuation of intimate partner violence from adolescence to young adulthood

    Journal of Marriage and the Family

    (2013)
  • C.B. Cunradi et al.

    Alcohol-related problems, drug use, and male intimate partner violence severity among US couples

    Alcoholism: Clinical and Experimental Research

    (2002)
  • R.P. Dobash et al.

    The myth of sexual symmetry in marital violence

    Social Problems

    (1992)
  • M.K. *Ehrensaft et al.

    Clinically abusive relationships in an unselected birth cohort: Men's and women's participation and developmental antecedents

    Journal of Abnormal Psychology

    (2004)
  • D. *Fergusson et al.

    Developmental antecedents of interpartner violence in a New Zealand birth cohort

    Journal of Family Violence

    (2008)
  • A.M. *Gomez

    Testing the cycle of violence hypothesis: Child abuse and adolescent dating violence as predictors of intimate partner violence in young adulthood

    Youth & Society

    (2011)
  • Graham-Kevan et al.

    Intimate terrorism and common couple violence: A test of Johnson's predictions in four British samples

    Journal of Interpersonal Violence

    (2003)
  • C.T. Halpern et al.

    Sexuality in emerging adulthood

  • J. Hardt et al.

    Validity of adult retrospective reports of adverse childhood experiences: Review of the evidence

    Journal of Child Psychology and Psychiatry

    (2004)
  • J.D. Hawkins et al.

    Youth problem behaviors 8 years after implementing the Communities that Care Prevention System: A community-randomized trial

    Journal of the American Medical Association Pediatrics

    (2014)
  • S.A. Hemphill et al.

    Modifiable determinants of youth violence in Australia and the United States: A longitudinal study

    Australia and New Zealand Journal of Criminology

    (2009)
  • T.I. *Herrenkohl et al.

    Youth violence trajectories and proximal characteristics of intimate partner violence

    Violence and Victims

    (2007)
  • T. *Ireland et al.

    Living in partner-violent families: Developmental links to antisocial behavior and relationship violence

    Journal of Youth & Adolescence

    (2009)
  • M.P. Johnson

    Apples and oranges in child custody disputes: Intimate partner terrorism vs. situational couple violence

    Journal of Child Custody

    (2005)
  • M.P. Johnson

    Conflict and control: Gender symmetry and asymmetry in domestic violence

    Violence Against Women

    (2006)
  • M.P. Johnson et al.

    Effects and non-effects of childhood experiences of family violence on adult partner violence

  • W.L. *Johnson et al.

    The age–IPV curve: Changes in the perpetration of intimate partner violence during adolescence and young adulthood

    Journal of Youth and Adolescence

    (2015)
  • M.P. Johnson et al.

    The differential effects of intimate terrorism and situational couple violence: Findings from the National Violence Against Women Survey

    Journal of Family Issues

    (2005)
  • J.B. Kelly et al.

    Differentiation among types of intimate partner violence: Research update and implications for interventions

    Family Court Review

    (2008)
  • Cited by (111)

    View all citing articles on Scopus
    View full text