Intimate Partner Violence Across the
Lifespan: Dyadic Theory and Risk and
Deborah M. Capaldi, Sabina Low, Stacey S. Tiberio, and
Joann Wu Shortt
Introduction ....................................................................................... 2
Theoretical Approaches ........................................................................... 3
Typological Approaches . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . 4
Dyadic Theoretical Approaches . . . . . ......................................................... 4
Theory Related to the Association Between Age and Aggression .. ........................ 5
Reviews of Risk Factors . ......................................................................... 6
Early Reviews on Risk Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Reviews of Risk Factors in the Last Decade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Risk Factor Summary from Capaldi et al. (2012) ............................... ............. 9
Demographic Risk Factors and the Role of Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Exposure to Interparent IPV and Experience of Maltreatment in Childhood ................ 9
Protective Family Factors ..................................................................... 10
Social Risk Factors . ........................................................................... 10
Behavioral Risk Factors ....................................................................... 11
Impulsivity, Anger, Hostility, and Traumatic Brain Injury ................................... 11
Substance Use . ................................................................................ 13
Relationship Factors .......................................................................... 13
Gender Issues .................................................................................. 14
Other Recent Reviews . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . 14
Cumulative and Compounded Risk Factors . . . ............................................... 15
Assortative Partnering ......................................................................... 16
Do Risk Factors Differ by Severity of Violence? . . . . ........................................ 16
Recent Work Testing the DDS Model ............................................................ 19
Key Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
D. M. Capaldi (*) · S. S. Tiberio · J. W. Shortt
Oregon Social Learning Center, Eugene, OR, USA
T. Denny Sanford School of Social and Family Dynamics, Arizona State University, Phoenix, AZ,
© The Author(s) 2019
R. Geffner et al. (eds.), Handbook of Interpersonal Violence Across the Lifespan,
Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Implications for Future Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Prevention Implications . . ..................................................................... 21
Cross-References . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 22
References . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
In this chapter, theory regarding the development of intimate partner violence
(IPV) and risk and protective factors for involvement in IPV are reviewed.
In deﬁning IPV, both perpetration of and victimization by acts of psychological,
physical, and sexual aggression are considered, along with injuries, which are a
key indicator of physical IPV. Organized within the levels of an ecological or
dynamic developmental systems model, risk and protective factors are considered
within the domains of (a) contextual characteristics of partners (demographic,
neighborhood, community, and school factors), (b) developmental characteristics
and behaviors of the partners (e.g., family, peer, psychological/behavioral, and
cognitive factors), and (c) relationship inﬂuences and interactional patterns. Find-
ings of a prior systematic review of risk factors (Capaldi DM, Knoble NB, Shortt
JW, Kim HK, Partner Abuse 3:231–280, 2012) are summarized and extended by
considering ﬁndings of recent reviews and empirical studies. Recommendations
for prevention and intervention based on the review ﬁndings are presented.
Intimate partner violence · Risk factors · Adolescent · Young adults ·
Developmental · Theoretical approaches
This chapter concerns childhood, adolescent, and early adult risk factors for intimate
partner violence (IPV) involvement as a perpetrator and/or victim in both adoles-
cence and adulthood. Understanding the role of risk factors at different develop-
mental stages is essential for designing evidence-based and developmentally
informed prevention programs and interventions. During the last few decades, the
ﬁeld has undergone signiﬁcant changes in the perceptions and prevailing paradigms
surrounding IPV; mainly, there has been a departure from the belief that IPV occurs
exclusively within marriages and involves only female victims who were trapped in
volatile marriages primarily due to ﬁnancial dependence and children (Gelles and
Straus 1979). Rather, Stets and Straus (1989) found that IPV began much earlier,
often within adolescent dating relationships, ushering in research that would chal-
lenge and revise models of risk and protective factors. In this chapter, theoretical
approaches to the development of IPV are discussed, and an integrative develop-
mental model of risk factors is presented. Prior reviews of risk factors are then
described chronologically, with particular focus on the extensive systematic review
2 D. M. Capaldi et al.
conducted by Capaldi et al. (2012) and the more recent review by Costa et al. (2015).
Recent ﬁndings from mediational or indirect effects models are then discussed. The
question of whether risk factors may differ between more moderate and severe IPV is
also considered. Finally, the chapter concludes with key points, conclusions, impli-
cations for future research, and prevention.
Corvo and Johnson (2013) discuss and evaluate some major current theories of
individual risk for committing IPV using well-known criteria for theoretical value –
namely, that the theory must generate hypotheses that can be tested empirically, that
it can predict outcomes of interest (i.e., explain variation in those outcomes), and that
it produces a goodness of ﬁt with known data. They add to these criteria that simpler
or more parsimonious theories are preferred. They consider that current major
theoretical views of IPV perpetration can be summarized as, ﬁrst, feminist/sociocul-
tural, where IPV is viewed solely as the result of patriarchy involving a more
powerful position of men versus women in society. This viewpoint underlies most
batterer treatment programs. Second, social learning theory highlights the impor-
tance of modeling, where IPV occurs as a result of intergenerational transmission
related either to witnessing parental IPV during childhood or being maltreated
directly by parents. Third, psychological/psychosocial explanations suggest that
IPV results from psychopathology symptoms and disorder and includes risk factors
such as impulsivity, antisocial behavior, depression, and substance use and abuse.
Corvo and Johnson (2013) evaluate each of these theoretical suppositions and
present compelling evidence that the patriarchy explanation is difﬁcult to test
according to the criteria for evaluating theories. Yet, due in part to its universal
nature, it has been perpetuated among those who feel strongly regarding the concept
of patriarchy and the position of women in society. As stated by Corvo and Johnson
(2013), “Although most national policies and the ‘batterer’treatment standards of
most states are premised upon domestic violence being the product of ‘patriarchy’,
the central causal construct in feminist/sociocultural theory, there is little empirical
evidence in support of this view”(p. 176). Second, regarding social learning studies
of intergenerational associations, the main issues raised by Corvo and Johnson are
the modest effect sizes obtained in the prediction of IPV perpetration and, thus, that
those who witness violence in the family of origin or experience it directed toward
them in childhood may or may not perpetrate IPV as adults. Therefore, knowledge is
needed of mediating, or more proximal, risk factors. Corvo and Johnson argue that
the psychological/psychosocial risk factors show stronger associations with IPV
perpetration, are more proximal, and provide a simpler or more parsimonious
theoretical explanation for perpetration of IPV than other explanations. From their
viewpoint, IPV is a maladaptive and primitive coping strategy, and treatment pro-
grams should focus on reducing these proximal risk factors.
Although there is much strength to these arguments, focusing on only the
proximal risk factors has two major drawbacks. First, while parsimonious, it is
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 3
incomplete from a developmental perspective, because it ignores the etiology of the
psychopathology. Second, ignoring development is problematic from a prevention
standpoint, as prevention programs prior to the onset of IPV are necessary.
Another major focus of studies has involved typological approaches to understand-
ing IPV. These are not theories per se, but may be thought of as a combination of
empirical approaches to determining variance-based groupings, similar to cluster
analyses, with some theoretical grounding related to the variables chosen to estimate
the groupings. They are based on the premise that unlike feminist theory, which
posits that IPV perpetrators are a homogeneous group, there is in fact heterogeneity
among IPV perpetrators and that understanding such heterogeneity can assist in
determining differential treatment approaches. As described by Capaldi et al. (2018),
advantages of typological approaches include simpliﬁcation of relatively complex
data into archetypical characterizations. Early typologies were based on aspects
of male perpetrator’s behavior –such as generally violent/antisocial, dysphoric/
borderline personality, and family-only violence (Holtzworth-Munroe and Stuart
1994)–or on motivations for IPV, such as patriarchal or intimate terrorism versus
common couple violence (Johnson 1995). A number of typological approaches have
been proposed (for a review, see Ali et al. 2016).
Capaldi and Kim (2007) posited that there are a number of weaknesses to
typological approaches. First, typological approaches are essentially static
approaches, which group the individuals on the basis of behavior at one point in
time; thus, they do not encompass the dynamic nature of partner violence. For
example, a follow-up study by Holtzworth-Munroe et al. (2003) found some evi-
dence for the stability of partner violence groupings. However, they also found
considerable evidence of change in the violent behaviors of the men in the different
groupings, rendering the types less distinguishable across time. A second critical
issue is that there is evidence that IPV is best characterized along a continuum from
low levels to severe, with the risk characteristics of those perpetrating such violence
also tending to range from low to severe rather than being related to qualitatively
different types. Finally, most typologies have been focused on characteristics and
motivations of the individual and do not encompass the fact that the behavior
involves an interaction between the partners in a romantic relationship, therefore,
neglecting the dyadic nature of the behavior.
Dyadic Theoretical Approaches
The mounting evidence of a high prevalence of bidirectional IPV has led to a
recognition of the need for theories based on the behavior of both partners in the
relationship or dyadic behavior. Langhinrichsen-Rohling et al. (2012) reviewed rates
of bidirectional versus unidirectional IPV across samples, sexual orientations, and
4 D. M. Capaldi et al.
race/ethnicities. Bidirectional violence was common across all types of samples
(population-based to criminal justice), at an average of 58% of IPV-involved cou-
ples. They concluded that the role of women in violent relationships is important to
consider, even if all aspects of women’s perpetration of IPV are not symmetrical
to men’s perpetration of IPV. Riggs and O’Leary (1989) and colleagues were early
proponents of theoretical approaches that included a focus on both partners; they
examined the role of partner factors, with a focus on predicting conﬂict and IPV,
among dating and engaged couples. The importance of a dyadic focus was empha-
sized by ﬁndings of their study of physical IPV among high school couples (O’Leary
and Slep 2003). They found signiﬁcant stability over a 3-month time period in IPV
perpetration for both the boys and girls. Moreover, in a model including both
partners’behaviors, they found that one partner’s physical IPV was predictive
of the other partner’s physical IPV at the later time point.
Findings from such studies and from the work of Capaldi and colleagues led to
the development of a dynamic developmental systems (DDS) approach to under-
standing IPV (Capaldi et al. 2012). This approach emphasizes developmental risks
and the bidirectional, as well as the changing, nature of IPV over time. Thus, IPV is
dynamic both in relation to occurrence during conﬂict episodes involving the
interaction of two people and over time (e.g., varying with age, stage of relation-
ship). The model also builds on prior work regarding ecological systems
(Bronfenbrenner 1995) and lifespan development (Caspi and Elder 1988). The
DDS model, focusing on the developmental risk pathways (rather than on aspects
of change in IPV over time), is summarized in Fig. 1. The model also encompasses
physiological systems which is one aspect of individual vulnerabilities (Kim et al.
Theory Related to the Association Between Age and Aggression
Age is related to the perpetration of aggression and impulsive behavior, in general,
including IPV. Impulsive behavior peaks in adolescence or young adulthood, with
declines starting at least by the late 20s (Kim et al. 2008). Similarly, delinquency and
crime involving physical aggression as well as other forms of societal rule-breaking
behavior (e.g., theft, vandalism) also decrease in the 20s (Blumstein et al. 1986). As
discussed by Capaldi et al. (2018), dual systems theory (Casey et al. 2008; Steinberg
2010) is a recent theory related to brain development focusing on the socioemotional
system regarding this age-related change. The theory posits that a surge in reward
seeking and risk taking in adolescence is inﬂuenced by a sharp increase in dopami-
nergic activity in the limbic and paralimbic areas of the brain. However, the increase
in reward seeking occurs prior to the structural maturation of the cognitive control
system (mainly involving the lateral prefrontal and parietal cortices and parts of the
anterior cingulate cortex to which they are connected) and its connections to
the socioemotional areas. Adolescence and early adulthood are therefore vulnerable
periods for poorly controlled arousal, including anger and aggression, and for
impulsive behaviors. Anecdotally, this theory provides a good ﬁt to the observed
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 5
association seen during problem-solving discussions of young couples in the OYS-
Couples Study where arousal and excitement could quickly turn to conﬂict and
physical aggression. This theory is essentially the only one that can adequately
account for the very sharp peak of risk-taking behaviors in later adolescence,
followed by a decline across the adult years; although, other factors such as less
association with delinquent peers might also contribute to this reduction.
A number of models have been tested based on the DDS approach, which support
the contribution of family of origin, prior developmental risk factors, and concurrent
contextual factors to partner violence involvement (e.g., Capaldi et al. 2001; Kim
and Capaldi 2004; Shortt et al. 2013), as well as declines in physical IPV (although
less so for psychological IPV) with age (Kim et al. 2008).
Reviews of Risk Factors
Early Reviews on Risk Factors
Turning to empirical ﬁndings regarding associations of risk factors with IPV, an early
review by Schumacher et al. (2001a) exclusively focused on risk factors for male-to-
female partner physical abuse. In summary, they found that perpetrator risk factors
with moderate to strong effect sizes included socioeconomic status (SES), education,
history of sexual victimization in childhood, exposure to parental physical and/or
verbal aggression, exposure to adults exhibiting violence in childhood, nonfamily
aggression by parent, elevated levels of state and trait anger and hostility, various
personality disorders, various types of Axis I psychopathology (including depres-
sion, alcohol, and drug abuse), deﬁcits in assertiveness, and attitudes condoning
men’s physical aggression toward women. There are several notable features of the
Schumacher et al. (2001a) review. First, numerous studies at that time focused on
risk from maltreatment experiences in childhood and witnessing violence between
parents. These two forms of violence experiences within the family of origin have
Fig. 1 Theoretical developmental model of intimate partner violence
6 D. M. Capaldi et al.
long been hypothesized to be inﬂuential on later IPV involvement, based on a social
learning theory framework. They served as a catalyst for studies on intergenerational
transmission of violence (Capaldi et al. 2012). Second, other than these factors and
background demographic factors, other risk factors were overwhelmingly assessed
during adulthood (e.g., personality and psychopathology); thus, there was a rather
long developmental gap between the childhood risk factors and more proximal risk
factors. However, it was also the case that the family-of-origin risk factors were
predominantly assessed by retrospective reports.
Since the Schumacher et al. (2001a) study, there has been an increase in studies
with a longitudinal developmental design that began in childhood or adolescence.
This has led to an increase in examination of developmental history, including
individual characteristics, thus helping to ﬁll the developmental gap. A third notable
ﬁnding of the review was that age was relatively strongly and negatively associated
with male-to-female physical aggression. For example, Pan et al. (1994) found for a
military sample of men that for every 10-year increase in age, risk of mild physical
IPV perpetration decreased by 29% and risk of severe physical IPV perpetration
decreased by 19%. Finally, the Schumacher et al. (2001a) review only examined
male perpetration of physical IPV toward women, reﬂecting the predominant view
of IPV in research in the late twentieth century being limited to female victims. More
recent work includes examination of female perpetration of IPV toward men
(Capaldi et al. 2012), as well as examination of IPV perpetration and victimization
in couples other than those identiﬁed as heterosexual (West 2012).
In a second review, Schumacher et al. (2001b) examined risk factors for male-to-
female partner psychological abuse in married and cohabiting couples, an important
complement to the historical focus on physical abuse. They were able to identify
only 10 such studies. The main conclusions were that SES variables did not appear to
be associated with a signiﬁcant increase in risk and that, whereas some couple
relationship variables were associated, it was hard to interpret these due to their
overlap with psychological aggression (e.g., negative communication patterns).
There has been considerably more work in the area of risk for psychological IPV
(e.g., Lohman et al. 2013) since the Schumacher et al. (2001b) review was
Reviews of Risk Factors in the Last Decade
In 2012, Capaldi et al. conducted a systematic review of risk factors for IPV
(physical, psychological, and sexual), including both adolescent and adult studies
and involving both men and women as perpetrators. The review was organized
around the dynamic developmental systems perspective that views couple
aggression as an outcome of couples’interactions, as well as a result of their
predispositions, which are a consequence of genetic vulnerabilities and learning
experiences. Thus, the developmental characteristics and behaviors of each partner –
as well as contextual factors, relationship inﬂuences, and processes –were key foci.
As part of the State of the Knowledge Project initiated and published by Partner
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 7
Abuse, involving reviews on differing aspects of IPV (e.g., prevalence), the reviews
were planned as comprehensive and therefore included cross-sectional as well as
longitudinal studies. Prospective and longitudinal studies provide a very strong
design for identifying risk factors. However, only 26 adolescent and 67 adult (age
18 years or older) longitudinal studies were identiﬁed (many of which involved
multiple studies on the same sample). Furthermore, a number of the longitudinal
adult studies did not assess risk factors during childhood and adolescence. Findings
for longitudinal versus cross-sectional studies are discussed separately by Capaldi et
al. (2012). Note that the 228 publications included in the review represented only
approximately 95 unique samples.
Inclusion criteria for the review were detailed by Capaldi et al. (2012). A key
criterion was that samples needed to be representative of a particular population –i.
e., a community sample (adequate studies of clinical samples with comparison
groups were not found). Samples of convenience and college-student samples
were excluded. A further key criterion was that the study participation rate had to
be 50% or higher, to eliminate studies where systematic bias related to non-
participation would be most likely to affect ﬁndings. Very few studies that met
criteria and were included in the review were published prior to 1996 (8%); close to
one half had been published between 2006 and 2011. The predominant design in the
later published studies was to examine both male and female perpetration of IPV.
These studies chieﬂy examined risk factors for acts of IPV, rather than examining
prediction to outcomes of IPV such as injury. Few studies involving sexual violence
were found that met criteria. Unfortunately, at that time, studies of same gender
couples were generally samples of convenience; thus, no studies with this focus met
the criteria for inclusion. Summary tables from the review are available at https://
domesticviolenceresearch.org/. Full references for reviewed studies are available in
the Partner Abuse State of the Knowledge tables.
Risk factors regarding the contextual characteristics of the partners included
demographic factors (i.e., age, gender, SES, race/ethnicity, acculturation, and stress),
neighborhood and community-level factors (e.g., social disorganization), and school
factors (e.g., perceived school safety). Developmental characteristics of the individ-
uals involved included family factors (exposure to IPV in family of origin, experi-
ence of child abuse, and parenting), peer associations and inﬂuences (association
with deviant peers, social and emotional support), psychological and behavioral
factors (conduct problems/antisocial behavior, personality disorder, depression, sui-
cide attempts, alcohol and drug use, and self-esteem), and cognitive factors (hostile
attributes, attitudes, and beliefs). Risk factors regarding the couples’relationship and
interaction patterns included relationship status, relationship satisfaction, attach-
ment, and negative emotionality and jealousy. Capaldi et al. (2012) reported that
no studies meeting criteria were found on dominance and empathy. Many studies
controlled for additional or confounding factors and several studies were based on
mediational developmental models whereby effects of more distal predictors (e.g.,
demographic factors, witnessing parental violence) were hypothesized to be medi-
ated by more proximal risk factors (e.g., conduct problems, substance use).
8 D. M. Capaldi et al.
Risk Factor Summary from Capaldi et al. (2012)
In the following sections, ﬁndings from the Capaldi et al. (2012) review are
summarized. In addition, ﬁndings from some recent studies are included where
they add new or relevant information.
Demographic Risk Factors and the Role of Age
Regarding review ﬁndings for demographic risk factors, older age is associated with
decreased risk for IPV, with the IPV peak seeming to occur quite early, in late
adolescence and young adulthood. Theoretical explanations for the association of
IPV with age are discussed in the theory section of this chapter.
Other demographic risk factors found predictive of IPV in the Capaldi et al.
(2012) review included deprivation (e.g., unemployment and low income) and
minority group membership, with some evidence that income differences explained
this association. For Hispanics, being born in the United States versus foreign born is
a risk factor, but degree of acculturation did not explain this association. Stress
related to demographic risk factors (e.g., acculturation stress, ﬁnancial and work-
related stress) was also found predictive of IPV.
Studies of community context, including neighborhood and school context, had
mixed ﬁndings, with no clear risk factors emerging. In a recent systematic review,
the association of neighborhood factors and dating violence among adolescents and
young adults less than age 26 year was examined (Johnson et al. 2015). The studies
(n=20) had three main foci, namely, demographic and structural characteristics,
neighborhood disorder, and social disorganization. Evidence indicated associations
between neighborhood disadvantage, perceived neighborhood disorder, low social
control, community connectedness, and IPV. Residence characteristics (e.g., racial
heterogeneity) were not associated with IPV.
Exposure to Interparent IPV and Experience of Maltreatment in
Social learning theory posits that dysfunctional and violent approaches to interper-
sonal problem-solving are learned in the family of origin. Based on this, exposure to
violence between parents in the family of origin and experience of child maltreat-
ment have been a focus of many studies, although they have usually been examined
separately. Overall, there is evidence of a low-to-moderate signiﬁcant association of
these two childhood experiences of violence and later perpetration or victimization
for IPV. However, much of the evidence was from retrospective reporting. An
important factor to consider is that the likelihood that a child will experience some
form of direct child maltreatment is signiﬁcantly higher when there is IPV in the
home (Dong et al. 2004). The co-occurrence of IPV and parent-to-child aggression
has been found in different types of samples. There have been a number of studies
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 9
examining whether there are further developmental links between these childhood
experiences of violence and later IPV involvement. In particular, studies have tested
hypotheses that such family experiences relate to the development of conduct
problems, which in turn is predictive of IPV involvement, and thus may mediate
the association. For example, White and Widom (2003) found that the effects of
child maltreatment on IPV involvement seemed to be mediated by problematic
development subsequent to the maltreatment, including antisocial behavior and
substance use problems.
Protective Family Factors
Capaldi et al. (2012) found that some parenting factors were protective against
children’s later IPV involvement. In particular, positive involvement in the adoles-
cent’s life (monitoring, support) and encouragement of positive problem-solving
were relatively robust, low-to-moderate predictors of lack of dating violence
involvement. Thus, ﬁndings regarding family risk factors for IPV indicate that
family programs to promote positive parental involvement and prevent both child
maltreatment and the development of child conduct problems may also prevent later
Social Risk Factors
Following family factors, social and behavioral factors in adolescence have been
a major focus of IPV risk factor research. They were a secondary focus of a number
of longitudinal studies of the etiology of adolescent behavior problems in the late
1990s and 2000s (e.g., Magdol et al. 1998), as well as the primary focus of a number
of studies conducted in this same period (e.g., Foshee et al. 2005). Theoretically, a
number of these adolescent social and behavioral risk factors can be viewed as
outcomes of early contextual and family risk and, thus, as mediators in the devel-
opmental pathways of risk for IPV involvement. Notably, as these social and
behavioral risk factors are more proximal both chronologically and conceptually to
the outcome of IPV, stronger associations emerge for some predictors as compared
with contextual and family factors.
Similar to ﬁndings for crime and violence, more generally, involvement with
aggressive peers is a relatively robust and strong predictor of involvement in dating
aggression during adolescence, whereas higher friendship quality is a protective
factor. Risk due to peer association is likely to be related ﬁrst to selection factors (i.e.,
differential peer afﬁliation) –for example, dating partners are often met on social
occasions involving friends –and second to social inﬂuence factors within the
friendships (Dishion and Patterson 2006). The nature of the mechanisms of inﬂuence
within the peer group and friendship relations have been less examined but were
tested in a prospective longitudinal study of individuals from ages 11 to 28–30 years
(Ha et al. 2019). The study examined whether the associations between early
10 D. M. Capaldi et al.
disruptive parenting and later problems observed during the interactions of young
couples (namely, negative/hostile interactions and low commitment) were mediated
by peer reinforcement of delinquent behaviors (observed during interactions with
peers). They found support for both the mediated pathways and an additional direct
pathway from early disruptive parenting to negative/hostile interactions.
Social isolation is a risk factor for IPV victimization; thus, the possible protective
factor of social support has been examined albeit in a limited number of studies.
Findings indicate that social support and tangible help are protective against victim-
ization as well as perpetration. In particular, parental support is protective against
adolescent IPV involvement (Capaldi et al. 2012). Further work is needed to
elucidate the protective value of different aspects of support within populations
with differing characteristics (e.g., risk levels).
Behavioral Risk Factors
Major domains of behavioral risk in adolescence and adulthood for IPV include
externalizing and internalizing behaviors and substance use. Conduct problems or
antisocial behaviors, including various forms of societal rule-breaking behavior (e.
g., lying, stealing, vandalism, and violence) and getting into trouble at school and
with the law, have emerged consistently as substantial risk factors for IPV involve-
ment for adolescents and adults who perpetrate IPV (Capaldi et al. 2012). External-
izing behaviors are found frequently to be a mediator for earlier risk factors such as
harsh parental treatment (White and Widom 2003). For the internalizing domain,
fewer studies were identiﬁed than for externalizing behaviors. The ﬁndings for
depressive symptoms (the main form of internalizing symptoms studied) indicate
that whereas they are associated with IPV perpetration and victimization, overall, the
association does not hold up well in multivariate analyses. There was, however,
evidence that depressive symptoms may be a stronger risk factor for IPV perpetra-
tion for women than for men. This is one of the few indications of gender differences
in risk factors identiﬁed in the Capaldi et al. (2012) review. The association between
depressive symptoms and IPV for women was assumed to be due to a causal
association from IPV victimization to depressive symptoms. However, the ﬁndings
that depressive symptoms may be present prior to IPV victimization suggest a
possible causal association in the opposite direction, from depressive symptoms to
IPV victimization. This may be related to the effects of symptoms such as irritability
and negative affect on interactions in intimate relationships. Furthermore, associa-
tions between depressive symptoms and IPV may be reciprocal.
Impulsivity, Anger, Hostility, and Traumatic Brain Injury
Corvo (2014) hypothesizes that of the neuropsychological risks for IPV perpetration,
deﬁcits in executive functioning may be inﬂuential; however, this topic has had little
empirical exploration. Impulsivity and poor inhibitory control are major underlying
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 11
risk factors for crime. Poor control of anger impulses may underlie both conduct
problems and IPV. There is evidence for this contention in a study by Cunradi et al.
(2013) of blue-collar couples (with an average age of just under 40 years) recruited
from a large construction union. They found that 29.6% of the sample reported IPV
and 47.6% of those couples reported bidirectional IPV. Additionally, they reported
that each partner’s level of impulsivity was associated with bidirectional partner
violence; male impulsivity was associated with male-to-female violence, and female
impulsivity was associated with female-to-male violence.
In a test of a mediational developmental model, Reyes et al. (2015) used three
waves of data for 13–14- to 14–15-year-olds to examine the hypotheses that the
prospective associations between witnessing family violence and directly experienc-
ing violence in the family of origin and physical dating aggression perpetration are
mediated by aggression norms, anger dysregulation (related to poor inhibitory
control), and depressive symptoms. Findings indicated that the association between
having been hit by an adult and dating aggression was mediated by changes in norms
about aggression and anger dysregulation, but not depression. There was no evi-
dence of indirect effects from witnessing family violence to dating aggression
through any of the proposed mediators.
Similarly, Giordano et al. (2016) posited that emotional processes, particularly
anger, are a component of IPV incidents –in addition to power and control issues.
In a prospective study of adolescents and young adults, they found that, controlling
for other factors, both partners’control attempts and measures of anger increased the
odds of reporting perpetration and that these ﬁndings did not differ by gender.
Giordano et al. concluded that there is similarity in the emotional processes associ-
ated with young men’s and women’s IPV involvement.
Attention deﬁcit hyperactivity disorder (ADHD) is an ofﬁcial diagnosis related to
poor inhibitory control. In a systematic review of the role of ADHD as a risk factor
for IPV, Buitelaar et al. (2016) posited that emotional lability and poor emotional
self-regulation/control are common features of ADHD and are also likely to be key
risk factors for perpetration of IPV. They concluded that cohort studies identiﬁed
hyperactive, impulsive, and inattention symptoms as risk factors for adult-IPV
A risk factor for IPV that is related to poor inhibitory control, yet rarely men-
tioned in reviews of psychosocial risk factors, is traumatic brain injury. Such
injuries, particularly involving damage to the frontal lobes, may be a cause of poor
inhibitory control (Tateno et al. 2003). A meta-analysis of the association of trau-
matic brain injury with violence (in general) found an odds ratio of 1.66, indicating
that those with such an injury were signiﬁcantly more likely to perpetrate violence
(Fazel et al. 2009). It is possible that traumatic brain injuries may be one link
between experience of maltreatment in childhood and later perpetration of IPV, as
traumatic brain injuries (e.g., due to excessive shaking or hitting the child on the
head) may be suffered due to physical maltreatment in childhood. Thus, traumatic
brain injury and related poor inhibitory control may be mechanisms involved in the
association of this childhood risk factor with later IPV.
12 D. M. Capaldi et al.
Findings in the area of substance use have challenged the conventional wisdom that
alcohol use is a major risk factor for IPV. Whereas some evidence was found for an
association in the Capaldi et al. (2012) review, it was of a low magnitude and not
found consistently, especially when controlling for other factors. However, differ-
ences in sampling, methodology, and consideration of contextual factors have
hindered efforts to determine reliably the magnitude and nature of the role of
substance use in IPV involvement. Given the strong overlap among antisocial
behavior and substance use, it remains important to isolate the independent role of
substance use (i.e., rule out a spurious association), although such studies are sparse.
Substance use is the second risk factor for which possible gender differences are
indicated, such that alcohol use might be a stronger risk factor for women’s IPV
perpetration than for men’s. Capaldi et al. (2012) did ﬁnd evidence that there could
be a stronger association between drug use and IPV than between alcohol use and
Given the potential for differences across gender and, furthermore, the potential
for each partner’s use of substances to be associated with both his/her own IPV
perpetration (i.e., actor effect) and IPV victimization (i.e., partner effect), Low et al.
(2017a) examined the substance use-IPV association in young adulthood using a
dyadic actor-partner framework. This study included alcohol, marijuana, and poly-
substance users. Findings suggested alcohol use was a stronger risk factor for men’s
IPV than for women’s IPV, as was polysubstance use; in all cases, the associations
were stronger for the perpetration of psychological rather than physical IPV.
Collectively, more studies are needed that account for the co-occurrence of sub-
stances and contextual, relational, and individual risk factors (e.g., antisocial behav-
ior). More nuanced methodological designs, such as diary or electronic momentary
assessment data, may also help advance knowledge of the precise role substances
play. Use of substances may be the source of conﬂict for many couples or could be a
coping strategy for couples in distress. Survey data is insufﬁcient for full under-
standing of the underlying dynamics that unfold in real time in day-to-day interac-
tions between partners. For example, for the OYS-Couples Study, physical
aggression was observed for some couples who did not report such aggression,
and observed reciprocated physical aggression, more often initiated by the women,
was associated with a greater likelihood of injury (Capaldi et al. 2007).
Relationship factors have received less research attention than have contextual and
developmental characteristics and behaviors of partners. Relationship status (e.g.,
married, cohabiting, separated) is related to IPV, with married individuals being at
lowest risk and separated women being particularly vulnerable. The proximal risk
factors of low relationship satisfaction and high discord or conﬂict are predictive of
IPV. It was outside the scope of the Capaldi et al. (2012) review to examine
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 13
associations among the forms of IPV, but it should be noted that there are signiﬁcant
associations among the forms of IPV. In particular, psychological IPV, which is
highly related to relationship conﬂict, is predictive of physical IPV (O’Leary 1999).
In more recent work regarding relationship risk factors, Collibee and Furman
(2016) examined acute and chronic relational risk factors (negative interactions,
jealousy, support, and relationship satisfaction) and their effects on physical and
psychological dating aggression from middle adolescence to young adulthood.
Findings indicated that both acute (within-person) and chronic (between-person)
levels of jealousy, negative interactions, and low relationship satisfaction were
associated with physical and psychological dating aggression. Signiﬁcant interac-
tions between chronic and acute risk in predicting physical aggression were indi-
cated for negative interactions, jealousy, and relationship satisfaction. Individuals
with higher levels of chronic risk are more vulnerable to increases in acute risk (e.g.,
for jealousy). Collibee and Furman conclude that prevention programs for higher-
risk youth and young couples would beneﬁt from addressing periods of acute risk (e.
g., discovering partner inﬁdelity, breakups).
A key ﬁnding of Capaldi et al. (2012) regarding gender was that regardless of any
differences in frequency and/or severity of engagement in IPV, overall, there were
more similarities than differences in risk factors by sex. The main area where there
was relatively robust evidence of gender differences was for internalizing problem
behaviors, which were more consistent risk factors for IPV perpetration for women
than for men. The second area where there was some evidence for gender differences
was that alcohol use is a greater risk factor for IPV perpetration for girls/women than
for boys/men; although, again, such patterns have not been consistent.
Other Recent Reviews
Costa et al. (2015) conducted a systematic review of longitudinal predictors of
domestic violence perpetration and victimization in adulthood. Only 25 studies
were identiﬁed, from 14 different cohorts. Note that of the 25 studies in Costa et
al., 14 were also included in the Capaldi et al. (2012) review. One exclusion criterion
used by Costa et al., which likely related to fewer adult longitudinal studies being
identiﬁed as compared to Capaldi et al., is that they required the ﬁrst data collection
and measurement of the risk factor to have been prior to age 18 years, whereas
Capaldi et al. included longitudinal studies where all assessments were during
adulthood. Thus, Costa et al. found relatively few studies of developmental risk
for IPV in adulthood with a strong developmental design (i.e., not involving
retrospective reports of risk factors). Eighteen of the studies were conducted in the
United States, and six of those were from the same data set, namely, the Survey of
Adolescent Health. Given the importance of understanding risk for IPV as a public
14 D. M. Capaldi et al.
health question, the sparsity of studies with such a basically adequate developmental
design is a concerning state of research. Costa et al. note that they adopted the
approach of Capaldi et al. in examining the contribution of each predictor to the
perpetration and victimization of IPV for men and women considered together.
Similar to Capaldi et al., they found few differences in predictors between men
and women or for the perpetration versus victimization of IPV. Furthermore, they
found that there were still few studies that comprehensively considered sexual,
psychological, and verbal abuse, which precluded evaluating speciﬁc types of IPV
Costa et al. (2015) found that ﬁve domains of risk were consistently predictive of
IPV victimization, namely, child and adolescent abuse experiences, family-of-origin
risks (again, the most frequently investigated and consistent predictors), child and
adolescent behavioral risks (including aggressive behavior, withdrawal, conduct
disorder, and alcohol and drug use), peer risks in adolescence (note, however, that
only two studies examined peer risks), and sociodemographic risks, although the
latter were relatively weak predictors. One conclusion reached by Costa et al. is that
intervening on parenting, both to prevent child abuse and the development of
antisocial behaviors, is indicated for prevention of later IPV involvement.
Cumulative and Compounded Risk Factors
Developmental pathways from childhood risk factors to more proximal risk factors
to IPV involvement such as antisocial behavior, as well as the inﬂuence of demo-
graphic factors such as gender and race, have been discussed. However, there is also
growing evidence of additive or cumulative effects of risk factors (Hamby and Grych
2013). Capaldi et al. (2012) noted that IPV shared many risk factors for other
problems that involve risky behavior in adolescence and adulthood such as crime,
substance use, and sexual risk behaviors. Thus, IPV may share a similar etiology to
other conduct problems. Relatedly, many of the risk factors examined tend to
be interrelated. A prior study by Capaldi et al. (2002) examined the association
of contextual risk factors for antisocial behavior across the early lifespan for the
Oregon Youth Study (OYS) sample. They found that the family risk factors of low
income, parental antisocial behavior, depressive symptoms, transitions (separations,
repartnerings), SES, high stress, and unemployment were signiﬁcantly associated
and formed a unidimensional meta-risk factor.
Regarding cumulative risk, Smith et al. (2015) examined adolescent risk for IPV
perpetration at ages 20–22 and 29–30 years using data from the longitudinal
Rochester Youth Development Study, which originally focused on risk for antisocial
behavior and crime. Similar to the Capaldi et al. (2012) review, the results indicated
that risk factors for IPV spanned several developmental domains and were substan-
tially similar for both genders. They also found that cumulative risk across a number
of developmental domains placed adolescents at particularly high risk of perpetrat-
ing partner violence. As discussed by Smith et al. (2015), ﬁndings indicate that
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 15
prevention efforts should focus on the identiﬁed risk factors, but should also be
directed at higher-risk groups showing multiple risk factors.
A key way in which cumulative risk affects risk for IPV is via the tendency for
romantic partners to show associations in levels of risk characteristics, particularly
conduct problems or antisocial behavior. As discussed by Capaldi et al. (2018), such
assortative partnering occurs due to a number of factors. First, partners usually met
and were selected during interactions or social occasions within the peer group, and
for individuals with elevated levels of antisocial behavior, the peer group was likely
to include others with elevated levels of antisocial behavior. Second, individuals tend
to be attracted to partners with interests and characteristics similar in some respects
to their own. Thus, a boy with strong academic interests may prefer a partner with
similar academic strengths, or a girl who likes to party and use substances is likely to
choose a partner who enjoys similar activities. Finally, individuals may grow more
similar in their behaviors once they are in a dating relationship. In the OYS-Couples
Study, signiﬁcant associations in antisocial behavior across partners in late adoles-
cence were found for the couples (r=0.44, p<0.001), and such assortative
partnering was related to considerably higher levels of physical IPV in the couples
(Kim et al. 2016). If both partners (at age 21 years on average) showed antisocial
behavior at or above the 67th percentile (standardized separately for men and
women), their mean physical IPV level on a standardized score was 0.60 versus
0.20 if neither partner was elevated or almost 0 if just one partner was elevated
(note that in the latter case results did not differ depending on whether it was the
young woman or the young man who was elevated in antisocial behavior). Thus,
assortative partnering by antisocial behavior is a case of cumulative risk for the
couple, not just for the individual, and relates to risk for bidirectional IPV and
escalation of IPV for couples.
Do Risk Factors Differ by Severity of Violence?
There has been an ongoing debate regarding whether severe IPV (e.g., more frequent
and chronic, resulting in injuries) differs qualitatively from moderate or lower level
IPV (e.g., less frequent, less likelihood of injury). This debate originated from
observations that samples recruited from the general population (e.g., phone survey
samples) had lower frequencies of IPV acts and a low prevalence of injuries than
samples such as women’s shelter samples (Capaldi and Owen 2001).
In their review of risk factors for IPV, Costa et al. (2015) argue that their risk
factors likely apply to the perpetration of lower-to-moderate levels of IPV, which
they and others (Johnson 1995) characterize as common couple violence, but may
not apply to more serious IPV characterized as intimate terrorism. According to
Johnson (1995), common couple violence is generally found in the larger-scale
16 D. M. Capaldi et al.
surveys and involves conﬂicts between partners that are poorly managed, occasion-
ally escalate to minor violence, and more rarely escalate to serious violence. Such
violence is hypothesized to be more likely to be bidirectional than more severe
violence, of lower frequency, and less likely to persist. Johnson argues that intimate
terrorism usually involves patterned violence with the purpose of maintaining
domination, is typically found in agency samples (e.g., women’s shelter samples,
IPV treatment samples), and is likely to be much more frequent, persistent, one-
sided, and mainly perpetrated by men.
The major argument regarding non-overlap of survey versus agency samples is
that samples of convenience and community samples fail to recruit more severely
violent people because such individuals are unlikely to volunteer for a study or to
agree to participate in a study when recruited or may be incarcerated, whereas
samples such as women’s shelter samples come from these same severely violent
families. There is some justiﬁcation for the claim that individuals with higher levels
of antisocial behavior and violence are less likely to participate in community
surveys. At the time of initial recruitment to the OYS (in 1984–1985), a number of
the fourth-grade families to be recruited did not have a home telephone. These
families were difﬁcult to recruit, requiring a cold home visit without being able to
call ahead and schedule an appointment. In a comparison of juvenile arrest histories
of boys from these families (13% of the sample) versus families with phones, it was
found that 62% of the no-phone family boys had three or more police referrals by age
18 years versus 19% of the other boys. Furthermore, 23% of the no-phone family
boys had 3–5 police referrals for violence versus 1% of the other boys. The most
difﬁcult to recruit families showed very high levels of problem behaviors among the
However, there are a number of ﬂaws to the argument that the kinds of violence
found in the two kinds of samples are qualitatively different (suggesting possible
qualitatively different risk factors), rather than that IPV is better characterized as a
continuum –running from relatively minor and infrequent to more frequent and
severe –as argued (Capaldi and Kim 2007). First, rather than IPV in shelter samples
being predominantly one-sided intimate terrorism, there is some evidence of
bidirectionality of violence even within shelter samples. McDonald et al. (2009)
conducted a study of children’s adjustment in families with severe violence by male
partners toward the mother, using a shelter sample of mothers who had experienced
at least one act of partner-to-mother physical violence in the past year. Over the
same period, it was found that 96% of the men and 67% of the women (according
to only the women’s reports) had engaged in severe violence toward a partner.
Thus, violence was bidirectional in the majority of these shelter cases. A study by
Langhinrichsen-Rohling (2010) indicated that for a sample of sheltered women IPV
victims, 25% reported stalking behaviors against the men who were violent toward
them. These women were signiﬁcantly more likely to be depressed and to leave the
shelter within the ﬁrst week than the non-stalking women and thus were more likely
to be considered treatment failures than the non-stalking women. Again, this study
provides evidence of bidirectionality of IPV within women’s shelter samples.
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 17
Jasinski et al. (2014) tested the Johnson typology as detailed by Johnson and
Leone (2005) using data regarding victimization of men and women (not perpetra-
tion by either) from the National Violence Against Women Study. To be consistent
with Johnson and Leone, they focused only on married respondents. Findings
indicated that men experienced higher levels of victimization on the violence scale
than did women and that there was no signiﬁcant gender difference for severe
violence. Furthermore, there were no gender differences on the nonviolent control
scale. Regarding intimate terrorism, by the Johnson and Leone criteria, Jasinski et al.
(2014) found that 36% of women and 35% of men were victims of intimate
terrorism. In addition, post-traumatic stress disorder symptom levels were similar
for men and women, but depressive symptoms were higher for females. Consistent
with ﬁndings of a number of other studies, gender differences were found for
injuries, which were experienced by 24% of the women and 11% of the men.
Further evidence calling into question the contention of two qualitatively differ-
ent groups of IPV –namely, individuals who experience common couples’violence
(more minor) versus partners who experience one-sided intimate terrorism (more
severe) –comes from a study of risk factors associated with IPV among patients
visiting an inner-city emergency department (Bazargan-Hejazi et al. 2014). Nearly
16% had experienced IPV. Overall, the individuals in the IPV group were younger,
had more childhood exposure to violence, and were more depressed and more
impulsive than the non-IPV group. They were also more likely to engage in binge
drinking and drug use. In the IPV group, 31% were perpetrators, 20% victims, and
49% both victims and perpetrators, with the latter group showing higher levels of
risk factors. Thus, even among a relatively high-risk group, namely, inner-city
emergency room patients, close to one half of those involved in IPV were involved
in bidirectional IPV. This group was at more risk of experiencing background risk
factors compared with those involved in unidirectional IPV. Bazargan-Hejazi et al.
conclude that alcohol and drug use, depressive symptoms, and childhood exposure
to violence may be factors and signs for which emergency physicians should screen
in the context of IPV.
In a prior study with the OYS sample, Capaldi and Owen (2001) found that the
proportion of couples where both partners were frequently aggressive (deﬁned as a
cutoff score of 19 or more acts) was 6 times higher than expected by chance. In
addition, for bidirectionally aggressive couples, injuries were likely to have occurred
for both partners at three times the rate expected by chance. Capaldi et al. (2007)
found that women and men appeared more likely to report injuries if the couple was
observed (by interaction task coders) to use bidirectional physical aggression during
a task where they discussed problems in their relationship. Findings from Whitaker
and his colleagues (Whitaker et al. 2007) also support the argument that injuries are
more likely to occur when aggression is bidirectional rather than one-sided.
Therefore, several key aspects of the Johnson hypothesis have yet to be
established: ﬁrst, that the motivations behind the violence are signiﬁcantly different
for more severe versus less severe violence and, second, that more severe violence is
one-sided rather than bidirectional. In the OYS-Couples Study sample –for which
ﬁndings regarding risk factors have been very consistent with ﬁndings from other
18 D. M. Capaldi et al.
survey samples –in addition to ﬁnding relatively high levels of IPV-related injuries
(compared with phone survey samples), searches of police records through ages
26–27 years revealed 40 arrests of the approximately 200 men for IPV, involving a
total of 28 men (some men had multiple arrests; Capaldi et al. 2009). The OYS-
Couples Study therefore included severe cases of IPV. The proposition that two
groups versus a continuum would emerge from population-based samples, including
severe levels of IPV, has not been shown. Thus, there is no compelling argument that
risk factors for severe versus moderate IPV differ in type. However, it would
certainly be expected that those exhibiting more frequent, severe, and persistent
IPV perpetration would have the most severe levels of risk factors, including
psychopathology, and tend to show multiple or cumulative risk factors.
Recent Work Testing the DDS Model
Recent work examining aspects of the developmental pathways in the DDS model
includes a study of the temporal relation between externalizing and internalizing
symptoms and IPV. Low et al. (2017b) tested a prospective mediational model
(spanning 9 years) in which adolescent psychopathology symptoms (i.e., internaliz-
ing, externalizing, and combined) mediated the association between exposure to IPV
in middle childhood and young-adult-IPV perpetration. Analyses also controlled for
proximal young-adult partner and relationship characteristics. Such indirect effect
models are surprisingly rare, as there are still relatively few prospective designs
utilized in studies of behavioral risk factors. Findings indicated that only external-
izing symptoms mediated pathways from IPV exposure to adult-IPV involvement.
Furthermore, gender moderation analyses of the mediational processes revealed
differences in sensitivity to exposure across developmental periods –for males,
effects of exposure intensiﬁed during the transition to adolescence; on the other
hand, for females, effects were ampliﬁed during the transition from adolescence to
adulthood. In both cases, the mediational role of psychopathology symptoms was no
longer signiﬁcant once partner antisocial behavior was controlled, indicating that the
risk was dyadic. These ﬁndings have important prevention implications (e.g., timing
of risk conveyance for boys vs. girls) and validate the robust role of externalizing in
the transmission of IPV. Given that risk for IPV is intensiﬁed earlier in development
for men, intervening in childhood to reduce immediate (or short-term) impacts could
be most beneﬁcial in curtailing risk trajectories for men. In contrast, interventions
targeted at adolescent girls who exhibit higher levels of externalizing behaviors
could be most impactful in mitigating progress toward IPV involvement in young
Key issues and ﬁndings regarding theory and risk factors related to the development
of IPV based on the studies discussed and reviewed in this chapter are as follows.
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 19
1. Much IPV is bidirectional within couples, and IPV is dynamic and changes over
time. Thus, theories regarding the development of IPV need to encompass these
fundamental characteristics of IPV. One such theory, the dynamic developmental
systems theory, was presented.
2. Whereas many risk factors (e.g., neighborhood factors, family risk factors) show
associations with IPV, individual risk factors in the externalizing domain (e.g.,
antisocial behavior and related criminal involvement, substance use including
drug abuse) tend to show the strongest associations.
3. Findings indicate developmental sequences of risk for IPV, where factors
in childhood (e.g., parental harsh physical discipline) place the individual in
jeopardy of future risk factors (e.g., antisocial behavior in adolescence), which
show a stronger association with IPV than do the earlier factors.
4. A key aspect of risk for IPV is that cumulative risk or risk from multiple domains
(e.g., neighborhood risk plus family risk plus externalizing behaviors) is associ-
ated with higher levels of risk for IPV than any particular risk factor alone.
A major form of cumulative risk is when both members of the dyad have risk
factors (e.g., a couple where both partners show higher levels of antisocial
5. Regardless of any differences in the occurrence, frequency, and severity of IPV
by gender, overall, ﬁndings regarding risk factors indicate considerably more
similarities than differences in risk factors by gender. The one exception for
which there is relatively robust evidence is that depressive symptoms appear to
be a stronger risk factor for IPV perpetration by women than by men.
Summary and Conclusions
Implications for Future Research
The review by Capaldi et al. (2012) offered a number of implications for future
research that are still relevant. These include the need for stronger study designs
involving representative community samples, both partners in an intimate relation-
ship, and multiple modes of measurement, rather than studies of self-reports from
one partner only (i.e., egocentric designs). Second, studies incorporating same-sex
couples and/or comparing same-sex and heterosexual couples are limited. A chal-
lenge in this regard is that it is very difﬁcult to obtain a representative sample (as
opposed to a convenience sample) of same-sex couples. Third is the need for
prospective developmental models of additive and cumulative risks, particularly
dyadic designs that study effects of assortative partnering and effects of changes in
partners or repartnering in IPV perpetration. Studies examining correspondence
in risk factors (e.g., substance use) and IPV over time, as well as between
partners, using multilevel modeling or time-sensitive methods are also needed. In
addition, creative experimental designs to examine these associations are needed (e.
g., Leonard and Roberts 1998), such as alcohol and marijuana administration
studies. Regarding childhood abuse experience, Costa et al. (2015) propose that
20 D. M. Capaldi et al.
future studies should examine issues including type of abuse, severity (such as
whether the abuse involved a traumatic brain injury), and duration –as well as
developmental period when the abuse was experienced (e.g., childhood vs. adoles-
cence) –as these may differentially relate to IPV.
Prevention and intervention programs to reduce IPV should reﬂect well-established,
malleable risk factors, rather than hypothesized but not adequately or rigorously
tested risk factors (Capaldi et al. 2012). Findings that foundations of risk are present
from early childhood support the contention of Costa et al. (2015) and Eddy et al.
(2018) that a lifespan approach should be used for the effective prevention of IPV.
Intervention efforts can start with early programs to support vulnerable parents (e.g.,
the Nurse-Family Partnership program, Olds et al. 1986; Promoting First Relation-
ships, Spieker et al. 2012) and continue with school-based programs to mitigate or
prevent poor behavioral control, antisocial behaviors, and associated risk behaviors
–including delinquent peer associations and substance use (including drug use),
with particular emphasis on children from higher-risk backgrounds. Given the
interconnections among types of violence (e.g., bullying and dating violence, com-
munity violence, and IPV) and common epidemiology, prevention programs should
consider aggression across contexts and address cumulative risk assessed through
comprehensive assessment. Prevention programs still heavily reﬂect the paradigm
that only women can be victims of domestic violence/IPV, and only men can be
perpetrators of IPV, despite data validating high levels of bidirectional aggression
and assortative partnering effects. In addition, parenting interventions disproportion-
ately target mothers. Taken together, the ﬁndings suggest that prevention efforts
seeking to interrupt the intergenerational transmission of IPV would beneﬁt from
early parenting supports and a dyadic framework.
Education efforts and prevention campaigns should start by the adolescent years
and provide accurate information that can help both partners to protect themselves.
Practitioners and those in advocacy roles should understand that women may be at
risk for perpetration, as well as victimization, and be aware of the factors that place
women at higher risk. Conversely, it is important to increase awareness that men may
be at risk for victimization, as well as perpetration. Both women and men should
understand the risks associated with bidirectional IPV, including injuries. Finally,
given the proximal risk factors of couple conﬂict and dissatisfaction, programs
promoting positive problem-solving skills and relationship satisfaction for couples
may be valuable for preventing IPV as well as fostering a more positive environment
for the children of couples at risk for IPV.
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 21
▶Prevention Across the Life Cycle
▶Intergenerational Transmission of Abuse
▶Intimate Partner Violence (IPV) and Health Systems Response
▶Intimate Partner Violence (IPV) and Mental Health Systems Response
▶Feminist Theories and Perspectives of Intimate Partner Violence and Abuse
▶Other Theories and Models of Intimate Partner Violence
▶Intimate Partner Violence and Substance Abuse
▶Male Victims of Intimate Partner Violence
Ali, P., Dhingra, K., & McGarry, J. (2016). A literature review of intimate partner violence and its
classiﬁcations. Aggression and Violent Behavior, 31,16–25. https://doi.org/10.1016/j.
Bazargan-Hejazi, S., Kim, E., Lin, J., Ahmadi, A., Khamesi, M. T., & Teruya, S. (2014). Risk
factors associated with different types of intimate partner violence (IPV): An emergency
department study. Journal of Emergency Medicine, 47, 710–720. https://doi.org/10.1016/j.
Blumstein, A., Cohen, J., Roth, J. A., & Visher, C. A. (Eds.). (1986). Criminal careers and career
criminals (Vol. 1). Washington, DC: National Academy Press.
Bronfenbrenner, U. (1995). The bioecological model from a life course perspective: Reﬂections of
a participant observer. In P. Moen, G. H. Elder Jr., & K. Luscher (Eds.), Examining lives in
context (pp. 559–618). Washington, DC: American Psychological Association.
Buitelaar, N. J. L., Posthumus, J. A., & Buitelaar, J. K. (2016). ADHD in childhood and/or
adulthood as a risk factor for domestic violence or intimate partner violence: A systematic
review. Journal of Attention Disorders.https://doi.org/10.1177/1087054715587099.
Capaldi, D. M., & Kim, H. K. (2007). Typological approaches to violence in couples: A critique and
alternative conceptual approach. Clinical Psychology Review, 27, 253–265. https://doi.org/
Capaldi, D. M., & Owen, L. D. (2001). Physical aggression in a community sample of at-risk,
young couples: Gender comparisons for high frequency, injury, and fear. Journal of Family
Psychology, 15, 425–440. https://doi.org/10.1037/0893-3126.96.36.1995.
Capaldi, D. M., Dishion, T. J., Stoolmiller, M., & Yoerger, K. (2001). Aggression toward female
partners by at-risk young men: The contribution of male adolescent friendships. Developmental
Psychology, 37,61–73. https://doi.org/10.1037/0012-16188.8.131.52.
Capaldi, D. M., DeGarmo, D. S., Patterson, G. R., & Forgatch, M. S. (2002). Contextual risk across
the early life span and association with antisocial behavior. In J. B. Reid, G. R. Patterson, & J.
Snyder (Eds.), Antisocial behavior in children and adolescents: A developmental analysis and
model for intervention (pp. 123–145). Washington, DC: American Psychological Association.
Capaldi, D. M., Kim, H. K., & Shortt, J. W. (2007). Observed initiation and reciprocity of physical
aggression in young, at-risk couples. Journal of Family Violence, 22, 101–111. https://doi.org/
Capaldi, D. M., Shortt, J. W., Kim, H. K., Wilson, J., Crosby, L., & Tucci, S. (2009). Ofﬁcial
incidents of domestic violence: Types, injury, and associations with nonofﬁcial couple aggres-
sion. Violence and Victims, 24, 502–519. https://doi.org/10.1891/0886-6708.24.4.502.
22 D. M. Capaldi et al.
Capaldi, D. M., Knoble, N. B., Shortt, J. W., & Kim, H. K. (2012). A systematic review of risk
factors for intimate partner violence. Partner Abuse, 3, 231–280. https://doi.org/10.1891/1946-
Capaldi, D. M., Shortt, J. W., Tiberio, S. S., & Low, S. (2018). Violence begets violence:
Addressing the dual nature of partner violence in adolescent and young adult relationships.
In D. A. Wolfe & J. R. Temple (Eds.), Adolescent dating violence: Theory, research, and
prevention (pp. 341–364). Cambridge, MA: Academic.
Casey, B. J., Getz, S., & Galvan, A. (2008). The adolescent brain. Developmental Review, 28,
Caspi, A., & Elder, G. H. (1988). Childhood precursors of the life course: Early personality and life
disorganization. In E. M. Hetherington, R. M. Lerner, & M. Perlmutter (Eds.), Child develop-
ment in life-span perspective (pp. 115–142). Hillsdale: Lawrence Erlbaum.
Collibee, C., & Furman, W. (2016). Chronic and acute relational risk factors for dating aggression in
adolescence and young adulthood. Journal of Youth and Adolescence, 45, 763–776. https://doi.
Corvo, K. (2014). The role of executive function deﬁcits in domestic violence perpetration. Partner
Abuse, 5, 342–355. https://doi.org/10.1891/1946-65184.108.40.2062.
Corvo, K., & Johnson, P. (2013). Sharpening Ockham’s razor: The role of psychopathology
and neuropsychopathology in the perpetration of domestic violence. Aggression and Violent
Behavior, 18, 175–182. https://doi.org/10.1016/j.avb.2012.11.017.
Costa, B. M., Kaestle, C. E., Walker, A., Curtis, A., Day, A., Toumbourou, J. W., & Miller, P.
(2015). Longitudinal predictors of domestic violence perpetration and victimization: A system-
atic review. Aggression and Violent Behavior, 24, 261–272. https://doi.org/10.1016/j.
Cunradi, C. B., Ames, G. M., & Duke, M. (2013). The relationship of alcohol problems to the risk
for unidirectional and bidirectional intimate partner violence among a sample of blue-collar
couples. Violence and Victims, 26, 147–158. https://doi.org/10.1891/0886-6708.26.2.147.
Dishion, T. J., & Patterson, G. R. (2006). The development and ecology of antisocial behavior in
children and adolescents. In D. Cicchetti & D. J. Cohen (Eds.), Developmental psychopathol-
ogy: Risk, disorder, and adaptation (Vol. 3, pp. 503–541). New York: Wiley.
Dong, M., Anda, R. F., Felitti, V. J., Dube, S. R., Williamson, D. F., Thompson, T. J., et al. (2004).
The interrelatedness of multiple forms of childhood abuse, neglect, and household dysfunction.
Child Abuse & Neglect, 28, 771–784.
Eddy, J. M., Schumer, J., Shortt, J. W., Capaldi, D. M., Tiberio, S. S., & Low, S. (2018). A life-
course model for the development of intimate partner violence. In A. T. Vazsonyi, D. J.
Flannery, & M. DeLisi (Eds.), The Cambridge handbook of violent behavior and aggression
(2nd ed., pp. 52–64). New York: Cambridge University Press.
Fazel, S., Philipson, J., Gardiner, L., Merritt, R., & Grann, M. (2009). Neurological disorders and
violence: A systematic review and meta-analysis with a focus on epilepsy and traumatic brain
injury. Journal of Neurology, 256, 1591–1602. https://doi.org/10.1007/s00415-009-5134-2.
Foshee, V. A., Ennett, S. T., Bauman, K. E., Beneﬁeld, T., & Suchindran, C. (2005). The association
between family violence and adolescent dating violence onset: Does it vary by race, socioeco-
nomic status, and family structure? Journal of Early Adolescence, 25, 317–344.
Gelles, R. J., & Straus, M. A. (1979). Determinants of violence in the family: Toward a theoretical
integration. In W. R. Burr, R. Hill, F. I. Nye, & I. L. Reiss (Eds.), Contemporary theories about
the family (pp. 549–581). New York: Free Press.
Giordano, P. C., Copp, J. E., Longmore, M. A., & Manning, W. D. (2016). Anger, control, and
intimate partner violence in young adulthood. Journal of Family Violence, 1,1–13. https://doi.
Ha, T., Otten, R., McGill, S., & Dishion, T. J. (2019). The family and peer origins of coercion within
adult romantic relationships: A longitudinal multimethod study across relationships contexts.
Developmental Psychology, 55, 207–215. https://doi.org/10.1037/dev0000630.
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 23
Hamby, S., & Grych, J. (2013). The web of violence: Exploring connections among forms of
interpersonal violence and abuse. New York: Springer.
Holtzworth-Munroe, A., & Stuart, G. L. (1994). Typologies of male batterers: Three subtypes and
the differences among them. Psychological Bulletin, 116, 476–497.
Holtzworth-Munroe, A., Meehan, J. C., Herron, K., Rehman, U., & Stuart, G. L. (2003).
Do subtypes of martially violent men continue to differ over time? Journal of Consulting and
Clinical Psychology, 71, 728–740. https://doi.org/10.1037/0022-006X.71.4.728.
Jasinski, J., Blumenstein, L., & Morgan, R. (2014). Testing Johnson’s typology: Is there gender
symmetry in intimate terrorism? Violence and Victims, 29,73–88. https://doi.org/10.1891/0886-
Johnson, M. P. (1995). Patriarchal terrorism and common couple violence: Two forms of violence
against women. Journal of Marriage and the Family, 57, 283–294.
Johnson, M. P., & Leone, J. M. (2005). The differential effects of intimate terrorism and situational
couple violence: Findings from the National Violence Against Women Survey. Journal of
Family Issues, 26, 322–349.
Johnson, R. M., Parker, E. M., Rinehart, J., Nail, J., & Rothman, E. F. (2015). Neighborhood factors
and dating violence among youth: A systematic review. American Journal of Preventive
Medicine, 49, 458–466. https://doi.org/10.1016/j.amepre.2015.05.020.
Kim, H. K., & Capaldi, D. M. (2004). The association of antisocial behavior and depressive
symptoms between partners and risk for aggression in romantic relationships. Journal of Family
Psychology, 18,82–96. https://doi.org/10.1037/0893-3220.127.116.11.
Kim, H. K., Laurent, H. K., Capaldi, D. M., & Feingold, A. (2008). Men’s aggression toward
women: A 10-year panel study. Journal of Marriage and Family, 70, 1169–1187. https://doi.
Kim, H. K., Tiberio, S. S., Capaldi, D. M., Shortt, J. W., Squires, E. C., & Snodgrass, J. J. (2015).
Intimate partner violence and diurnal cortisol patterns in couples. Psychoneuroendrocrinology,
Kim, H. K., Shortt, J. W., Tiberio, S. S., & Capaldi, D. M. (2016). Aggression and coercive
behaviors in early adult relationships: Findings from the Oregon Youth Study–Couples Study.
In T. J. Dishion & J. J. Snyder (Eds.), Oxford handbook of coercive relationship dynamics (pp.
169–181). New York: Oxford University Press.
Langhinrichsen-Rohling, J. (2010). Controversies involving gender and intimate partner violence in
the United States. Sex Roles, 62, 179–193. https://doi.org/10.1007/s11199-009-9628-2.
Langhinrichsen-Rohling, J., Misra, T. A., Selwyn, C., & Rohling, M. L. (2012). Rates of bidirec-
tional versus unidirectional intimate partner violence across samples, sexual orientations, and
race/ethnicities: A comprehensive review. Partner Abuse, 3, 199–230. https://doi.org/10.1891/
Leonard, K. E., & Roberts, L. J. (1998). The effects of alcohol on the marital interactions of
aggressive and nonaggressive husbands and their wives. Journal of Abnormal Psychology, 107,
Lohman, B. J., Neppl, T. K., Senia, J. M., & Schoﬁeld, T. J. (2013). Understanding adolescent and
family inﬂuences on intimate partner psychological violence during emerging adulthood and
adulthood. Journal of Youth and Adolescence, 42, 500–517. https://doi.org/10.1007/s10964-
Low, S., Tiberio, S. S., Shortt, J. W., Capaldi, D. M., & Eddy, J. M. (2017a). Associations of
couples’intimate partner violence in young adulthood and substance use: A dyadic approach.
Psychology of Violence, 7, 120–127. https://doi.org/10.1037/vio0000038.
Low, S., Tiberio, S. S., Shortt, J. W., Mulford, C., Eddy, J. M., & Capaldi, D. M. (2017b).
Intergenerational transmission of violence: The mediating role of adolescent psychopathology.
Development and Psychopathology, 31, 233. https://doi.org/10.1017/S0954579417001833.
Advanced online publication.
24 D. M. Capaldi et al.
Magdol, L., Mofﬁtt, T. E., Caspi, A., & Silva, P. A. (1998). Hitting without a license: Testing
explanations for differences in partner abuse between young adult daters and cohabitors.
Journal of Marriage and the Family, 60,41–55.
McDonald, R., Jouriles, E. N., Tart, C. D., & Minze, L. C. (2009). Children’s adjustment problems
in families characterized by men’s severe violence toward women: Does other family violence
matter? Child Abuse and Neglect, 33,94–101. https://doi.org/10.1016/j.chiabu.2008.03.005.
O’Leary, K. D. (1999). Psychological abuse: A variable deserving critical attention in domestic
violence. Violence and Victims, 14,3–23.
O’Leary, K. D., & Slep, A. M. S. (2003). A dyadic longitudinal model of adolescent dating
aggression. Journal of Clinical Child and Adolescent Psychology, 32, 314–327.
Olds, D. L., Henderson, C. R., Chamberlin, R., & Tatelbaum, R. (1986). Preventing child abuse and
neglect: A randomized trial of nurse home visitation. Pediatrics, 78,65–78.
Pan, H. S., Neidig, P. H., & O’Leary, K. D. (1994). Predicting mild and severe husband-to-wife
physical aggression. Journal of Consulting and Clinical Psychology, 62, 975–981.
Reyes, H. L. M., Foshee, V. A., Fortson, B. L., Valle, L. A., Breiding, M. J., & Merrick, M. T.
(2015). Longitudinal mediators of relations between family violence and adolescent dating
aggression perpetration. Journal of Marriage and Family, 77, 1016–1030. https://doi.org/
Riggs, D. S., & O’Leary, K. D. (1989). A theoretical model of courtship aggression. In M. A. Pirog-
Goood & J. E. Stets (Eds.), Violence in dating relationships: Emerging social issues (pp.
53–71). New York: Praeger Publications.
Schumacher, J. A., Feldbau-Kohn, S. R., Slep, A. M. S., & Heyman, R. E. (2001a). Risk factors for
male-to-female partner physical abuse. Aggression and Violent Behavior, 6, 281–352.
Schumacher, J. A., Slep, A. M. S., & Heyman, R. E. (2001b). Risk factors for male-to-female
partner psychological abuse. Aggression and Violent Behavior, 6, 255–268.
Shortt, J. W., Capaldi, D. M., Kim, H. K., & Tiberio, S. S. (2013). The interplay between
interpersonal stress and psychological intimate partner violence over time for young at-risk
couples. Journal of Youth and Adolescence, 42, 619–632. https://doi.org/10.1007/s10964-013-
Smith, C. A., Greenman, S. J., Thornberry, T. P., Henry, K. L., & Ireland, T. O. (2015). Adolescent
risk for intimate partner violence perpetration. Prevention Science, 16, 862–872. https://doi.org/
Spieker, S. J., Oxford, M. L., Kelly, J. F., Nelson, E. M., & Fleming, C. B. (2012). Promoting ﬁrst
relationships: Randomized trial of a relationship-based intervention for toddlers in child welfare.
Child Maltreatment, 17, 271–286. https://doi.org/10.1177/1077559512458176.
Steinberg, L. (2010). A dual systems model of adolescent risk-taking. Developmental Psychobiol-
ogy, 52, 216–224. https://doi.org/10.1002/dev.20445.
Stets, J. E., & Straus, M. A. (1989). The marriage license as a hitting license: A comparison of
assaults in dating, cohabiting and married couples. Journal of Family Violence, 4, 161–180.
Tateno, A., Jorge, R. E., & Robinson, R. G. (2003). Clinical correlates of aggressive behavior after
traumatic brain injury. Journal of Neuropsychiatry and Clinical Neurosciences, 15, 155–160.
West, C. M. (2012). Partner abuse in ethnic minority and lesbian, gay, bisexual, and transgender
populations. Partner Abuse, 3, 336–357. https://doi.org/10.1891/1946-6518.104.22.1686.
Whitaker, D. J., Haileyesus, T., Swahn, M., & Saltzman, L. E. (2007). Differences in frequency of
violence and reported injury between relationships with reciprocal and nonreciprocal intimate
partner violence. American Journal of Public Health, 97, 941–947. https://doi.org/10.2105/
White, H. R., & Widom, C. S. (2003). Intimate partner violence among abused and neglected
children in young adulthood: The mediating effects of early aggression, antisocial personality,
hostility and alcohol problems. Aggressive Behavior, 29, 332–345.
Intimate Partner Violence Across the Lifespan: Dyadic Theory and Risk and... 25