Impact of witnessing abuse of their mother and childhood trauma on men’s perpetration of intimate partner violence in the cross-sectional UN Multi-country Study on Men and Violence in Asia and the Pacific

Full text

Impact of witnessing abuse of their mother and childhood trauma on men’s
perpetration of intimate partner violence in the cross-sectional UN
multi-country study on men and violence in Asia and the Pacic
Rachel Jewkes
a,b,c,*
, Ruchira Tabassum Naved
d
, Naeema Abrahams
a,e
,
Nwabisa Shai
a,c
, Leane Ramsoomar
a
, Bianca Dekel
a,b
, Andrew Gibbs
a,f
,
Jani N¨
othling
a
, Samantha Willan
a
, Esnat Chirwa
a,c
a
Gender & Health Research Unit, South African Medical Research Council, Pretoria, South Africa
b
Ofce of the President, South African Medical Research Council, Tygerberg, South Africa
c
School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Parktown, Johannesburg, South Africa
d
Gender Equity and Sexual and Reproductive Health and Rights Unit, International Centre for Diarrhoeal Diseases Research, Dhaka, Bangladesh
e
School of Public Health, Faculty of Health Sciences, University of Cape Town, South Africa
f
Department of Psychology, University of Exeter, Exeter, UK
ARTICLE INFO
Handling editor: A C Tsai
Keywords:
Witnessing abuse
Childhood trauma
Masculinities
Intimate partner violence
Structural equation modelling
ABSTRACT
Trauma exposure and witnessing intimate partner violence (IPV) in childhood are recognised risk factors for IPV.
Using the UN Multi-country Study on Men and Violence in Asia and the Pacic dataset, we describe the pathways
through which they inuence adult IPV perpetration.
Methods: In nine sites, from six countries, data were collected in a two-stage, randomly-selected household
survey, with one man aged 18–49 years interviewed per house. 8379 interviews were completed with ever
partnered men in Bangladesh, Cambodia, China, Indonesia, Papua New Guinea (Bougainville) and Sri Lanka. We
present a Structural Equation Model (SEM) to understand paths through which childhood trauma and witnessing
IPV impacted perpetration of physical or sexual IPV in adulthood.
Results: Among the men, 25.5% had witnessed IPV, 47.0% had perpetrated physical or sexual IPV. Both wit-
nessing IPV and childhood trauma elevated the likelihood of such perpetration. The SEM showed four paths from
witnessing IPV and childhood trauma to the latent variable for physical/sexual IPV perpetration. One was direct
and three indirect. Paths were mediated by food insecurity, depression, and a latent variable measuring gender
inequitable and anti-social masculinities. The masculinity variable indicators were drug use, harmful alcohol use,
bullying, gang membership, ghting with other men, having sex with a sex worker and having raped a non-
partner. The direct and indirect effects showed both childhood trauma and witnessing maternal IPV to be
important, but childhood trauma the more so.
Conclusions: Both childhood trauma and witnessing IPV were important in driving IPV perpetration, with in-
dependent effects, however, broader childhood trauma exposure was most strongly associated. The effects were
mediated by food insecurity, depression and gender inequitable and anti-social masculinities, all recognised risk
factors for IPV perpetration. Thus, gender transformative IPV prevention interventions that include mental
health and economic elements can mitigate the inuence of these key exposures.
1. Introduction
Violence against women substantially impacts the health and well-
being of women who experience it and has lifelong consequences for
exposed children (Ellsberg et al., 2008; García-Moreno C et al., 2014;
Oram et al., 2022). Intimate partner violence (IPV), that is physical,
sexual, emotional or economic violence against a current or ex-wife or
girlfriend/partner, is the most common form of violence against women
perpetrated globally. Children exposed to physical, sexual or emotional
abuse and neglect in childhood, are more likely to perpetrate or
* Corresponding author: Gender & Health Research Unit, South African Medical Research Council, Pretoria, South Africa.
E-mail address: rjewkes@mrc.ac.za (R. Jewkes).
Contents lists available at ScienceDirect
SSM - Mental Health
journal homepage: www.journals.elsevier.com/ssm-mental-health
https://doi.org/10.1016/j.ssmmh.2025.100391
Received 14 October 2024; Received in revised form 8 January 2025; Accepted 10 January 2025
SSM - Mental Health 7 (2025) 100391
Available online 16 January 2025
2666-5603/© 2025 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ).

experience IPV when they have their own relationships (Fulu E et al.,
2017; Gibbs et al., 2020; Jewkes, 2002; Jewkes R et al., 2017; Wadji
et al., 2024; Yount et al., 2018). There are several suggested mechanisms
underpinning such intergenerational cycling of violence. These are best
understood as operating and intersecting at multiple levels: individual,
peer, family and community. The theorisation is better developed for
boys with respect to future perpetration, than for understanding how
girls are more likely to experience violence themselves.
Factors driving intergenerational cycling that operate at an indi-
vidual level include those that cause developmental, cognitive and
mental health problems (Bayarri et al., 2011; Carpenter and Stacks,
2009; V. C. Fong et al., 2019; Lamers-Winkelman et al., 2012; Malamuth
et al., 1995; Roustit et al., 2009; Trocm´
e, 2010). In essence, children are
directly affected through trauma exposure and its effects on brain
morphology, neurophysiology and mental health. Children of
IPV-exposed mothers are also indirectly impacted by the IPV negatively
affecting their mother’s mental health, alcohol and substance use,
parenting practices and harsh parenting practices (Fulu E et al., 2017).
Whilst there is a need to further explicate the pathways, children who
have been exposed to violence in the home exhibit more externalising
problems than other children, including physical aggression, deance,
conduct disorder and hyperactivity (Vanessa C Fong et al., 2019; Jagasia
et al., 2024). These all can inuence the propensity for the use of
violence in their own relationships, or tolerance of violence experienced.
Exposed children also experience more depression and anxiety (Bayarri
et al., 2011; Lamers-Winkelman et al., 2012), as well as adjustment,
social behaviour, and intellectual/academic performance challenges
(Supol et al., 2021). All of which have been shown to increase the pro-
pensity for the use of violence in some circumstances or reduce
self-protective behaviours.
Children’s experiences of being raised in homes where their mother
is exposed to IPV are shaped by the impact of IPV experience on how
they are parented (Bancroft et al., 2012; Greene et al., 2020). Experience
of IPV may lower mothers’ condence and patience as parents, and
result in difculties with disciplining children without using harsh
practices. (Lapierre et al., 2018; Levendosky and Graham-Bermann,
2001), although some mothers develop resilience and a strong positive
focus on their children as a coping strategy (Fogarty et al., 2019; Greene
et al., 2020). Harsh parenting by IPV-exposed mothers (Fulu E et al.,
2017), often results in insecure mother/child attachment, causing chil-
dren to grow up fearing intimacy, exhibit hostility or aggression within
their relationships, or seeking intimacy in maladaptive ways (Bowlby,
1969; Bowlby, 1973, 1980; Marshall WL & Barbaree, 1990). When
older, insecurity may result in boys or men having more dismissive and
hostile attitudes towards women (Stirpe et al., 2006) and both men and
women distancing from emotional relationships (Ward et al., 1997).
Both of which increase conict and tensions in relationships and can
lead to violence. There is much less theorisation of fathering among IPV
perpetrators, however violent parents have been observed to encourage
their children to associate with more aggressive peers (Morris RE et al.,
2002) and are known to more often drink heavily with impact on their
families (Wilson et al., 2024).
Violence exposure in childhood can also impact subsequent exposure
to violence of different types, and IPV perpetration may be through
pathways mediated by peer aggression, integrally linked to adolescent
and young men’s ideals of masculinity. Children exposed to IPV have a
greater propensity towards use, and experience, of peer violence
(Jagasia et al., 2024; Karmaliani R et al., 2017) and to join gangs and
other anti-social peer groups (Casey et al., 2009; Jewkes et al., 2020;
Malamuth et al., 1995). Trauma-exposed and more aggressive young
men seem to both aggregate as friends, and become more similar in their
aggression, replacing unsatisfactory relationships at home, and often
rejection by other children, with strong, like-minded peer support
(Snyder J et al., 1997; Steglich et al., 2010). In a similar way, children
who hang out with delinquent peers, or join gangs, as they seek power,
respect and self-esteem through such connections, and engage in risky
and often illicit and anti-social practices (Mathews S et al., 2015; Merrin
GJ et al., 2015; Sutton, 2017). In these contexts, misogynistic and vio-
lent constructions of masculinity often hold sway, and power and
respect that is sought is often used at the expense of women (Bourgois,
1996; Jewkes R & Morrell, 2017; Steinberg, 2006).
Ideas and behaviours related to the use of violence and gender are
socially constructed, starting from childhood, and are strongly impactful
on later expressed norms and practices. Children learn about gender
norms and the use of violence against women from home, and carry such
learning into their future relationships (Bandura, 1977). They also learn
a range of related norms and practices about the use of alcohol and other
substances, both strongly related to IPV perpetration (Wilson et al.,
2024). Other structural factors that are important include the
socio-economic conditions in which children are raised, as generally it is
recognised that poverty is a driver of IPV, associated with lower
maternal education and parenting challenges, as well as traditional so-
cial norms on gender (Gibbs, Dunkle et al., 2020; Yount, James-Hawkins
et al., 2018; Greene, Haisley et al., 2020).
Despite research on intergenerational cycling of violence, there is
much about the pathways and mechanisms that are not well understood,
notwithstanding the fact that breaking the cycle is foundational for
ending IPV. The relative importance of the exposure to witnessing IPV
and childhood trauma have not been well established. There are
consistent ndings in relation to childhood trauma exposure, but nd-
ings in relation to maternal IPV exposure are less consistent (Fulu et al.,
2013a, 2013b; Jewkes R et al., 2017; Kimber et al., 2018). A previous
analysis of the UN Multi-country study in Asia and the Pacic dataset
showed for men an association between witnessing abuse of his mother
and subsequent perpetration of at least one type of IPV in per country
analyses in four of the six countries (Fulu et al., 2013a, 2013b). How-
ever, multinomial logistic regression was used for the analysis with ve
categories, which may have impacted the power to describe associa-
tions, further a pooled regression analysis was not presented (Fulu et al.,
2013a, 2013b). In some of the other datasets, especially earlier ones,
only witnessing maternal IPV was measured, without measurement of
exposure to other childhood trauma, and this may have explained the
focus on associations with witnessing maternal IPV rather than child-
hood trauma more generally (Wadji et al., 2024). It is also possible that
some of the conclusions based on datasets with lower IPV prevalence
may have been a consequence of there being lower power in the study to
detect an association with a smaller effect size (Jewkes et al., 2017). A
systematic review also concluded that many of the available datasets for
analyses of the relationship between witnessing maternal IPV and later
perpetration were of low quality and non-representative samples
(Kimber et al., 2018).
Our paper seeks to extend understanding of intergenerational cycling
of violence, examining both the role of childhood trauma exposure and
witnessing maternal IPV, in a large multi-country, representative sample
survey. To maximise the power for the analysis, the full multi-country
dataset has been analysed, in contrast to the approach of Fulu (Fulu
et al., 2013a, 2013b). The objectives are to determine whether wit-
nessing maternal IPV was an independent risk factor for lifetime IPV
perpetration in the pooled dataset, after adjusting for childhood trauma
exposure, and to determine the pathways of impact, mediating factors
and comparable size of the effects of the two exposures.
The dataset of the UN Multi-country Study on Men and Violence
provided an opportunity to explore some of the hypothesised associa-
tions and pathways, despite not all the variables needed to explore the
pathways described being in the dataset. Within the limitations of the
dataset, our paper presents a secondary analysis of data and seeks to
describe the proportion of men who had witnessed abuse of their mother
as a child and experienced childhood trauma, the social and de-
mographic characteristics, depression and behaviours associated with
having witnessed abuse and perpetrated IPV, and the paths through
which having witnessed abuse of their mother and experience of
childhood trauma impact men’s IPV perpetration.
R. Jewkes et al.
SSM - Mental Health 7 (2025) 100391
2

2. Methods
We conducted a secondary analysis of data from the UN Multi-
country Study of Men and Violence in Asia and the Pacic (Fulu E
et al., 2013). The study was developed by Partners for Prevention, a
United Nations Development Programme (UNDP), United National
Population Fund (UNFPA), UN Women and United Nations Volunteers
(UNV) regional joint programme for gender-based violence prevention
in Asia and the Pacic, in collaboration with the South African Medical
Research Council and the country research teams. The rst author was
lead Technical Adviser for the multi-country study and Principal
Investigator in Bougainville, Papua New Guinea. The second author was
Principal Investigator in Bangladesh. Field research was conducted in
2011–12 in nine diverse sites in six countries in Asia and the Pacic:
Bangladesh, China, Cambodia, Sri Lanka, Indonesia and Papua New
Guinea (Bougainville). The sample was representative of Cambodia and
the island of Bougainville. In Bangladesh there was a site in the capital
city, and one provincial site. In Indonesia, there was a site in Jakarta and
a rural site (Java) and site with peri-urban and rural settlements
(Papua). The Chinese site was a county with a town and rural area, and
in Sri Lanka, Colombo and three contrasting districts, were surveyed.
Full details on the study methods, sampling, response rates, question-
naires and the ndings on intimate partner violence (IPV) in these same
study sites are presented elsewhere (Fulu et al., 2013a, 2013b). The sites
differed greatly. They were culturally and religiously diverse, and
differed in their socio-economic prole, housing conditions, political
and educational systems, access to services and information, urbanity,
family/kinship structures and in their history of civil conict. All these
factors inuence normative expectations of what it means ‘to be a man’
in a given society and the possibilities of achieving these expectations,
but also various stress factors that may impact on IPV perpetration.
A two-stage sampling strategy was used to identify census enumer-
ation areas, with a probability proportionate to size, and randomly
selected households within these areas. In each household, a man aged
18–49 years (where necessary, randomly selected) was invited for an
interview, with a trained male interviewer. The sample comprised
mostly heterosexual men, but this was not an inclusion requirement.
Most interviews were face-to-face, but answers to most sensitive ques-
tions were provided by self-completion on Audio-enhanced Personal
Digital Assistants (APDAs). In China a household list of individuals in
each cluster by age and sex was available and so used for sampling
within selected clusters and the entire questionnaire was self-completed.
10 178 men were interviewed, between 799 and 1776 per site. Among
these 8379 had ever been partnered. The proportion of enumerated and
eligible men interviewed per site varied between 59% and 95%, but only
in urban Bangladesh and Sri Lanka was it below 80% (Fulu et al., 2013a,
2013b). The variables used in the analysis are presented in Table 1.
2.1. Statistical analysis
The datasets were combined using Stata 13.0 and analysed using
Stata 17.0. Procedures took into account the multi-stage structure of the
dataset, with stratication by site within a country and enumeration
areas as clusters. Variables were summarised as percentages (or means),
with 95% condence limits calculated using standard methods (Taylor
linearisation). To show associations between independent variables that
were putative risk factors and outcomes, we rst present a series of
adjusted analyses (one per variable) with witnessing IPV as a child as an
independent variable for each analysis, all adjusted for country. Lifetime
physical or sexual IPV perpetration was the outcome in further analyses.
These are adjusted for country, but otherwise are bivariable analyses.
Multivariable logistic regression (in one case in Table 3, linear regres-
sion) was used. The variables selected mapped on to the known drivers
of IPV (Gibbs et al., 2020) to the extent that these were available in the
dataset.
Structural Equation Modelling (SEM) was conducted using MPlus to
assess the interrelationship between variables associated with a latent
variable for physical and/or sexual IPV. The model outcome was a
lifetime IPV latent variable comprising a summative score of physical
and sexual IPV questions. The gender inequitable and anti-social
Table 1
Operational denitions of variables.
Explanatory variables Denition
Social characteristics
Ever partnered Derived from two questions asking about having
been ever married and ever having a girlfriend
No high school Only attended primary school or had no formal
schooling.
Current food insecurity Sometimes or often people at home go without food
because of a lack of money
Employment status Single item asking if he was employed, had been
previously but not in the last year or had never
worked
Victimization history and childhood
Childhood trauma Men were asked about their experiences of trauma
before the age of 18 using a modied version of the
Childhood Trauma Questionnaire.
37,38
and whether
they happened never, sometimes, often or very
often: experienced emotional abuse as a child (e.g.
was insulted or humiliated by someone in his family
in front of other people); experienced neglect as a
child (e.g. parents were too drunk or drugged to take
care of him); experienced physical abuse as a child
(e.g. beaten at home with a belt or stick or whip or
something else which was hard); experienced sexual
abuse as a child (e.g. had sex with someone because
he was threatened or frightened or forced). Items
used as ‘ever’ exposed, and as a summative score of
responses to questions.
Witnessed abuse of his
mother
Before age 18 the respondent saw or heard his
mother being beaten by her husband or a boyfriend.
Used as a binary variable.
Teased or bullied others as a
child
Single item asking if the respondent ‘bullied, teased
or harassed’ others when a child. Used as a bInary
variable.
Psychological factors and
substance abuse

Depression 20 item CES-D depression measure of symptoms in
the past week cut point of 16+taken as indicating a
high level of depressive symptomatology.
Alcohol problems Based on AUDIT scale: frequency of drinking
currently, number of drinks usually consumed,
frequency of binge drinking (6+drinks), and
feelings of guilt or remorse after drinking and failure
to do what was normally expected from you because
of drinking in the past year [43].
Engagement in violence outside the home and drug use
Involvement in gangs Has ever participated in a gang
Involved in ghts with
weapons
Has ever been involved in a ght with a knife, gun or
other weapon
Past year drug use Has ever used drugs in the last 12 months
Ever sex with sex worker Respondent has ever had sex with a male, female or
transgender sex worker
Violence against women:
Non-partner rape ever 2 items asked about having ‘forced a woman who
was not your wife or girlfriend at the time to have
sex’ or having ‘had sex with a woman who was too
drunk or drugged to indicate whether she wanted
it’. Two more asked about having done these with
other men. A binary variable was derived from
responses categories as ever v. never
Physical and sexual IPV 5 items on physical intimate partner violence
perpetration ever were South African adaptations
for men from the WHO Multi-country Study
measure
2, 14
. 2 items on sexual violence against a
partner asking about ‘forced sex’ or sex ‘forced when
he knew she didn’t want it but believed she should
agree because she was his wife/partner’ (the second
question was not included in Bangladesh). Each had
never, once, few, many response options. Each
variable was derived by summing the responses.
R. Jewkes et al.
SSM - Mental Health 7 (2025) 100391
3

masculinity latent variable was constituted from binary variables for
substance abuse, problem drinking, bullying, gang membership, ghting
with a weapon, having had sex with a sex worker and having raped a
non-partner. With the exception of bullying and problem drinking, these
variables were used in denition of latent classes in this dataset in a
previous analysis (Jewkes et al., 2020). Having bullied and problem
drinking are well recognised practices among more gender inequitable
men and strongly associated with different latent classes in the previous
analysis. A measurement model was tted, with the variables allowed to
freely correlate. To assess model t of the data, we used the comparative
t index (CFI) (>0.95); Tucker-Lewis Index (TLI) (>0.9) for acceptable
t and (>0.95) as indicative of good t (Tucker LR & Lewis, 1973); and
root mean square error of approximation (RMSEA) (of 0.05 or less)
(Brown, 2006; Steiger, 1990).
We tted a SEM, built based on theory and statistically meaningful
modications, from the exogenous variables to IPV and used backwards
elimination to exclude endogenous variables that did not mediate any
path (with signicance set at the p <.05 level) from the exogenous
variables to IPV, to ensure model parsimony. Model t was good
(Goodness of t: RMSEA 0.045, CFI 0.95, TLI 0.90). The model included
error covariances among the variables comprising the gender inequi-
table and anti-social masculinities latent variable. The indicators
comprising the latent variable reected men’s practices that have been
shown to track together in previous analyses, in this, and other, datasets
using latent class analysis (Jewkes R & Morrell, 2017; Jewkes et al.,
2020). All models were run using a full-information maximum likeli-
hood method to deal with missing values. This method was chosen over
multiple imputations as it has been shown to yield superior results in
structural equation modelling (Enders CK & Bandalos, 2001). The model
presented is not adjusted for age, because the dataset lacked an ideal
variable as age was measured as a categorical variable in two countries,
and continuous in others. We tested the model with the categorical
variable for age, and found it had minimal impact on the direct and
indirect effects, and resulted in a small change to the t statistics (CFI =
0.94, TLI =0.89 & RMSEA =0.045), suggesting that the t was some-
what less good, likely because age was a categorical variable. We present
both standardised and unstandardised effects with their 95% Condence
Intervals.
2.2. Ethics
We followed ethical and safety guidelines for research with men on
perpetration (Jewkes R et al., 2012). To protect men providing sensitive
disclosure, we presented the study as a family and health study. The
interviewees received an information sheet and provided signed
informed consent. We did not retain identifying details of respondents.
Ethical approval was provided by the Medical Research Council of South
Africa; the College of Humanities, Beijing Forestry University; Interna-
tional Centre for Diarrhoeal Disease Research, Bangladesh; National
Ethics Committee for Health Research of Cambodia; the Faculty of
Medicine at the University of Colombo, Sri Lanka; and the Faculty of
Medicine, Gadjah Mada University, Indonesia. The funding sources had
no role in study design; in the collection, analysis, and interpretation of
data; in the writing of the report; and in the decision to submit the paper
for publication.
3. Results
Across all countries, 25.0% of men disclosed having witnessed abuse
of their mother in their childhood (Table 2). The proportion ranged from
8.5% in Rural Java, Indonesia to 56.5% in Bougainville, Papua New
Guinea. The background characteristics of the pooled sample are shown
in Table 3. In all, 51.7% of the men were under 35 years. 90.7% were
currently married or cohabiting. Two-thirds (67.7%) had attended high
school and 88.5% were currently working. Men who had witnessed
abuse in their childhood were younger than those not having done so,
and less likely to have attended high school and to have ever worked.
More were food insecure, depressed and drank alcohol harmfully. Many
of the men had experienced abuse or neglect in childhood, with 53.7%
having experienced some parental neglect, 48.3% some emotional
abuse, 54.1% some physical abuse and 17.0% sexual abuse. Men who
had witnessed maternal IPV in childhood were much more likely to
report exposure to all forms of child abuse. 32.9% of men had ever
perpetrated physical IPV and 24.3% disclosed perpetration of sexual
IPV. IPV perpetration was signicantly more common among men who
had witnessed maternal IPV in childhood (physical IPV aOR 2.53 (95%
CI 2.24, 2.85) and sexual IPV aOR 1.87 (95%CI 1.65, 2.12).
Table 4 shows a bivariable analysis of factors associated with
perpetration of physical and sexual IPV. Perpetration of physical IPV
was more strongly associated with being older, whereas sexual IPV was
not differentially reported by age. Having attended or completed high
school was associated with a lower likelihood of disclosing physical IPV
perpetration, but did not inuence the likelihood of disclosing sexual
IPV. The wife being more educated than her husband increased the
likelihood of men reporting sexual IPV perpetration. Experience of
current food insecurity was associated with a greater likelihood of
perpetration of both physical and sexual IPV. Both witnessing abuse of
their mother and their own experience of childhood trauma increased
the risk of perpetrating physical (aOR 2.52 (95% CI 2.25, 2.82) and
sexual IPV (aOR 1.94 (95% CI 1.73, 2.19). There was a positive asso-
ciation between childhood trauma scores and physical IPV perpetration
(aOR 1.24 (95% CI 1.22, 1.26)) and sexual IPV perpetration (aOR 1.19
(95% CI 1.17, 1.21)). Perpetration of both physical and sexual IPV were
associated with a greater likelihood of depression and reported alcohol
and drug abuse. IPV perpetration was also associated with a greater
likelihood of having bullied others, fought with weapons and been in a
gang, as well as a much higher likelihood of having also perpetrated
non-partner rape, and of having had sex with a sex worker.
3.1. Model of paths from men witnessing abuse of their mother to
perpetration of past year IPV
The SEM model had two endogenous variables, namely witnessing
their mother being beaten and other childhood trauma experiences
(Fig. 1). These covaried. Each variable had a direct path to IPV perpe-
tration. The standardised coefcients for these direct paths were 0.093
(95% CI 0.059, 0.126) for witnessing IPV and 0.171 (95% CI 0.131,
0.212) for childhood trauma exposure. Examination of the total effects
(direct plus indirect paths), showed that the impact of witnessing
maternal IPV was about 40% of that of all other childhood trauma
exposure (standardised coefcient 0.195 (95%CI 0.161, 0.229) for
witnessing IPV v. 0.510 (95%CI 0.475, 0.545) for childhood trauma),
indicating that it was a very important additional exposure in under-
standing IPV perpetration by men. The model showed that both
Table 2
Prevalence of witnessing maternal IPV by country and site of the ever-partnered
men.
N =8379 Witnessed IPV N
=2098
Did not witness
IPV N =6109
Witnessed abuse of
mother as a child
2134
(25.5%)
 
Country/setting   
Bangladesh urban 830 197 (23.7%) 633 (76.3%)
Bangladesh rural 742 256 (34.5%) 486 (65.5%)
Cambodia 1473 347 (23.6%) 1126 (73.4%)
China 965 204 (21.1%) 761 (78.9%)
Indonesia: Urban
Jakharta
820 75 (9.2%) 745 (90.8%)
Rural Java 768 65 (8.5%) 703 (91.5%)
West Papua 854 210 (24.6%) 644 (75.4%)
Papua New Guinea 738 417 (56.5%) 321 (43.5%)
Sri Lanka 1154 363 (31.5%) 791 (68.5%)
R. Jewkes et al.
SSM - Mental Health 7 (2025) 100391
4

witnessing their mother being beaten and other childhood trauma ex-
periences, had a strong positive association with IPV perpetration
through paths mediated by combinations of adult food insecurity,
depression and embracing practices associated with gender inequitable
and anti-social masculinities. For both witnessing IPV and other child-
hood trauma experiences, there were indirect paths only mediated by
current food insecurity, one by depression in the past week, and the
others were mediated by the gender inequitable and anti-social mascu-
linities latent variable, both as the sole mediator, and in combination
with depression, or food insecurity and depression. The direct effects
indirect effects and total effects are presented in Table 5. The analysis
shows that the highest total effects for childhood experiences were for
the path from childhood trauma to IPV perpetration (standardised co-
efcient 0.510 (95%CI 0.475,0.545), however the path from witnessing
IPV in childhood to adult IPV perpetration was almost two-fths of this
(38.2%) (standardised coefcient 0.195 (95%CI 0.161, 0.229). The table
shows that witnessing IPV in childhood had a modest inuence on food
insecurity (standardised coefcient 0.046 (95%CI 0.023, 0.069)) and
depression (standardised coefcient 0.055 (95%CI 0.033, 0.077)),
compared to the strong impact of childhood trauma exposure on both.
The impact of witnessing IPV in childhood on gender inequitable and
anti-social masculinities was larger (standardised coefcient 0.138
(95%CI 0.113, 0.162)), and a third (32.2%) of the standardised coef-
cient (0.429 (95% CI 0.405, 0.453) for childhood trauma exposure.
4. Discussion
In this very large, multi-country study, across the sites in the six
countries, a quarter of men disclosed having witnessed maternal IPV
when they were children. This proportion is higher than the estimate
from a systematic review (16%) that included many studies from higher
income populations (Whitten et al., 2024). We have shown that both
men who witnessed IPV as children, and those who experienced, the
often co-occurring, exposure of childhood trauma, were at much higher
risk of perpetrating IPV. The SEM has shown that both witnessing IPV in
childhood and experiencing other childhood trauma, such as emotional
abuse and neglect, physical abuse and sexual abuse, directly elevated the
likelihood of perpetrating IPV as an adult, as well as impacting through
indirect pathways. The total effects of childhood trauma exposure were
larger than those for witnessing IPV, which is not surprising given the
large and diverse number of forms of trauma measured in the childhood
trauma variable. Nonetheless the analysis has shown that witnessing IPV
in childhood is an important independent exposure which elevated the
IPV perpetration risk in this large multi-country sample of men.
It is likely that, at least at some level, the direct pathway from wit-
nessing IPV in childhood to IPV perpetration reects the social learning
of the use of violence (social learning theory) (Bandura, 1977), although
the path may also be mediated by unmeasured psychological health
variables and contextual mediators. The mediators of the pathways –
food insecurity, depression and gender inequitable and anti-social
masculinities - were variables that are well recognised as associated
with perpetration of IPV, in this, and other datasets with men (Fulu
et al., 2013a, 2013b; Gibbs et al., 2020; James-Hawkins et al., 2018).
The analysis has shown that present day food insecurity mediated paths
from both witnessing IPV and experiencing childhood trauma to
perpetration of IPV. The role of poverty as a risk factor for violence in
general, and specically IPV, is now well recognised, and it is likely to
explain a substantial part of the high prevalence of witnessing IPV in
childhood in lower resource settings. Interventions to alleviate poverty
have an important role to play in IPV prevention (Gibbs et al., 2020;
Kerr-Wilson A et al., 2020). Unfortunately, there were no indicators of
socio-economic status in childhood in the dataset, but the role of poverty
in elevating risk of childhood trauma is also well established (Hatcher
et al., 2019). Thus, it is likely that the present food insecurity for many
men would have been a continuation of a lifetime exposure to economic
precarity.
Table 3
Prevalence of social characteristics, mental health, IPV perpetration and child-
hood trauma exposures of ever-partnered men, with adjusted associations be-
tween witnessing IPV in childhood and these factors (modelled as outcomes).
N =
8379
Witnessed
IPV N =
2098
Did not
witness
IPV N =
6109
aOR (95%
CI)*
p-value
Independent
variables
    
Age group     
18–24 yrs 1409
(16.8%)
408
(19.1%)
994
(16.0%)
ref 
25–34yrs 2924
(34.9%)
792
(37.1%)
2120
(34.1%)
0.84 (0.72,
0.98)
0.023
35–49yrs 4046
(48.3%)
934
(43.8%)
3096
(49.9%)
0.67 (0.58,
0.78)
<0.0001
Currently
married or
cohabiting
7565
(90.7%)
1915
(89.8%)
5650
(91.0%)
0.89 (0.75,
1.06)
0.185
Attended high
school or
higher
5645
(67.7%)
1319
(61.8%)
4326
(69.7%)
0.87
(0.77,0.97)
0.016
Employment
status
    
working now 7381
(88.5%)
1834
(86.0%)
5547
(89.3%)
ref** 
not in the last
12 m
493
(5.9%)
168 (7.9%) 325
(5.2%)
1.24 (0.99,
1.57)
0.063
Never worked 469
(5.6%)
131 (6.1%) 338
(5.4%)
1.31
(1.03,1.66)
0.027
Ever without
food due to
lack of
money
2418
(29.8%)
827
(39.7%)
1591
(26.4%)
1.59 (1.40,
1.80)
<0.0001
Depression
score
13.98
(SD
5.36)
14.48 (SD
6.31)
13.81
(SD
4.97)
1.76 (1.46,
2.08)***
<0.0001
Has problem
drinking,
based on
modied
AUDIT
979
(12.1%)
391
(18.8%)
588
(9.8%)
1.96 (1.72,
2.24)
<0.0001
Experienced
parental
neglect as a
child
4468
(53.7%)
1594
(75.0%)
2874
(46.4%)
3.26 (2.87,
3.70)
<0.0001
Experienced
emotional
abuse as a
child
4026
(48.3%)
1495
(70.1%)
2531
(40.8%)
3.39 (3.00,
3.85)
<0.0001
Experienced
physical
abuse as a
child
4500
(54.1%)
1532
(72.0%)
2968
(48.0%)
3.31 (2.90,
3.77)
<0.0001
Experienced
sexual abuse
as a child
1416
(17.0%)
685
(32.1%)
731
(11.8%)
3.01 (2.63,
3.45)
<0.0001
Childhood
trauma
score
15.67
(SD
3.04)
17.69 (SD
3.48)
14.97
(SD
2.52)
1.35 (1.32,
1.38)
<0.0001
Lifetime
physical IPV
2697
(32.9%)
1067
(51.2%)
1630
(26.7%)
2.53 (2.24,
2.85)
<0.0001
Lifetime
sexual IPV
1983
(24.3%)
726
(34.6%)
1257
(20.7%)
1.87 (1.65,
2.12)
<0.0001
     
* adjusted for
country
    
**
multinomial
regression
model
    
*** coefcient
from linear
regression
    
R. Jewkes et al.
SSM - Mental Health 7 (2025) 100391
5

Depression is often associated with IPV perpetration (Nduna et al.,
2010; Oram et al., 2022), and was the only form of adverse mental
ill-health outcome measured in the dataset. Other forms of psychological
distress, particularly commonly post-traumatic stress disorder (PTSD),
complex PTSD and the adverse effects of childhood trauma exposure on
personality development, are also well recognised risk factors for IPV
perpetration, but these were not measured in the UN multi-country
study (Oram et al., 2022; Ramsoomar et al., 2023). Harmful alcohol
drinking is also an established risk factor for IPV perpetration and
interestingly this pertained in our dataset, which includes many coun-
tries that have a very low prevalence of harmful drinking due to strict
laws and religious prohibitions (Ramsoomar et al., 2021). The analysis
also showed that men who had witnessed IPV, and been exposed to other
childhood trauma types, had their risk of IPV perpetration mediated by
their adoption of practices associated with more gender inequitable and
anti-social masculinities. Studies from a range of other settings have
similarly shown that men who perpetrate IPV, often have a cluster of risk
factors linked to a more violent, gender inequitable and anti-social
masculinity, which is also linked to traumatic experiences in child-
hood (Bourgois, 1996; Gibbs et al., 2018; Jewkes R & Morrell, 2017).
Previous analyses of the dataset have shown that men who had perpe-
trated IPV were more likely to have engaged in a range of behaviours
that are associated with the use of violence (Jewkes et al., 2020). These
included bullying, gang afliation and the related practices of having
fought with a weapon, and having had relations with women in which
violence or lack of consent played a key role, and men seeking sex with
Table 4
Bivariate analysis of factors associated with physical and sexual IPV perpetration by ever partnered men from countries in Asia and the Pacic.
Perpetrated Physical IPV Perpetrated
sexual IPV
  Yes (N =
2704)
No (N =
5528)
  Yes (N =2262) No(N =5940)  
Factor Level n (%)/mean
(sd)
n (%)/mean
(sd)
aOR (95%CI) p-value n (%)/mean (sd) n (%)/mean
(sd)
aOR (95%CI) p-value
Age group 18–24 yrs 314 (23.8%) 1006 (76.2%) Ref 429 (31.0%) 953 (69.0%) Ref 
25–34yrs 920 (31.9%) 1968 (68.1%) 1.41
(1.21,1.66)
<0.001 834 (29.0%) 2039 (71.0%) 1.06
(0.91,1.23)
0.438
35–49yrs 1470 (36.5%) 2554 (63.5%) 1.58
(1.36,1.84)
<0.001 999 (25.3%) 2948 (74.7%) 0.91
(0.79,1.05)
0.213
Attended high school or
higher
No 1035 (38.8%) 1633 (61.2%) Ref 741 (27.9%) 1912 (72.1%) Ref 
Yes 1667 (30.0%) 3894 (70.0%) 0.78
(0.7,0.87)
<0.001 1519 (27.4%) 4027 (72.6%) 1.01
(0.89,1.13)
0.933
Employment status working now 2414 (33.0%) 4895 (67.0%) Ref 1928 (26.6%) 5320 (73.4%) Ref 
not in the last
12 m
174 (35.8%) 312 (64.2%) 0.99
(0.8,1.22)
0.911 187 (38.3%) 301 (61.7%) 1.10 (0.9,1.36) 0.357
Never worked 116 (26.6%) 320 (73.4%) 0.98
(0.78,1.24)
0.891 147 (31.6%) 318 (68.4%) 1.04
(0.84,1.29)
0.693
Experienced food
insecurity
No 1639 (29.2%) 3970 (70.8%) Ref 1433 (25.2%) 4257 (74.8%) Ref 
Yes 985 (41.4%) 1397 (58.6%) 1.66
(1.48,1.86)
<0.001 805 (33.3%) 1616 (66.7%) 1.75
(1.55,1.98)
<0.001
Childhood trauma score 16.9 (3.4) 15.1 (2.6) 1.24
(1.22,1.26)
<0.001 16.9 (3.5) 15.2 (2.7) 1.19
(1.17,1.21)
<0.001
Witnessed abuse of
mother as child
No 1630 (26.7%) 4482 (73.3%) Ref 1445 (23.7%) 4642 (76.3%) Ref 
Yes 1067 (51.1%) 1022 (48.9%) 2.52
(2.25,2.82)
<0.001 817 (38.8%) 1286 (61.2%) 1.94
(1.73,2.19)
<0.001
Depression score 16.3 (5.4) 14.2 (4.6) 1.08
(1.07,1.09)
<0.001 16.2 (5.3) 14.4 (4.7) 1.08
(1.07,1.10)
<0.001
Has problem drinking No 2156 (30.8%) 4854 (69.2%) Ref 1690 (23.8%) 5420 (76.2%) Ref 
Yes 452 (47.2%) 505 (52.8%) 2.47
(2.12,2.88)
<0.001 541 (55.2%) 439 (44.8%) 2.75
(2.37,3.19)
<0.001
Any past year drug use No 2317 (32.0%) 4919 (68.0%) Ref 1931 (26.3%) 5410 (73.7%) Ref 
Yes 308 (40.1%) 461 (59.9%) 1.83
(1.53,2.2)
<0.001 311 (39.4%) 478 (60.6%) 2.14
(1.79,2.57)
<0.001
Has teased or bullied
others
No 1683 (28.2%) 4289 (71.8%) Ref 1310 (21.9%) 4661 (78.1%) Ref 
Yes 1011 (45.5%) 1211 (54.5%) 2.03
(1.82,2.27)
<0.001 951 (42.7%) 1277 (57.3%) 2.03
(1.82,2.27)
<0.001
Has participated in a
gang
No 2283 (31.4%) 4991 (68.6%) Ref 1792 (24.3%) 5587 (75.7%) Ref 
Yes 347 (46.6%) 397 (53.4%) 1.98
(1.67,2.34)
<0.001 455 (59.6%) 309 (40.4%) 3.45
(2.93,4.06)
<0.001
Has been in ght with a
weapon
No 2158 (30.8%) 4852 (69.2%) Ref 1700 (23.9%) 5400 (76.1%) Ref 
Yes 473 (47.0%) 533 (53.0%) 2.2 (1.9,2.56) <0.001 546 (52.4%) 495 (47.6%) 2.68
(2.33,3.09)
<0.001
Has perpetrated non-
partner rape
No 2079 (29.5%) 4965 (70.5%) Ref 1497 (20.9%) 5658 (79.1%) Ref 
Yes 549 (56.6%) 421 (43.4%) 3.15
(2.7,3.67)
<0.001 752 (76.1%) 236 (23.9%) 9.04
(7.68,10.64)
<0.001
Has had sex with a sex
worker
No 2077 (30.9%) 4647 (69.1%) Ref 1644 (24.2%) 5156 (75.8%) Ref 
Yes 520 (43.2%) 684 (56.8%) 1.92
(1.67,2.2)
<0.001 583 (47.5%) 644 (52.5%) 3.16
(2.76,3.61)
<0.001
aOR =adjusted for country.
R. Jewkes et al.
SSM - Mental Health 7 (2025) 100391
6

sex workers. In our analysis we have extended previous quantitative
research on masculinities by showing that it was possible to derive a
latent variable indicative of the spectrum of hegemonic masculinity and
hypermasculinity (Herek, 1987) that was built from these measured
indicator practices. This has methodologically extended our under-
standing of the application of quantitative methods to understanding
masculinities, and the role of men embracing different masculine iden-
tities and associated practices in their risk of IPV perpetration.
5. Limitations
The exposure variable witnessing abuse of their mother was a single
item variable ‘Before I reached 18, I saw or heard my mother being
beaten by her husband or boyfriend’. We recognise that there may have
been some physical IPV witnessed that would not have been reported
because the word ‘beaten’ was used (e.g. hair pulling, slaps), that there
may have been IPV experienced by men’s mothers that the men may not
have known about, and/or disclosed in the interview, including sexual
IPV, that may have nonetheless impacted their mother in important
ways (such as through impacting her mental health and parenting).
Further that the question did not ask about witnessing emotional IPV
and controlling behaviour. This will have resulted in an under-
estimation of the proportion of men who had witnessed and IPV
directed against their mother. We can only speculate on its impact on the
ndings. It is also likely that with extended families, children may
witness a range of IPV perpetrated against other family members, as well
as violence from in-laws directed at women (Jewkes et al., 2019), all of
which would impact social norms and learning around violence. Future
research on the impact of witnessing IPV needs to employ a broader
measure of the exposure. The study ndings reect the sampled sites,
whether one can generalise beyond these is unclear. The combined
dataset is not intended to reect the whole Asia Pacic region. Since the
research was cross-sectional, temporality may be questioned. The key
exposures were ‘before age 18’ and most IPV would have occurred after
age 18, thus clarifying the temporal relationship in respect of these
variables. However, depression and food insecurity were measured for
the recent period. Which raises a question of whether they were recent
problems and may have followed IPV perpetration. We recognise that
there is uncertainly here about the temporal sequence and are limited by
the variables in the dataset. However, depression is generally recognised
as a chronic problem, especially in the absence of available therapy,
which would generally have been the case for the men in the study.
Thus, it is reasonable to assume that for many men current depressive
symptoms would be a reasonable indicator of higher levels of prior
depression symptomatology. Furthermore, it is reasonable to assume
limited social mobility and thus there is a strong likelihood that food
insecurity in the current period would reect an enduring problem. They
are well established risk factors for IPV perpetration, which adds to our
condence about their likely role as mediators. All the indicators of
masculinity are recognised as being temporarily co-manifest with IPV in
other datasets.
6. Conclusions
We have shown that exposure to childhood trauma and exposure to
maternal IPV in childhood are both associated with a greater risk of
lifetime perpetration of IPV in adulthood by men in Asia and the Pacic
region. They impact the likelihood of perpetration, both directly, and
independently, as well as impacting adult food insecurity, depression
and the likelihood of men embracing more gender inequitable and anti-
social masculinities. The ndings are compatible with explanations of
direct impact of witnessing maternal IPV through social learning, as well
as consequences of growing up with adversity in childhood manifest in
self-esteem, personality and mood, and resulting in men embracing
gender inequitable and anti-social masculinities, and relatedly using
violence towards their wife or girlfriend. These pathways are all
amenable to impact through gender transformative interventions that
also seek to strengthen mental health and reduce poverty. The direct
pathways are more challenging and require intervention with exposed
children, and other family members, to reap benets through reductions
in IPV experienced/perpetrated in future generations of adults.
CRediT authorship contribution statement
Rachel Jewkes: Writing – original draft, Project administration,
Methodology, Investigation, Formal analysis, Data curation, Conceptu-
alization. Ruchira Tabassum Naved: Writing – review & editing, Su-
pervision, Project administration, Methodology, Investigation,
Conceptualization. Naeema Abrahams: Writing – review & editing,
Conceptualization. Nwabisa Shai: Writing – review & editing, Investi-
gation. Leane Ramsoomar: Writing – review & editing, Conceptuali-
zation. Bianca Dekel: Writing – original draft, Conceptualization.
Andrew Gibbs: Writing – review & editing. Jani N¨
othling: Writing –
review & editing, Conceptualization. Samantha Willan: Writing – re-
view & editing, Conceptualization. Esnat Chirwa: Writing – review &
editing, Methodology, Formal analysis, Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing nancial
Fig. 1. Structural equation model of paths from witnessing maternal IPV and experience of childhood trauma to IPV perpetration.
R. Jewkes et al.
SSM - Mental Health 7 (2025) 100391
7

interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgements
The UN Multi-country Study on Men and Violence was a collabora-
tive effort and only made possible by the commitment, dedication, and
hard work of all the organisations and individuals involved, both
internationally and in each of the study countries. First and foremost, we
would like to acknowledge and give gratitude to the women and men
who gave their time to participate in our study and generously shared
their life experiences with us. We wish to express our profound thanks to
our partner institutions and organisations in each of the study countries,
as well as all the interviewers and supervisors who worked tirelessly,
and often under difcult circumstances, to collect the data for this study.
We would like to thank the studies’ technical advisors, members of the
steering committee, and members of the national working groups. We
would like to thank all the members of the Partner for Prevention
Technical Advisory Group and Regional Steering Committee who guided
the overall study and the SVRI for managing the availability of the
dataset. UN Multi-country Study on Men and Violence Study Team: Core
research team: Emma Fulu (P4P) (Study Coordinator), Rachel Jewkes
(Medical Research Council, South Africa), Xian Warner (P4P), Stephanie
Miedema (P4P), Tim Roselli (P4P), and James Lang (P4P). Country study
teams—China: Dr Wang Xiangxian (PI) (Tianjin University, China); Fang
Gang (Beijing Forestry University); Li Hongtao (Chinese Women’s Col-
lege and Anti-Domestic Violence Network); Zeljka Mudrovcic, Wen Hua,
Arie Hoekman, Elina Nikulainen, Bernard, Coquelin, and Mariam Khan
(UNFPA China); Cambodia: Wenny Kusuma, Clara Magari˜
no Manero,
and Freya Larsen (UN Women Cambodia); Emma Fulu (PI) and Xian
Warner (P4P); and Saba Moussavi (independent consultant); Sri Lanka:
Neloufer de Mel (PI) (University of Colombo); Pradeep Peiris (Social
Scientists’ Association); Shyamala Gomez (independent consultant);
Social Indicator Team; and Kamani Jinadasa (CARE Sri Lanka); Papua
New Guinea (Bougainville): Rachel Jewkes (PI), Yandisa Sikweyiya, and
Nwabisa Shai (Medical Research Council, South Africa); Francesca
Drapuluvik-Tinabar (National Statistics Ofce, PNG); Peterson Magoola
and Anthony Agyenta (UNDP PNG); Thomas Shanahan and Tracy
Vienings (UNDP Regional Pacic Centre). Steering committeeÐRachel
Jewkes (MRC, South Africa), Claudia Garcia-Moreno (WHO), Ruchira
Table 5
Structural equation model: Direct, indirect and total effects.
Goodness of Fit: CFI =0.945, TLI =0.901, RMSEA =0.045
Direct effects Indirect Effects Total Effects
Path StdCoef (95%
CI)
UnStdCoef
(95%CI)
p-value StdCoef (95%
CI)
UnStdCoef
(95%CI)
p-value StdCoef (95%
CI)
UnStdCoef
(95%CI)
p-value
Food insecurity → IPV
perpetration
0.079
(0.048,0.110)
0.181
(0.110,0.251)
<0.001 0.033
(0.025,0.040)
0.075
(0.058,0.091)
<0.001 0.112
(0.081,0.143)
0.255
(0.186,0.325)
<0.001
Depression score → IPV
perpetration
0.089
(0.056,0.123)
0.019
(0.012,0.026)
<0.001 0.089
(0.071,0.107)
0.019
(0.015,0.023)
<0.001 0.178
(0.144,0.213)
0.038
(0.03,0.045)
<0.001
Anti-SocialMasculinity → IPV
perpetration
0.679
(0.625,0.733)
9.398
(8.071,10.73)
<0.001    0.679
(0.625,0.733)
9.398
(8.071,10.725)
<0.001
Childhood trauma experience
→ IPV perpetration
0.171
(0.131,0.212)
0.059
(0.045,0.073)
<0.001 0.338
(0.308,0.369)
0.116
(0.105,0.128)
<0.001 0.510
(0.475,0.545)
0.175
(0.164,0.187)
<0.001
Witness abuse of mother as
child → IPV perpetration
0.093
(0.059,0.126)
0.222
(0.143,0.301)
<0.001 0.102
(0.083,0.121)
0.244
(0.199,0.290)
<0.001 0.195
(0.161,0.229)
0.466
(0.386,0.547)
<0.001
Childhood trauma experience
→ Food insecurity
0.209
(0.186,0.231)
0.031
(0.028,0.035)
<0.001    0.209
(0.186,0.231)
0.031
(0.028,0.035)
<0.001
Witness abuse of mother as
child → Food insecurity
0.046
(0.023,0.069)
0.048
(0.024,0.072)
<0.001    0.046
(0.023,0.069)
0.048
(0.024,0.072)
<0.001
Food insecurity → Depression
score
0.183
(0.163,0.204)
1.986
(1.764,2.207)
<0.001    0.183
(0.163,0.204)
1.986
(1.764,2.207)
<0.001
Childhood trauma experience
→ Depression score
0.301
(0.280,0.322)
0.491
(0.455,0.526)
<0.001 0.038
(0.032,0.044)
0.062
(0.053,0.072)
<0.001 0.339
(0.318,0.360)
0.553
(0.517,0.589)
<0.001
Witness abuse of mother as
child → Depression score
0.046
(0.025,0.068)
0.527
(0.284,0.771)
<0.001 0.008
(0.004,0.013)
0.096
(0.047,0.145)
<0.001 0.055
(0.033,0.077)
0.624
(0.376,0.871)
<0.001
Depression score → Anti-
SocialMasculinity
0.131
(0.107,0.155)
0.002
(0.002,0.002)
<0.001    0.131
(0.107,0.155)
0.002
(0.002,0.002)
<0.001
Childhood trauma experience
→ Anti-SocialMasculinity
0.385
(0.360,0.410)
0.010
(0.008,0.011)
<0.001 0.044
(0.036,0.053)
0.001
(0.001,0.001)
<0.001 0.429
(0.405,0.453)
0.011
(0.009,0.012)
<0.001
Witness abuse of mother as
child → Anti-
SocialMasculinity
0.130
(0.106,0.155)
0.023
(0.018,0.028)
<0.001 0.007
(0.004,0.010)
0.001
(0.001,0.002)
<0.001 0.138
(0.113,0.162)
0.024
(0.019,0.029)
<0.001
Latents         
IPV → Sexual IPV 0.604
(0.575,0.633)
1<0.001      
IPV → Physical IPV 0.461
(0.435,0.488)
0.882
(0.821,0.942)
<0.001      
Anti-SocialMasculinity→ Past
year drug use
0.255
(0.229,0.282)
1<0.001      
Anti-SocialMasculinity→
Teased/bullied others
0.579
(0.545,0.613)
3.409
(2.993,3.826)
<0.001      
Anti-SocialMasculinity→Gang
involvement
0.449
(0.425,0.473)
1.733
(1.538,1.928)
<0.001      
Anti-SocialMasculinity→ Fight
with weapon
0.419
(0.394,0.443)
1.852
(1.636,2.067)
<0.001      
Anti-SocialMasculinity→ Non-
partner rape
0.662
(0.641,0.683)
2.865
(2.536,3.194)
<0.001      
Anti-SocialMasculinity→ Sex
with sex worker
0.391
(0.367,0.415)
1.861
(1.632,2.09)
<0.001      
Anti-SocialMasculinity→
Problem drinking
0.42
(0.396,0.445)
1.816
(1.604,2.029)
<0.001      
R. Jewkes et al.
SSM - Mental Health 7 (2025) 100391
8

Tabassum Naved (ICDDR,B), Kamani Jinadasa (CARE Sri Lanka), Tracy
Vienings (UNDP Regional Pacic Centre) and Wenny Kusuma (UN
Women Cambodia). Technical advisory groupÐRachel Jewkes (MRC,
South Africa), Raewyn Connell (University of Sydney, Australia), Gary
Barker (Instituto Promundo, USA & Brazil), Alan Greig (Independent
consultant, USA), Rahul Roy (AAKAR, India), Ravi Verma (ICRW),
Kalyani Menon Sen (Independent consultant). PDA programmer: Scott
Johnson (University of Kentucky).
References
Bancroft, L., Silverman, J., Ritchie, D., 2012. Shock waves: the batterer’s impact on the
home. Author, the Batterer as Parent: Addressing the Impact of Domestic Violence on
Family Dynamics, pp. 69–105.
Bandura, A., 1977. Social Learning Theory. Prentice-Hall.
Bayarri, E., Ezpeleta, L., Granero, R., 2011. Exposure to intimate partner violence,
psychopathology, and functional impairment in children and adolescents: moderator
effect of sex and age. J. Fam. Violence 26 (7), 535–543.
Bourgois, P., 1996. In search of masculinity - violence, respect and sexuality among
Puerto Rican crack dealers in East Harlem. Br. J. Criminol. 36, 412–427.
Bowlby, J., 1969. Attachment and Loss. In: Attachment, vol. 1. Basic Books.
Bowlby, J., 1973. Attachment and Loss, vol. 2. Basic Books.
Bowlby, J., 1980. Attachment and Loss: Vol. 3 Loss, Sadness and Depression. Basic Books.
Brown, T., 2006. Conrmatory Factor Analysis for Applied Research. Guilford Press.
Carpenter, G.L., Stacks, A.M., 2009. Developmental effects of exposure to intimate
partner violence in early childhood: a review of the literature. Child. Youth Serv.
Rev. 31 (8), 831–839.
Casey, E.A., Beadnell, B., Lindhorst, T.P., 2009. Predictors of sexually coercive behaviour
in a nationally representative sample of adolescent males. J. Interpers Violence 24,
1129–1147.
Ellsberg, M., Jansen, H.A.F.M., Heise, L., Watts, C.H., Garcia-Moreno, C., Hlth, W.M.S.
W., 2008. Intimate partner violence and women’s physical and mental health in the
WHO multi-country study on women’s health and domestic violence: an
observational study. Lancet 371 (9619), 1165–1172.
Enders, C.K., Bandalos, D., 2001. The relative performance of full information maximum
likelihood estimation for missing data in structural equation models. Struct. Equ.
Model. 8 (3), 430–457.
Fogarty, A., Woolhouse, H., Giallo, R., Wood, C., Kaufman, J., Brown, S., 2019.
Promoting resilience and wellbeing in children exposed to intimate partner violence:
a qualitative study with mothers. Child Abuse Negl. 95, 104039.
Fong, V.C., Hawes, D., Allen, J.L., 2019. A systematic review of risk and protective
factors for externalizing problems in children exposed to intimate partner violence.
Trauma Violence Abuse 20 (2), 149–167. https://doi.org/10.1177/
1524838017692383.
Fulu, E., Jewkes, R., Roselli, T., Garcia-Moreno, C., 2013a. Prevalence of and factors
associated with male perpetration of intimate partner violence: ndings from the UN
Multi-country Cross-sectional Study on Men and Violence in Asia and the Pacic.
Lancet Global Health 1 (4), e187–e207.
Fulu, E., Warner, X., Miedema, S., Jewkes, R., Roselli, T., Lang, J., 2013b. Why Do Some
Men Use Violence against Women and How Can We Prevent it.
Fulu, E., Miedema, S., Roselli, T., Ko, Ling Chan E., Jewkes, R., 2017. Pathways between
violence against children and intimate partner violence and child maltreatment:
ndings from the cross-sectional UN Multi-country Study on Men and Violence in
Asia and the Pacic. Lancet Global Health 5 (5), e512–e522.
García-Moreno, C., Hegarty, K., Lucas d’Oliveira, A.F., Koziol-MacLain, J.,
Colombini, M., Feder, G., 2014. The health-systems response to violence against
women. Lancet. https://doi.org/10.1016/S0140-6736(14)61837-7.
Gibbs, A., Dunkle, K., Washington, L., Willan, S., Shai, N., Jewkes, R., 2018. Childhood
traumas as a risk factor for HIV-risk behaviours amongst young women and men
living in urban informal settlements in South Africa: a cross-sectional study. PLoS
One 13 (4), e0195369. https://doi.org/10.1371/journal.pone.0195369.
Gibbs, A., Dunkle, K., Ramsoomar, L., Willan, S., Jama Shai, N., Chatterji, S., Naved, R.,
Jewkes, R., 2020. New learnings on drivers of men’s physical and/or sexual violence
against their female partners, and women’s experiences of this, and the implications
for prevention interventions. Glob. Health Action 13 (1), 1739845. https://doi.org/
10.1080/16549716.2020.1739845.
Greene, C.A., Haisley, L., Wallace, C., Ford, J.D., 2020. Intergenerational effects of
childhood maltreatment: a systematic review of the parenting practices of adult
survivors of childhood abuse, neglect, and violence. Clin. Psychol. Rev. 80, 101891.
https://doi.org/10.1016/j.cpr.2020.101891.
Hatcher, A.M., Gibbs, A., Jewkes, R., McBride, R.S., Peacock, D., Christodes, N., 2019.
Effect of childhood poverty and trauma on adult depressive symptoms among young
men in peri-urban South African settlements. J. Adolesc. Health 64 (1), 79–85.
https://doi.org/10.1016/j.jadohealth.2018.07.026.
Herek, G., 1987. On heterosexual masculinity: some psychological consequences of the
social construction of gender and sexuality. In: Kimmel, M. (Ed.), Changing Men:
New Directions in Research on Men and Masculinity. Sage.
Jagasia, E., Bloom, I., Nelson, K.E., Campbell, J., 2024. Early adolescent development in
the face of violence: a systematic review running. Child Abuse Negl. 151, 106751.
https://doi.org/10.1016/j.chiabu.2024.106751.
James-Hawkins, L., Cheong, Y.F., Naved, R.T., Yount, K.M., 2018. Gender norms,
violence in childhood, and men’s coercive control in marriage: a multilevel analysis
of young men in Bangladesh. Psychol Violence 8 (5), 580–595. https://doi.org/
10.1037/vio0000152.
Jewkes, R., 2002. Intimate partner violence: causes and prevention. Lancet 359 (9315),
1423–1429.
Jewkes, R., Morrell, R., 2017. Hegemonic Masculinity, Violence and Gender Equality:
using latent class analysis to investigate the origins and correlates of differences
between men. Men Masculinities 21 (4), 547–571. https://doi.org/10.1177/109718
4X1769617.
Jewkes, R., Dartnall, E., Sikweyiya, Y., 2012. Ethical and Safety Recommendations for
Research on Perpetration of Sexual Violence.
Jewkes, R., Fulu, E., Naved, R.T., Chirwa, E., Dunkle, K., Haard¨
orfer, R., Garcia-
Moreno, C., 2017. Women’s and men’s reports of past-year prevalence of intimate
partner violence and rape and women’s risk factors for intimate partner violence: a
multicountry cross-sectional study in Asia and the Pacic. PLoS Med. 14 (9),
e1002381.
Jewkes, R., Corboz, J., Gibbs, A., 2019. Violence against Afghan women by husbands,
mothers-in-law and siblings-in-law/siblings: risk markers and health consequences
in an analysis of the baseline of a randomised controlled trial. PLoS One 14 (2),
e0211361. https://doi.org/10.1186/s12889-019-6673-9.
Jewkes, R., Jordaan, E., Myrttinen, H., Gibbs, A., 2020. Masculinities and violence: using
latent class analysis to investigate the origins and correlates of differences between
men in the cross-sectional UN Multi-country Study on men and violence in Asia and
the Pacic. J Glob Health 10 (2), 020439. https://doi.org/10.7189/jogh.10.020439.
Karmaliani, R., McFarlane, J., Somani, R., Khuwaja, H.M.A., Shehzad Bhamani, S., Saeed
Ali, T., Gulzar, S., Somani, Y., Chirwa, E.D., Jewkes, R., 2017. Peer violence
perpetration and victimization: prevalence, associated factors and pathways among
1752 sixth grade boys and girls in schools in Pakistan. PLoS One 12 (8), e0180833.
https://doi.org/10.1371/journal.pone.0180833.
Kerr-Wilson, A., Fraser, E., Gibbs, A., Ramsoomar, L., Parke, A., Maqbool, H., Jewkes, R.,
2020. What Works to Prevent Violence against Women and Girls? Evidence Review
of Interventions to Prevent Violence against Women and Girls 2000-19. South
African Medical Research Council.
Kimber, M., Adham, S., Gill, S., McTavish, J., MacMillan, H.L., 2018. The association
between child exposure to intimate partner violence (IPV) and perpetration of IPV in
adulthood-A systematic review. Child Abuse Negl. 76, 273–286. https://doi.org/
10.1016/j.chiabu.2017.11.007.
Lamers-Winkelman, F., Willemen, A.M., Visser, M., 2012. Adverse childhood experiences
of referred children exposed to intimate partner violence: consequences for their
wellbeing. Child Abuse Negl. 36 (2), 166–179.
Lapierre, S., Cˆ
ot´
e, I., Lambert, A., Buetti, D., Lavergne, C., Damant, D., Couturier, V.,
2018. Difcult but close relationships: children’s perspectives on relationships with
their mothers in the context of domestic violence. Violence Against Women 24 (9),
1023–1038.
Levendosky, A.A., Graham-Bermann, S.A., 2001. Parenting in battered women: the
effects of domestic violence on women and their children. J. Fam. Violence 16 (2),
171–192.
Malamuth, N., Linz, D., Heavey, C.L., Barnes, G., Acker, M., 1995. Using the conuence
model of sexual aggression to predict men’s conict with women: a 10 year follow
up study. J. Pers. Soc. Psychol. 69, 353–369.
Marshall, W.L., Barbaree, H., 1990. An integrated theory of the etiology of sexual
offending. In: Marshall, W., Laws, D., Barbaree, H. (Eds.), Handbook of Sexual
Assault: Issues, Theories and Treatment of Offenders. Plenum, pp. 257–275.
Mathews, S., Jewkes, R., Abrahams, N., 2015. “So now I’m the man”: intimate partner
femicide and its interconnections with expressions of masculinities in South Africa.
Br. J. Criminol. 55 (1), 107–124.
Merrin, G.J., Hong, J.S., Espelage, D., 2015. Are the risk and protective factors similar for
gang-involved, pressured-to-join, and non-gang-involved youth? A social ecological
analysis. Am. J. Orthopsychiatry 85 (6), 522–535.
Morris, R.E., Anderson, M.M., Knox, G., 2002. Incarcerated adolescents’ experiences as
perpetrators of sexual assualt. Achives of Pediatric and Adolescent Medicine 156,
831–835.
Nduna, M., Jewkes, R., Dunkle, K., Shai, N., Colman, I., 2010. Associations between
depressive symptoms, sexual behaviour and relationship characteristics: a
prospective cohort study of young women and men in the Eastern Cape, South
Africa. J. Int. AIDS Soc. 13 (1), 44. http://www.jiasociety.org/content/13/1/44.
Oram, S., Fisher, H.L., Minnis, H., Seedat, S., Walby, S., Hegarty, K., Rouf, K.,
Ang´
enieux, C., Callard, F., Chandra, P.S., Fazel, S., Garcia-Moreno, C.,
Henderson, M., Howarth, E., MacMillan, H.L., Murray, L.K., Othman, S.,
Robotham, D., Rondon, M.B., Howard, L.M., 2022. The Lancet Psychiatry
Commission on intimate partner violence and mental health: advancing mental
health services, research, and policy. Lancet Psychiatr. 9 (6), 487–524. https://doi.
org/10.1016/s2215-0366(22)00008-6.
Ramsoomar, L., Gibbs, A., Chirwa, E.D., Dunkle, K., Jewkes, R., 2021. Pooled analysis of
the association between alcohol use and violence against women: evidence from four
violence prevention studies in Africa. BMJ Open 11 (7), e049282. https://doi.org/
10.1136/bmjopen-2021-049282.
Ramsoomar, L., Gibbs, A., Chirwa, E.D., Machisa, M.T., Alangea, D.O., Addo-Lartey, A.A.,
Dunkle, K., Jewkes, R., 2023. Pooled analysis of the association between mental
health and violence against women: evidence from ve settings in the Global South.
BMJ Open 13 (3), e063730. https://doi.org/10.1136/bmjopen-2022-063730.
Roustit, C., Renahy, E., Guernec, G., Lesieur, S., Parizot, I., Chauvin, P., 2009. Exposure
to interparental violence and psychosocial maladjustment in the adult life course:
advocacy for early prevention. J. Epidemiol. Community 63 (7), 563–568.
Snyder, J., Horsch, E., Childs, J., 1997. Peer relationships of young children: afliative
choice and the shaping of aggressive behaviour. J. Clin. Child Psychol. 26, 145–156.
R. Jewkes et al.
SSM - Mental Health 7 (2025) 100391
9

Steglich, C.E.G., Snijders, T.A.B., Pearson, M., 2010. Dynamic networks and behaviour:
separating selection from inuence. Socio. Methodol. 40, 329–393.
Steiger, J., 1990. Structural model equation and modication: an interval estimation
approach. Multivariate Behav. Res. 25 (173–80).
Steinberg, J., 2006. The number. Jonathan Ball.
Stirpe, T., Abracen, J., Stermac, L., Wilson, R., 2006. Sexual offenders’ state-of-mind
regarding childhood attachment: a controlled investigation. Sex. Abuse 18, 289–302.
Supol, M., Satyen, L., Ghayour-Minaie, M., Toumbourou, J.W., 2021. Effects of family
violence exposure on adolescent academic achievement: a systematic review.
Trauma Violence Abuse 22 (5), 1042–1056.
Sutton, T., 2017. The lives of female gang members: a review of the literature. Aggression
and Violence Behaviour 37, 142–152.
Trocm´
e, N., 2010. Canadian Incidence Study of Reported Child Abuse and Neglect, 2008:
Major Findings. Public Health Agency of Canada.
Tucker, L.R., Lewis, C., 1973. A reliability coefcient for maximum likelihood factor
analysis. Psychometrika 38, 1–10.
Wadji, D.L., Pirro, T., Langevin, R., 2024. A Systematic Review and Meta-Analysis of the
Association between Childhood Exposure to Intimate Partner Violence and Intimate
Partner Violence Victimization/Perpetration in Adulthood in Africa. Trauma
Violence Abuse, 15248380241287144. https://doi.org/10.1177/
15248380241287144.
Ward, T., Hudson, S., McCormack, J., 1997. Attachment style, intimacy decits and
sexual offending. In: Schwartz, B., Cellini, H. (Eds.), The Sex Offender: New Insights,
Treatment Innovations and Legal Developments. Civic Research Institute.
Whitten, T., Tzoumakis, S., Green, M.J., Dean, K., 2024. Global prevalence of childhood
exposure to physical violence within domestic and family relationships in the
general population: a systematic review and proportional meta-analysis. Trauma
Violence Abuse 25 (2), 1411–1430. https://doi.org/10.1177/15248380231179133.
Wilson, I.M., Willoughby, B., Tanyos, A., Graham, K., Walker, M., Laslett, A.M.,
Ramsoomar, L., 2024. A global review of the impact on women from men’s alcohol
drinking: the need for responding with a gendered lens. Glob. Health Action 17 (1),
2341522. https://doi.org/10.1080/16549716.2024.2341522.
Yount, K.M., James-Hawkins, L., Cheong, Y.F., Naved, R.T., 2018. Men’s perpetration of
partner violence in Bangladesh: community gender norms and violence in childhood.
Psychol. Men Masc. 19 (1), 117–130. https://doi.org/10.1037/men0000069.
R. Jewkes et al.
SSM - Mental Health 7 (2025) 100391
10