Skip Navigation


Journal of Public Health Advance Access originally published online on November 6, 2007
Journal of Public Health 2008 30(1):14-22; doi:10.1093/pubmed/fdm071
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
30/1/14    most recent
fdm071v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Gil-González, D.
Right arrow Articles by Álvarez-Dardet, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gil-González, D.
Right arrow Articles by Álvarez-Dardet, C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007, Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved

Childhood experiences of violence in perpetrators as a risk factor of intimate partner violence: a systematic review



Diana Gil-González, Lecturer in Public Health1,

Carmen Vives-Cases, Lecturer in Public Health1,2,3

María Teresa Ruiz, Professor of Public Health1,2,3

Mercedes Carrasco-Portiño, PhD Student in Public Health1

Carlos Álvarez-Dardet, Professor of Public Health1,2,3
1 Preventive Medicine and Public Health Area, University of Alicante, San Vicente del Raspeig 03080, Alicante, Spain
2 Network for Research on Gender and Health (RISG), Spain
3 CIBER Epidemiología y Salud Pública (CIBERESP), Spain

Address correspondence to Diana Gil-González, E-mail: Diana.gil{at}us.es

Background Perpetrators' experiences of violence during childhood are considered a risk factor for intimate partner violence (IPV). The objective of this study is to systematically review the characteristics and quality of papers which analyse the association between being battered during childhood, witnessing marital violence as a child within the family of origin and having an absent or rejecting father and the occurrence of IPV.

Methods Nine scientific databases were consulted (1960–2004). After applying the exclusion criteria, 10 studies were analysed. Variables are sample characteristics, directionality/study design, IPV and perpetrators' childhood measurements, findings, limitations and interventions.

Results All the studies found an association. Conflict Tactics Scale was the main tool used to measure the IPV. Different instruments were used to measure violent childhood experiences in men as an exposure. Recall bias (seven papers) and retrospective data (four papers) were reported as the main methodological limitations. Despite these, 50% of the studies proposed treatments/preventive measures.

Conclusion Our findings support the results of previous studies, implying that action recommendations within IPV prevention are still not evidence based. Methodological problems of the papers reviewed should be solved to obtain more useful data. Scientific evidence about the aetiology of IPV should be increased to guide effective prevention programmes.


   Introduction
TOP
 Introduction
Methods
 Results
 Discussion
 Funding
 References
 
The increasing relevance of intimate partner violence (IPV) in the scientific and political agenda as a public health issue1–4 has given different explanations of its aetiology. Lori Heise's5 ecological framework for violence against women is the most cited one in the literature. This model explains that a suitable approach to the phenomena should focus on the complexity of different levels: individual, family/relationship, community and societal. Several studies about each of the IPV risk factors identified in Heise's model have been recently published; such is the case of those papers addressing the role of perpetrators' alcohol consumption6 and socioeconomic status7 as risk factors for IPV.

According to Heise's model, being exposed to violence during childhood is considered a risk factor for IPV at the individual level. This determinant distinguishes three situations: being battered during childhood, witnessing marital violence as a child within the family of origin and having an absent or rejecting father. Two empirical reviews prior to 1990s8,9 on this topic conclude that for violent adult males, the most consistent risk factor for perpetrating husband-to-wife violence was having witnessed violence as a child in the family of origin.

However, a theoretical review carried out 10 years ago by Feldman10 on violence against women calls for caution in drawing causal conclusions based on these particular studies on a perpetrator's childhood as a risk factor. The author reports the conclusions of different papers focused on the aetiology of violence against women, and states that the authors of the reviewed articles recognised two important limitations in their studies: the weaknesses inherent in the use of retrospective study designs with adults and recall and social desirability biases.

The rising prevalence of IPV has promoted the creation of prevention and treatment programmes targeted at the rehabilitation of the aggressors.11 Given that the last review of this issue offers a theoretical perspective,10 we propose a public health evidence-based approach to prove the influence of violent childhood experiences on perpetrators.5 This could provide valuable information for both policy-makers and health professionals involved in the prevention of this problem. Consequently, we have systematically reviewed the characteristics and quality of the published papers which specifically analyse the association between being battered during childhood, witnessing marital violence as a child within the family of origin, having an absent or rejecting father during childhood (exposure) and the occurrence of IPV (outcome).


   Methods
TOP
 Introduction
 Methods
Results
 Discussion
 Funding
 References
 
The fields of Social and Behavioural Sciences, Clinical Medicine and Life Sciences were searched using the following databases in all the available years: Medline (1966–2004), Science Citation Index (1945–2004), Social Science Citation Index (1956–2004), Current Contents (1998–2004), PsyINFO (1887–2004), Sociological Abstracts (1963–2004), ERIC Database (1966–2004) and Social Service Abstracts (1960–2004). To identify the keywords, the thesaurus (MESH) was used in all the available databases, except in the Science Citation Index and Current Contents.

The inclusion criteria for this review were: (i) empirical studies, (ii) quantitative methodological studies, (iii) analysis of the variable risk factors ‘being a man battered during childhood, witnessing marital violence within the family of origin or having an absent or rejecting father’ (exposure markers) and its effect on the occurrence of ‘IPV against women’ (outcome).

The main exclusion criteria in this review were defined and justified due to the high prevalence of this problem in men as perpetrators and women as victims of IPV. Also, a scientific approach based on male perpetrators may help for the analysis of the political, social and psychological mechanisms that legitimize the social position of men and women and their impact on social problems such as IPV. All those papers covering the following two main topics were excluded: (i) women victims of violence during childhood and the impact of this on their becoming violent towards their partners in adulthood and (ii) women and men who suffer violence during childhood and how this increases their chances of becoming victims of violence in adulthood (revictimization).

Furthermore, papers on the following issues were also excluded from the review process: treatments and programmes against IPV, other violence-related issues (e.g. alcohol and drug consumption in perpetrators), aggressors' mental illness and its impact on IPV, consequences of childhood violence on children's health and qualitative articles and reviews.

The search equations used were:

  1. Battered women and battered children—battered child syndrome'.
  2. Battered females and battered child syndrome.
  3. Violence against women and men childhood, childhood factors'.
  4. Domestic violence and childhood.
  5. Family violence and childhood experiences, childhood violence, battered children, battered child syndrome, battered children.
  6. Partner abuse and battered child syndrome.
  7. Spouse abuse and battered child syndrome, childhood experiences, childhood experiences, childhood factors.
  8. IPV and men childhood and violence.
The titles, abstract and full text of each paper were examined, omitting those studies determined by the exclusion criteria. A content analysis of the papers included was performed and codified independently by two authors (D.G.-G. and C.V.-C.), using an ad hoc developed checklist that included the studies' methodological characteristics, their findings and the self-acknowledged limitations.
  1. Year of publication.
  2. Characteristics of the studied population: sample, sex and age.
  3. Classification of the type of analytical study design used: cross-sectional, case-control and cohort.
  4. Directionality: retrospective/cross-sectional/prospective.
  5. Inclusion of Heise's model variables (non-excluding variables):
    1. Being battered during childhood (yes/no).
    2. Having witnessed marital violence as a child within the family of origin (yes/no).
    3. Having an absent or rejecting father during childhood (yes/no).

  6. Scale or instrument to measure outcome (IPV).
  7. Scale or instrument to measure exposure (violent childhood experience in men).
  8. Other variables taken from Heise's model at the family/relationship, community and societal levels (yes/no).
  9. Limitations mentioned in the papers.
  10. Main findings
  11. Conclusions about the acknowledged limitations.
  12. Types of suggested measures: preventive and/or treatment interventions.
A simple concordance study was implemented in order to estimate the level of disagreement among the researchers who reviewed the papers. The codifiers and classifiers (D.G.-G. and C.V.-C.) independently analysed a sample of 10 papers. A high level of concordance (90%) was obtained.

A formal meta-analysis was not conducted for two reasons: first, due to the heterogeneity of the studies as regards design, study populations and exposure and outcome measurements, and secondly given the lack of studies including odds ratios or similar effect sizes which could be weighted into a combined estimate. A descriptive study (frequencies and percentages) of the variables included in the review was performed using SPSS 11.5 and Excel-2000 commercial software.


   Results
TOP
 Introduction
 Methods
 Results
Discussion
 Funding
 References
 
Three hundred and fourteen papers were identified from all the years available in the reviewed bibliographic databases. After applying the exclusion criteria, 10 papers were selected.12–21 The 304 excluded papers were related to the following issues: women victims of violence during childhood and the impact of this on their becoming violent towards their partners in adulthood (n = 40; 13%), women and men who suffer violence during childhood and how this increases their chances of becoming victims of violence in adulthood (revictimization) (n = 20; 7%), treatments and programmes against violence (n = 19; 6%), other violence-related issues (e.g. alcohol and drug consumption in perpetrators), aggressors' mental illness (n = 136; 44%), consequences of childhood violence on children's health (n = 78; 26%), and qualitative papers and reviews (n = 11; 4%).

The papers that fulfilled the inclusion criteria were published between 199512 and 2004.20,21 The relations between the experience of being battered during childhood and IPV was explored in eight studies.12–15,17,19–21 Six of these studies12–15,17,19 and two others16,18 considered the relationship between the variable witnessing marital violence as a child in the family and IPV. Only one study measured the relations between IPV and having an absent or rejecting father in childhood.18

Table 1 outlines the methodological characteristics of the studies and their main findings. The studied populations are highly heterogeneous, and cross-sectional designs and retrospective data are used in all the studies.12–21 Many of these studies are not based on a general population sample: some analyse information from groups such as Vietnam veterans and their partners,18 undergraduate men at university17 or Navy recruit trainees.14 The number of individuals analysed in the different studies ranges from 9917 to 8629.19 Six studies compile information from male and female population groups12,14,15,19,20,21 whereas four other studies compile information only from male population samples.13,16–18 In addition, the age range of these populations varies significantly.


View this table:
[in this window]
[in a new window]

  		Table 1 Methodological characteristics and findings of the studies about the intergenerational learning of IPV (1995–2004)

 
Only variables related to the individual were taken into account as possible confounding or interaction factors in the analysis model: impulsivity and depression,13 negative emotionality,20 harsh discipline from grandparents,12 antisocial behaviour,12 stress15 and alcoholism.21 The Conflict Tactics Scale was the main methodological tool used by the studies to measure IPV as an outcome. However, different instruments were used to measure violent childhood experiences in men as an exposure factor and did not take into account the same risk factors. They included variables about diverse violent acts related to different contexts within men's lives (personal experiences, family issues, and social environment). This is the case of the Child Trauma Questionnaire,18 a brief version of the severe variables related to child abuse adapted from the Conflict Tactics Scale21 or the Antisocial Behaviour Trait Scale.12 Other family/relationship, community and societal risk factor variables from Heise's model were not taken into account in the reviewed papers.

Table 2 shows the main limitations highlighted by the studies' authors, most of them related to the information (recall) bias12,14,16–20 and linked to the usage of retrospective data.16,18–20 The generalization of the findings to the whole population (external validity) is a problem identified by 50% of the studies14–17,21 as only men were interviewed; most of these samples are not based on the general population and they are, therefore, heterogeneous. The small size of the samples is also a recognized limitation mentioned in two studies.15,20 The use of the Conflict Tactics Scale22 is questioned as it oversimplifies the patterns of violence by emphasizing physical assaults on a partner without taking into account the social context of the affected population and other kinds of IPV risk factors.17


View this table:
[in this window]
[in a new window]

  		Table 2 Limitations mentioned in the scientific literature about the intergenerational learning of IPV (1995–2004)

 
The authors conclude their papers by proposing treatments focused on different prevention levels13,16,18–20 Three studies recommend screening followed by treatment targeted at men, especially for those with a high risk of perpetrating violence.13,16,19 Some authors suggest that primary health care prevention measures should be applied to reduce the incidence of physical abuse while behavioural family therapy treatments should be implemented to decrease the IPV risk factors.18,20 As regards further studies, 70% of the papers state that new lines of research are required to deal with this problem, including an analysis of women's perceptions of violence.13–16,18,20,21 Meanwhile, new studies using prospective data are also considered to be necessary20 as well as studies to measure the association between IPV and other risk factors.18 Finally, it is considered that steps must be taken to reshape the way the medical community perceives domestic violence so that a multidisciplinary perspective may be achieved.13


   Discussion
TOP
 Introduction
 Methods
 Results
 Discussion
Funding
 References
 
Main finding of this study
Our study found a consistent association between perpetrators' childhood experiences of violence and the occurrence of IPV. The studies acknowledge problems in the quality of their own scientific research regarding this issue, due on the one hand to the use of cross-sectional designs and, on the other, to the retrospective nature of data and recall bias. There are also limitations as regards the external validity of the data. The other family/relationship, community and societal risk factor variables from Heise's model are not considered in the studies. Although all of the studies recommend improving research on this public health problem, half of the papers put forward proposals for different types of treatments focused on the early recognition of risk factors related to the IPV.

What is already known on this topic
The findings of this current review support the results of Feldman's latest work (1997), implying that action recommendations within IPV prevention are still not evidence based. Thus, the screening of violent men and different therapies may be proposed based on insufficient information as regards the true role of IPV determinants.3 Retrospective data and information bias were, and still are, a crucial measurement problem. In fact, some of the reviewed papers highlight the need for new studies using prospective data. Moreover, as Feldman argued not only is a prospective compilation of data necessary, but also an adequate design with the inferential power to increase the quality of scientific research on this issue.

What this study adds
The variety of scales used to measure exposure severely affects the ability to compare the data. The analysed studies compile different information about violent experiences during childhood. Some scales gather data about being battered by the father during childhood, which implies that IPV may be a consequence of the patriarchal ideology, whereas other instruments measured witnessing father-to-mother abuse during childhood, which suggests that IPV may be a consequence of learned sexist behaviour.

The most frequent application of the Conflict Tactics Scale—as the authors of the scale argue22—has been to obtain individual information on physical assaults on a partner. In addition to the criticism made about this scale by one of the studies analysed in our review, the original authors underline that this instrument does not consider social factors related to the IPV context. Furthermore, it is important to improve the tool's capacity in order to obtain more information about IPV.

The authors consider childhood experiences of violence in perpetrators as an individual risk factor without taking into account the family/relationship, community and societal context. This could mean that IPV determinants are addressed at different levels but analysed separately,5 without considering all the aforementioned variables as a whole. To understand learned violence, it is important to focus not only on the individual but also on the societal context.23,24

Scientific evidence about the aetiology of IPV should be increased to address prevention programmes.25 Methodological problems discussed 10 years ago by authors like Feldman10 and acknowledged by the authors of the papers reviewed in this article should be solved to obtain more useful data. Given the multicausal nature of this public health problem, a scientific approach based on different variables, such as the individual, family/relationships, community and society, could be of value to analyse the aetiology of the problem.5,26 Also, the role of patriarchal and sexist27 patterns learned in society and within the family could be taken into account in order to understand the complexity of IPV. Policies against gender violence are increasingly included in political agendas4 and in the mass media.28 Therefore, greater knowledge is required on how IPV is affected by risk factors such as childhood experiences of violence in perpetrators.

Limitations of this study
A limitation to this study may be that grey literature has not been included here, because policy-makers make ample use of such information when designing their prevention programmes. The fact that 90% of the studies were designed in the USA could be a potential constraint in the extrapolation of the results to the populations of other countries. Moreover, as the aim of the systematic review is to summarize the results of the papers and obtain combined conclusions, the heterogeneity of the studies could be a handicap.


   Funding
TOP
 Introduction
 Methods
 Results
 Discussion
 Funding
References
 
The authors thank the Valencian School for Health Studies (EVES) for its financial support to this research (Grant: ‘Intimate partner violence against women. Causes, determinants and risk factor for evidence-based policies’), and the Network for Research on Gender and Health (RISG) on its support.


   References
TOP
 Introduction
 Methods
 Results
 Discussion
 Funding
 References
 

  1. Centre for Disease Control and Prevention (CDC). National Centre for Injury Prevention and Control (NCIPC). Intimate Partner Violence: Fact Sheet. (2004).
  2. Krug E, Dahlberg L, Mercy J. World Report on Violence and Health (2002) Geneva: World Health Organization.
  3. World Health Organization. WHO Multi-country study on women's health and domestic violence. Preliminary results on prevalence, health-related events and women's responses to said violence. (2005) Geneva.
  4. Vives-Cases C, Gil-Gonzalez D, Carrasco-Portiño M, et al. Gender based violence in the Spanish parliamentary agenda [1979–2004]. Gac Sanit (2006) 20:142–8.[CrossRef][Medline]
  5. Heise L. Violence against women. An integrated, ecological framework. Violence Against Women (1998) 4:262–90.[Abstract/Free Full Text]
  6. Gil-Gonzalez D, Vives-Cases C, Álvarez-Dardet C, et al. Alcohol and intimate partner violence: do we have enough information to act? Eur J Public Health (2006) 16:279–85.[Medline]
  7. Vives-Cases C, Gil-González D, Carrasco-Portiño M, Álvarez-Dardet C. Systematic review of studies on the socioeconomic status of men who batter their intimate partners. Gac Sanit (2007) 21:425–30.[CrossRef][Medline]
  8. Hotaling GT, Sugarman DB. Prevention of wife assault. In: Treatment of Family Violence: A Sourcebook—Ammerman RT, Hersen H, eds. (1990) New York, NY: Plenum Press. 3–14.
  9. Hotaling GT, Sugarman DB. An analysis of risk markers in husband to wife violence: the current state of knowledge. Violence Vict (1986) 1:101–24.[Medline]
  10. Feldman CM. Childhood precursors of adult interpartner violence. Clin Psychol (1997) 4:307–34.[Web of Science]
  11. Eckhardt CI, Murphy C, Black D, et al. Intervention programs for perpetrators of intimate partner violence: conclusions from a clinical research perspective. Public Health Rep (2006) 121:369–81.[Web of Science][Medline]
  12. Simons RL, Wu C, Jonson C, et al. A test of various perspectives on the intergenerational transmission of domestic violence. Criminology (1995) 33:141–72.[CrossRef][Web of Science]
  13. Oriel KA, Fleming MF. Screening men for partner violence in a primary care setting. A new strategy for detecting domestic violence. J Fam Pract (1998) 46:493–8.[Web of Science][Medline]
  14. Merrill LL, Hervig LK, Milner JS. Childhood parenting experiences, intimate partner conflict resolution, and adult risk for child physical abuse. Child Abuse Negl (1996) 20:1049–65.[CrossRef][Web of Science][Medline]
  15. Kesner JE, McKenry PC. The role of childhood attachment factors in predicting male violence toward female intimates. J Fam Violence (1998) 13:417–32.[CrossRef][Web of Science]
  16. Martin SL, Moracco KE, Garro J, et al. Domestic violence across generations: findings from northern India. Int J Epidemiol (2002) 31:560–72.[Abstract/Free Full Text]
  17. Carr JL, VanDeuse KM. The relationship between family of origin violence and dating violence in college men. J Interpers Violence (2002) 17:630–46.[Abstract/Free Full Text]
  18. Orcutt HK, King LA, King DW. Male-perpetrated violence among Vietnam veteran couples: relationships with veterans early life characteristics, trauma history, and PTSD symptomatology. J Trauma Stress (2003) 16:381–90.[CrossRef][Web of Science][Medline]
  19. Whitfield CL, Anda RF, Dube SR, et al. The violent childhood and the risk of intimate partner violence in adults. Assessment in a large health maintenance organization. J Interpers Violence (2003) 18:166–85.[Abstract]
  20. Herrenkhol TI, Mason WA, Kosterman R, et al. Pathways of physical childhood abuse to partner violence in young adulthood. Violence Vict (2004) 19:123–36.[CrossRef][Medline]
  21. Schafer J, Caetano R, Cunradi C B. A path model of risk factors for intimate partner violence among couples in the United States. J Interpers Violence (2004) 19:127–42.[Abstract/Free Full Text]
  22. Strauss MA, Hamby SL, Boney-McCoy S, et al. The revised Conflict Tactics Scales (CTS2). Development and Preliminary Psychometric Data. J Fam Issues (1996) 17:283–316.[CrossRef]
  23. Dodge KA, Bates JE, Pettit GS. Mechanisms in the cycle of violence. Science (1990) 220:1678–83.
  24. Doumas D, Margolin G, John RS. The intergenerational transmission of aggression across generations. J Fam Violence (1994) 9:157–75.[CrossRef][Web of Science]
  25. Edleson JL. Judging the success of treatments with men who batter. In: Family Violence: Research and Public Policy Issues—Besharov DJ, ed. (1990) New York, NY: AEI Press. 130–45.
  26. Krieger N. Theories for social epidemiology in the 21st century: an ecosocial perspective. Int J Epidemiol (2001) 30:668–77.[Free Full Text]
  27. Inhorn MC, Whittle KL. Feminism meets the "new" epidemiologies: toward an appraisal of antifeminist biases in epidemiological research on womens health. Soc Sci Med (2001) 53:553–67.[CrossRef][Web of Science][Medline]
  28. Vives-Cases C, Ruiz MT, Álvarez-Dardet C, et al. Recent history of the news coverage of violence against women in Spain [1997–2001]. Gac Sanit (2005) 19:22–28.[CrossRef][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
30/1/14    most recent
fdm071v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Gil-González, D.
Right arrow Articles by Álvarez-Dardet, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gil-González, D.
Right arrow Articles by Álvarez-Dardet, C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?