Measuring and Addressing the
Vos T, PhD MSc
Astbury J, PhD
Piers L S, MBBS, MD, PhD, MPH
Magnus A, BEc. Ed. G.
Heenan M, PhD
Walker L, MSW
Webster K, BSW
Author responsible for correspondence
In the Australian state of Victoria, as is the case both nationally and in other countries around the world, intimate partner violence (IPV) is increasingly recognised as a human rights violation with serious health, social, legal and economic consequences (Australian Medical Association [AMA], 1998; Office of Women’s Policy [OWP], 2002; Public Health Association Australia [PHAA], 2002; United Nations [UN], 2006; World Health Organization [WHO], 2002). Although much work remains to be done, in recent decades there have been significant advances in social and legal responses to men who use violence and the women and children affected by it (Department of Victorian Communities [DVC], 2005; Victorian Law Reform Commission [VLRC], 2006). However less attention has been given to primary prevention; that is to preventing violence before it occurs.
Although the causes of IPV are complex, factors in our social, cultural and economic environments play a significant part (Heise, 1998; UN, 2006; WHO, 2002). This suggests that there are prospects for preventing violence by addressing these factors.
The Victorian Health Promotion Foundation (VicHealth) is an independent statutory authority which works in partnership with others to improve the health of all Victorians by addressing social, economic and behavioural factors contributing to poor health. In its work, it has adopted a public health approach involving the application of multiple and reinforcing strategies at a population level. With the primary influences on health often lying outside of the health care system (Wilkinson & Marmot, 2003), much of this work is undertaken in partnership with actors in other sectors, such as employment, education or sports and recreation. Significant health gains have been made through such an approach in addressing other public health issues. Prominent examples include tobacco control and road safety, where major reductions in avoidable death, injury and illness have been achieved through a combination of legislative reform, law enforcement, social marketing, organisational development, research and services and programs to support individuals.
VicHealth is currently seeking to apply this approach to the goal of reducing violence against women. This work is being undertaken as part of a program of activity aimed at promoting mental health, given the strong evidential link between poor mental health and exposure to this form of violence (WHO, 2000). Intimate partner violence has also been identified as a significant contributor to gender disparities in some mental health problems such as depression (Astbury, 2001). Critical to a public health approach is accurate data to determine the magnitude and health impacts of the problem concerned and to raise awareness to ensure that it is given appropriate priority. Data is also important for planning and for monitoring the impact and assessing cost effectiveness of interventions (Graffunder, Noonan, Cox, & Wheaton, 2004).
This study was conducted as a partnership of VicHealth, the Victorian Department of Human Services and researchers with expertise in addressing violence and in burden of disease methodology. The study had three primary objectives: (a) to document the prevalence of IPV; (b) to identify the health problems associated with this form of violence; and (c) to estimate the contribution IPV makes to illness, injury and premature death among Victorian women using burden of disease methodology.
Burden of disease methodology
Burden of disease methodology is an internationally accepted approach to estimating the impacts of particular health problems across a population, taking into account illness, disability and premature death. The burden of disease is a measurement of the gap between the current health status of a population and an ideal situation where everyone lives into old age free of illness and disability (Murray & Lopez, 1997; WHO, 2001). The advantage of this methodology compared with other commonly used health indicators (such as mortality or health service utilisation data) is that it enables estimation of all health risks, including morbidity and disability. Burden of disease measures are used extensively by governments, researchers, health planners and advocates world wide and provide a standard measure for:
Burden of disease estimates were first developed for Victoria in 1996 for 176 diseases and 10 risk factors (Vos & Begg, 1999a, 1999b). More recently, results were being updated to the year 2001, enabling an estimate for IPV to be included for the first time.
While the prevalence of IPV is notoriously difficult to determine, this study drew on the most reliable national population-based survey of its time. The Women’s Safety Survey (WSS), conducted in 1996, involved face to face interviews with over 6000 Australian women (Australian Bureau of Statistics [ABS], 1996a, 1996b). Prevalence-based, rather than incident-based data for Australia were used since they involved larger numbers of women and hence provided a basis for a more precise estimate. To establish the type, severity and extent of health problems which have been found in research to result from violence, an extensive review of Australian and international studies and published and unpublished government reports and research was undertaken.
A peer reviewed paper providing a detailed technical description of the methods used to calculate burden of disease estimates can be found here (Vos et al., 2006). In summary, estimates of the contribution to illness, injury and premature death made by IPV were developed for each of the main health problems found in the review described above. These estimates were made using prevalence data from the WSS and data on the relative risk of adverse health outcomes through exposure to IPV. Most of these were based on the Australian Longitudinal Study on Women’s Health (ALSWH), a study that began in 1996 and enrolled 40,000 women with the intention of periodically surveying them about their health over a 20-year period. Risk data for femicide were based on a study of a coronial data base conducted by the Australian Institute of Criminology (Mouzos, 1999), and those for physical injuries were derived from a 5-year study of hospital emergency department data conducted in the Australian state of Queensland (Roberts, O’Toole, Raphael, Lawrence, & Ashby, 1996).
To find out how much of the disease burden can be attributed to IPV, these data sources were used to estimate how much less disease there would have been in the whole population if no woman had been the subject of IPV. This proportion was then multiplied by the overall estimates of the burden for each of the relevant diseases. The latest year for which these were available was 2001.These results were summed to calculate the burden of disease that occurs due to women’s exposure to IPV. Statistical testing of the data and assumptions (described in greater detail in the technical paper referred to above) was undertaken to maximise the robustness of the estimates. This included sensitivity analysis.
The study focused on IPV perpetrated against women by their male partners, on the understanding that the health consequences of this form of violence are particularly acute for women owing to the gendered patterns of both the experience and consequences of violence. In particular, violence perpetrated by men against their female partners is more likely than that perpetrated by women to be part of a pattern of severe, prolonged and repeated abuse. Women victims are also more likely than their male counterparts to sustain injuries, to fear for their lives and to be sexually assaulted (ABS 2003; Bagshaw & Chung 2000; Statistics Canada, 2003; Flood 2003, Belknap & Melton 2005, Gordon 2000 cited in Flood 2006). While recognising that IPV occurs on a continuum of economic, psychological and emotional abuse through to physical and sexual violence, this study could only reliably consider physical and sexual violence owing to the paucity of data pertaining to other forms of violence.
Data from the WSS (see Table 1) indicated that almost 195,000 or 2.9% of Australian women had experienced recent physical or sexual violence (in the 12 months prior to the survey), while over one million or 17.0% had experienced past physical or sexual violence (over the life time). Prevalence was found to be higher in younger women than in those aged over 45 years.
The large body of literature identified in the review of existing research demonstrates that IPV has wide ranging and persistent effects on women’s physical and mental health (summarised in Table 2), with the impact over time of different types and multiple episodes of abuse appearing to be cumulative (Golding, 1999; Taft, 2003; WHO, 2000).
Table 2: Summary of known health outcomes of intimate partner violence
This form of violence was found to be associated with an increased risk of premature death and injury. Between 1989 and 1998, over 57 per cent of deaths in Australian women resulting from homicide or violence were perpetrated by an intimate partner, with women being over five times more likely to be killed by an intimate partner than men (Mouzos, 1999). Women experiencing IPV were found to be more likely to experience mental health problems (Parker & Lee, 2002; Roberts, Williams, Lawrence, & Raphael, 1998; WHO, 2000), with one study of women attending general practitioners indicating a fivefold increased risk of depression (Hegarty, Gunn, Chondros, & Small, 2004). Women reporting IPV are also significantly more likely to use medication for depression and anxiety (Campbell, 2002; Coker et al., 2002; Hathaway et al., 2000; Janssen et al., 2003; Loxton, Schofield & Hussain, n.d.; Resnick, Acierno & Kilpatrick, 1997).
There is also evidence that the mental health impacts of violence persist over many years, with women who have experienced violence in the past having lower rates of mental health problems than women reporting current IPV, but significantly higher rates than those who have never experienced this type of violence (Golding, 1999; Loxton, Schofield & Hussain, n.d.).
IPV was found to increase the risk of women engaging in practices and behaviours harmful to their health, including the use of tobacco, non-prescription drugs, amphetamines and solvents and risky levels of drinking (Golding, 1999; Quinlivan & Evans, 2001; Roberts, Lawrence, O’Toole, & Raphael, 1997; Roberts, Lawrence, et al., 1998; Roberts, Williams, et al., 1998). Women reporting IPV were found to be more likely to have had an abnormal pap smear and to report having a vaginal or endo cervical infection (Quinlivan & Evans, 2001) and to have a higher risk of complications of pregnancy and birth (Quinlivan & Evans, 2001; Taft, 2002). This is a particular concern given data indicating that some 42% of Australian women reporting that they have been exposed to IPV at some time in their lives were pregnant at the time of the violence (WHO, 2000).
Contribution to burden of disease
Using the burden of disease methodology described above (see Figure 3), it was found that in women under the age of 45 years of age, IPV is responsible for an estimated 7.9% of disease burden. It is less for women over 45 years of age (1.5%) and 2.9% in women overall (i.e. when all ages are combined). The greatest proportion of the disease burden is associated with mental health problems. Depression, anxiety and suicide together contribute 73% of the total disease burden associated with IPV. Harmful health related behaviours, including tobacco, alcohol and illicit drug use, accounted for another 22% (see Figure 1).
IPV was found to be a greater risk for ill-health among women aged under 45 years than seven other major risk factors included in contemporary burden of disease estimates, including body weight, high cholesterol and high blood pressure (see Figure 2). It was more than twice the risk of the next most important factor, illicit drug use, which contributed to less than 4%. For all ages combined IPV was associated with more disease among women than alcohol and illicit drugs (see Figure 3).
Burden of disease estimates are derived in part from prevalence data. While the best available data sources were chosen, existing measures may under estimate the problem, with studies showing that compared with other forms of violence, women are less likely to disclose IPV or to name the act as violence (Lievore, 2003; WHO, 2002). This is particularly the case for women from non-English speaking backgrounds (Lievore, 2003). Women from non-English speaking and Indigenous backgrounds and women with disabilities were under represented in the prevalence data used in this study. These groups may be particularly vulnerable to violence or its impacts (Mouzos & Mikkai, 2004; National Crime Prevention, 2001, Rees & Pease, 2006). Their under representation also worked against comparing the burden experienced by these groups with the burden for all women. Both the prevalence data and burden of disease estimates only included physical and sexual violence, with the result that the health impacts associated with other forms of violence were not included despite evidence of their health consequences (Coker et al., 2002).
Calculating the burden of disease associated with a risk factor involves using data from different sources. In many areas of health care, data collection is imperfect or sources of data are not directly compatible. This was also the case for data pertaining to IPV. However, the strength of burden of disease methodology is that it provides a standard framework for addressing these issues so that plausible estimates can be made from the available information. The data issues encountered in this study and the steps taken to address them are discussed in greater detail in the technical paper available here (Vos et al., 2006).
The WSS, from which prevalence data was derived, was the most recent reliable data source at the time of the study. However, it was somewhat dated and an assumption of the study was that prevalence had not changed between 1996 and 2001. Some questions in the WSS have since been replicated in the Personal Safety Survey (involving both male and female respondents) conducted in 2005 (ABS, 2006). Data used in our study are not directly comparable with those published to date for the Personal Safety Survey. However, between 1996 and 2005 there were no sizeable changes in the proportion of women reporting exposure to all forms of interpersonal violence over the lifetime.
Implications of findings
The findings of this study are a stark indication of the impact of IPV for women’s health and human rights, and present significant challenges for communities, service providers and governments. Addressing this issue will be important not only to reduce the contemporary health burden but also that of future generations, given evidence that IPV also has an impact on the children of women exposed (Edelson, 1999). Increasing the attention given to IPV is clearly warranted. The findings also suggest the importance of building the knowledge and skill base amongst health care providers to respond to those affected by the problem, especially in settings concerned with substance abuse and reproductive and mental health.
Given the prevalence and serious health consequences documented in this study, efforts to address IPV should not be limited to improving responses to the victims and perpetrators of violence. Rather there is a need to strengthen efforts in primary prevention through strategies implemented at the population level. Work currently being undertaken by VicHealth indicates that the prospects for the effectiveness of such an approach are promising (VicHealth, in preparation).
Primary prevention is particularly critical given not only the size of the burden but also its composition. Mental health problems, which account for nearly three- quarters of the IPV disease burden, comprise a substantial proportion of the total disease burden of both developed and developing nations (Mathers & Loncar, 2006). Depression alone is predicted to be among the top three contributors to global disease burden by 2030 (Mathers & Loncar, 2006). A reduction in IPV prevalence is not only likely to result in health gains in this area, but also in respect to other risk behaviors in which IPV is implicated and which themselves contribute to high rates of chronic disease (e.g. tobacco and illicit drug use). For its part, VicHealth is continuing to work with others to strengthen the application of a public health approach to the primary prevention of IPV. Activities include a partnership with the Victorian Government to develop an evidence informed framework to guide policy and program development in this area, intervention research to build the evidence and knowledge base for primary prevention, periodic community attitudes surveys to inform media and educational activity, a small grants program to support communities and organizations to undertake primary prevention activity, and an organizational development and media program being undertaken in partnership with Victoria’s largest sporting code, the Australian Football League.
Through this approach VicHealth hopes to complement efforts being made to prevent the reoccurrence of violence and to mitigate its effects with sound strategies implemented across sectors to prevent it from occurring in the first instance.
Thanks are extended to the group of researchers at the University of Newcastle and the University of Queensland who are conducting the Australian Longitudinal Study on Women’s Health.
This study was funded by the Australian Department of Health and Ageing and the Victorian Health Promotion Foundation.
ABS. (1996a). Women’s safety Australia. Cat. No. 4128.0. Canberra: Australian Bureau of Statistics.
ABS. (1996b). Women’s safety Australia: User guide 1996. Cat. No. 4129.0. Canberra: Australian Bureau of Statistics.
ABS. (2003). Crime and safety, Australia, 2002. Cat. No. 4509.0. Canberra: Australian Bureau of Statistics.
ABS. (2006). 2005 Personal safety survey. Cat. No. 4906.0. Canberra: Australian Bureau of Statistics.
AMA. (1998). Position statement on domestic violence. Australian Medical Association. Retrieved September 24, 2006, from http://www.ama.com.au/web.nsf/doc/SHED-5FK5FQ
Astbury, J. (2001). Gender disparities in mental health. Background paper, Ministerial Rountables, 2001 World Health Assembly on Mental Health. Geneva: World Health Organization.
Bagshaw, D., & Chung, D. (2000). Gender politics and research: Male and female violence in intimate relationships. Women Against Violence: An Australian Feminist Journal, 8, 4-23.
Campbell, J. C. (2002). Health consequences of intimate partner violence. Lancet, 359, 1331-1336.
Coker, A. L., Davis, K. E., Arias, I., Desai, S., Sanderson, M., Brandt, H. M., et al. (2002). Physical and mental health effects of intimate partner violence for men and women. American Journal of Preventive Medicine, 23, 260-268.
DVC. (2005). Changing lives. A new approach to family violence in Victoria. Melbourne: Department of Victorian Communities.
Edleson, J. (1999). Children’s witnessing of adult domestic violence Journal of Interpersonal Violence, 14(8) 839-870
Flood, M., (2006) Violence against women and men in Australia. What the Personal Safety Survey can and can’t tell us Domestic Violence and Incest Resource Centre Newsletter Edition 4, Summer.
Golding, J. M. (1999). Intimate partner violence as a risk factor for mental disorders: A meta-analysis. Journal of Family Violence, 14(2), 99-132.
Graffunder, C. M., Noonan, R. K., Cox, P., Wheaton, J. (2004). Through a public health lens. Preventing violence against women. An update from the US Center for Disease Control and Prevention. Journal of Women’s Health, 13(1), 5-14.
Hathaway, J., Mucci, L., Silverman, J., Brooks, D., Mathews, R., Pavlos, C. (2000). Health status and health care use of Massachusetts women reporting partner abuse. American Journal of Preventive Medicine, 19, 302-307.
Hegarty, K., Gunn, J., Chondros, P., Small R. (2004). Association between depression and abuse by partners of women attending general practice. Descriptive cross sectional survey. British Medical Journal, 328(7440), 621-624.
Heisse, L. L. (1998). Violence against women: An international, ecological framework. Violence Against Women, 4(3), 262-290.
Janssen, P., Holt, V., Sugg, N., Emanuel, I., Critchlow, C., Henderson, A. (2003). Intimate partner violence and adverse pregnancy outcomes: A population based study. American Journal of Obstetrics and Gynecology, 188, 1341-1347.
Lievore, D. (2003). Non-reporting and hidden reporting of sexual assault: An international literature review. Canberra: Australian Institute of Criminology.
Loxton, D., Schofield, M., Hussain, R. (n.d.). Psychological health in mid-life among women who have ever lived with a violent partner or spouse. Analysis of data from the mid aged cohort of the Australian longitudinal study on women’s health. Unpublished paper.
Mathers, C. D., & Loncar D. (2006). Projections of global mortality and burden of disease from 2002 to 2030. PLoS Medicine, 3, 442.
Mouzos, J. (1999). Femicide: An overview of major findings. Trends and issues in crime and criminal justice, No. 124. Canberra: Australian Institute of Criminology.
Mouzos, J., & Mikkai, T. (2004). Women’s experiences of male violence. Findings from the Australian component of the international violence against women survey. Research and Public Policy Series, No. 56. Canberra: Australian Institute of Criminology.
Murray, C. J., & Lopez , A. D. (1997). Global mortality, disability and the contribution of risk factors: Global burden of disease study. Lancet, 349, 1436-42.
National Crime Prevention. (2001). Violence in Indigenous communities. Full Report. Canberra: National Crime Prevention.
OWP. (2002). Women’s safety strategy: A policy framework. Melbourne: State Government of Victoria.
Parker, G., & Lee, C. (2002). Violence and abuse: An assessment of mid-aged Australian women’s experiences. Australian Psychologist, 37(2), 142-148.
PHAA. (2002). Domestic and family violence policy. Public Health Association Australia.
Quinlivan, J. A., Evans, S. F. (2001). A prospective cohort study of the impact of domestic violence on young teenage pregnancy outcomes. Journal of Pediatric and Adolescent Gynecology, 14, 17-23.
Rees, S., & Pease, B. (2006). Refugee resettlement, safety and wellbeing. Exploring family and domestic violence in refugee communities. Melbourne: Immigrant Women’s Domestic Violence Service.
Resnick, H., Acierno, R., Kilpatrick, D. (1997). Health impact of interpersonal violence. 2: Medical and mental health outcomes. Behavioral Medicine, 23, 65-78.
Roberts, G. L., Lawrence, J. M., O’Toole B. I., Raphael, B. (1997). Domestic violence in the emergency department: 1. Two case control studies of victims. General Hospital Psychiatry, 19, 5-11.
Roberts, G. L., O’Toole, B. I., Raphael, B., Lawrence, J. M., Ashby, R. (1996). Prevalence study of domestic violence victims in an emergency department. Annals of Emergency Medicine, 27, 741-753.
Roberts, G. L., Williams, G. M., Lawrence, J. M., Raphael, B. (1998). How does domestic violence affect women’s mental health? Women and Health, 28, 117-129.
Statistics Canada. (2003). Family violence in Canada: A statistical profile 2003. Ottawa, Canadian Centre for Justice Statistics.
Taft, A. (2002). Violence in pregnancy and after childbirth. Issues paper No. 6. Sydney: Australian Domestic and Family Violence Clearing House.
Taft, A. (2003). Promoting women’s mental health: The challenge of intimate partner/domestic violence. Issues paper No. 8. Sydney: Australian Domestic and Family Violence Clearing House.
UN. (2006). Secretary General’s study on violence against women. Background documentation for 61st session of the General Assembly. Item 60(a) on advancement of women. Retrieved December, 2006, from http://www.un.org/womenwatch.daw/vaw/violenceagainstwomenstudy.doc.pdf
VLRC. (2006). Review of family violence laws. Melbourne: Victorian Law Reform Commission.
Vos, T., & Begg, S. (1999a). The Victorian burden of disease study: Morbidity. Melbourne: Department of Human Services.
Vos, T., & Begg, S. (1999b). The Victorian burden of disease study: Mortality. Melbourne: Department of Human Services.
Vos, T., Astbury J., Piers, L. S., Heenan, M., Walker, L., Webster, K. (2006). Measuring the impact of intimate partner violence on the health of women in Victoria, Australia. Bulletin of the World Health Organization, 84(9), 739-744.
WHO (2000). Women and mental health: An evidence based review. Geneva: World Health Organization.
WHO. (2001). World health report 2001. Mental health. New understanding, new hope. Geneva: World Health Organization.
WHO. (2002). World report on violence and health. Geneva: World Health Organization.
Wilkinson, R., & Marmot, R. (2003). The solid facts. Social determinants of health. Geneva: World Health Organization.