Violence against Women: A literature review commissioned by the National Group to Address Violence Against Women
3. Consequences of violence against women
3.1 For women
3.1.1 Fear of violence
Given the prevalence rates for violence against women, it is perhaps not surprising that women feel less safe than men. However, women's perceptions of their own safety are significantly at odds with the realities of where the risk to them is located. Despite consistent research reports that women are most at risk from men known to them, the myth of the dangerous stranger prevails. Women feel most at risk on the streets, at night - 16% of women in the 2000 Scottish Crime Survey said they felt 'very unsafe' when walking alone after dark; 40% of women felt 'very or a bit unsafe'. By comparison, only 3.5% of men reported to the same survey that they felt 'very unsafe' walking alone after dark, despite the fact that they are more than three times more likely than women to experience an assault (Scottish Executive, 2000). Men are also more likely to be assaulted by strangers - only 17% of the 'stranger assaults' reported in a recent survey in England and Wales were committed against women. Conversely, 73% of the assaults designated 'domestic' were reported by women (Simmons and Dodd, 2003).
Elizabeth Stanko offers an interesting juxtaposition of the experiences of women and men and how those experiences impact on their feelings about safety and danger:
"Women's lives rest upon a continuum of unsafety. This does not mean that all women occupy the same position in relation to safety and violence. Many other features of their lives...will mean that their circumstances differ. Somehow, though, as all women reach adulthood, they share a common awareness of their particular vulnerability. Learning the strategies for survival is a continuous lesson about what it means to be female." (Stanko, 1990: 85)
In contrast, she says:
"For men, there are no tips about personal safety in crime prevention handbooks. It is assumed that men either know about avoiding dimly-lit alley ways and bus stops, or that they are able to protect themselves. While we may assume men already know how to protect themselves, they don't seem to be very successful: men's recorded levels of victimisation are much higher than women's." (Stanko, 1990: 109).
Women live with a consciousness of being 'at risk' of violence which is not experienced by most white heterosexual men, although Stanko acknowledges the connections between racist and homophobic violence and violence against women. It is this consciousness of ever present risk, she argues, which underpins the strategies women adopt to deal with the threat of violence in their daily lives, whether at home, at work, or on the street.
3.1.2 Health consequences of violence against women
The World Health Organisation defined violence as a public health issue in 1996, noting that it impacted especially on the health of women and children (WHO, 1997). Subsequently, a plan of action on violence against women drawn up by WHO identified areas of work needed in order to prevent violence and reduce violence-related morbidity and mortality among women. This work included the development of multi-country research, the need to document and test the efficacy of existing health interventions, and the need to raise awareness among health professionals of the impact of violence against women (WHO, 1997).
The latest outcome of this action plan is the World report on violence and health, a review of world literature and research on violence, including violence against women. This comprehensive report explores all aspects of the health consequences of violence against women, including injuries, pregnancy and reproductive health, chronic physical health issues, and the effects on mental health and wellbeing (Krug et al, 2002).
The 1996 national U.S. survey on violence against women found that in around a third of all rapes and physical assaults against women, the woman was injured. About one in three of those injured needed medical attention. Most of these injuries consisted of relatively minor bruising, scratches and welts (Tjaden and Thoennes, 1998). However, more severe injuries, including broken bones or fractures, burns and lacerations, are also recorded, particularly in relation to assaults by partners (Guthrie, 1998; McWilliams and McKiernan, 1993; Williamson, 2000)).
Rape and sexual assault may result in women acquiring sexually transmitted infections, including HIV and hepatitis (Winn et al, 2003; WHO, 2000). Pregnancy as a result of rape may cause psychological distress to the woman, including having to make decisions about whether to continue with the pregnancy (Lathrop, 1998; Rape Crisis Centre, 2003). If the woman decides to keep the child, there can also be difficulty for both the woman and the child in the long term (Rape Crisis, 1993). Sexual assault is linked to a range of gynaecological complications, including vaginal infection, bleeding, recurrent urinary tract infections, and chronic pelvic pain (Golding, 1996). These symptoms are reported by female survivors of both child and adult sexual assault, including women abused through prostitution (Farley and Barkan, 1998; Golding,1996).
In a sample of 892 women in two London hospitals, 2.5% of women reported experiencing domestic violence in their current pregnancy (Mezey et al, 2001). A recent study in the north of England found that 3.4% of the 475 respondents had experienced domestic violence during their current pregnancy. In both of these studies, the lifetime prevalence stood at 13-17% of the women surveyed. An earlier (1993/94) Canadian study showed slightly higher results, with 5.7% of a sample of 728 women reporting domestic violence during their pregnancy (Muhajarine and D'Arcy, 1999). Violence against women during pregnancy has been associated with miscarriage, premature birth, low birth weight, fetal injury and maternal death (RCOG, 2001, cited in Johnson et al, 2003).
Women survivors of violence also report a range of chronic health conditions, including gastro-intestinal problems (Goodwin et al, 2003), reproductive health problems, respiratory difficulties, migraine (Goodwin et al, 2003), impaired hearing or sight, joint pain, other chronic pain, and disability (Springer et al, 2003; Williamson 2000). Clinical research may focus on discrete populations with clearly defined characteristics, which has implications for the kind of research done in relation to violence against women. For example, there are a number of studies which consider the links between domestic abuse, and/or childhood abuse, and gastrointestinal disorders, but an extensive search of the same databases has failed to identify any comparable, or comparative, research with rape or sexual assault survivors. It may be that gastro-intestinal disorders are not commonly reported by sexual assault survivors. It may also be the case that, in the absence of routine screening, sexual assault survivors who do have gastro-intestinal disorders are less likely to disclose their assault history. A recent study looked at the relationship between women's history of abuse and subsequent (non-acute) surgical interventions. The study found that the number of surgeries undergone by women survivors of childhood abuse or domestic violence was significantly higher than for women with no reported history of abuse - 88.9% of women survivors of childhood abuse and 95% of survivors of domestic violence, compared with 67.3% of the control group (Hastings and Kaufman Kantor, 2003). Twice as many survivors of domestic violence as women with no abuse history had undergone major surgery. The authors acknowledge that the sample size was small (n=53) and possibly difficult to generalise from because it was predominantly made up of white women. However, they argue that the results are significant enough to warrant further investigation, particularly around the need to review perioperative nursing practice with a view to improving the identification and subsequent treatment of women survivors of abuse who are presenting for surgery (Hastings and Kaufman Kantor, 2003).
Women who have experienced violence report significant mental health difficulties (Carlson et al, 2003; Thomson, 1998; Women's Health Team, GGNHSB, 2003; Thompson et al 2002; Williamson, 2000). The mental health impact of child abuse and neglect has been relatively well documented, although research findings are not always consistent. A 1999 study of women attending GP surgeries in north London looked at the correlation between women's reported experiences of physical and sexual abuse at different points in the life cycle, and their mental health status. The study found that childhood experiences of physical abuse were associated with several mental health indicators, including depression, anxiety and self-harm. The study found that women reporting childhood experiences of sexual abuse were five times more likely to suffer from post traumatic stress disorder (Coid et al, 2003). However, it found no association between childhood experiences of sexual abuse and other adverse mental health outcomes. This is at odds with findings in other studies, which have identified a correlation between sexual abuse in childhood and adult experiences of mental health difficulty, including depression and anxiety (Fleming et al, 1999; Follette et al, 1996; Mullen et al, 1996). Coid et al acknowledge the difference in outcome between their study and others, and suggest it may be due in part to differences in methodology and in part to a different 'starting position', i.e. they made no assumptions about child sexual abuse being "the primary abusive experience associated with the psychopathological symptoms measured in adulthood" (Coid et al, 2003: 336). An invited response to this study strenuously refutes the findings in relation to child sexual abuse and psychiatric morbidity, while acknowledging the value of looking at child sexual abuse in the context of other forms of abuse (Mullen, 2003).
The north London study also reported associations between sexual assault in adulthood and substance misuse, and between rape and anxiety, depression and PTSD. Domestic violence had the strongest links across all of the mental health measures used in the study, with the exception of self injury (Coid et al, 2003).
Rates of PTSD in women survivors of violence are high (Farley and Barkan, 1998; Ullman and Brecklin, 2003). Other recent research has established an association between early onset child sexual abuse and both borderline personality disorder and complex PTSD, with the possibility that some women should be considered under the latter diagnosis rather than the former (McLean and Gallop, 2003). There is also some evidence that repeated experiences of violence or abuse have a cumulative effect, resulting in higher rates of PTSD, and more intense reactions (see Farley and Barkan, 1998, on PTSD in women working in prostitution; Ferguson (publication pending) on Complex PTSD in survivors of domestic abuse and child sexual abuse; Follette et al, 1996; Herman, 1992). Large numbers of women in acute mental health settings have histories of child sexual abuse (Nelson and Phillips, 2001).
The Adverse Childhood Experiences (ACE) Study, carried out in a large primary health care organisation in the U.S. in 1995/96, is a large scale study of the health impact of a range of experiences, including physical and sexual abuse. The study gives a very clear, if depressing, picture of the degree to which adverse childhood experiences - including sexual abuse, physical abuse, and witnessing violence towards mothers - are related to some of the leading causes of death in adults - including alcoholism, drug abuse, depression and suicide attempts, ischemic heart disease, cancer, chronic lung disease, skeletal fractures and liver disease (Felitti et al, 1998).
Much of the chronic ill health reported by survivors of child sexual abuse has traditionally been identified as somatic. Relatively little research has been done to establish how much of what was considered 'somatic' is actually rooted in women's experiences of violence and the impact of this on their bodies. Nelson (2002) explores some of the connections between the nature of the abuse experienced in childhood and the physical health difficulties experienced subsequently. She concludes that an increased awareness of what actually happens to children who are being abused may shed light on a range of "medically unexplained symptoms" (Nelson, 2002:51).
The consequences of violence against women for women's health can be severe. However, acknowledging the potential impact on women's health must be balanced against the danger of pathologising both the causes and the consequences of violence against women.
3.1.3 Socioeconomic consequences
Given the established relationship between poor health and poverty, it might be expected that violence against women will also have socioeconomic consequences.
Women living with ongoing abuse lose work days and earnings as a result of being injured (Browne et al, 1999; Lloyd and Taluc, 1999). Similarly, women may lose time at work, and sometimes be forced to leave jobs, as a result of sexual assault. Poverty can be seen as both a causal factor and a consequence of violence against women; findings from the U.S. National Comorbidity Study suggest that women living below poverty levels appear to be more vulnerable to assault, but also that women living above poverty levels at the time of an assault are twice as likely to lose income as a result (Byrne, Resnick et al, 1999).
That women become homeless as a result of domestic abuse is perhaps self-evident. Recent Scottish research has demonstrated that as a consequence of leaving a violent partner, women may go through a lengthy period of disruption before settling in a new permanent home (Edgar et al, 2003; Fitzpatrick et al, 2003). The difficulties they have been left with as a result of the violence they have experienced may be compounded by the disruption and trauma of living in temporary accommodation, perhaps moving many times before achieving safety.
The relationship between other forms of violence against women and homelessness is less well documented. Burgess and Holmstrom noted as far back as 1974 that 44 of the 92 women who took part in their landmark study on rape trauma syndrome moved house shortly after the assault. However, little has been done since then to examine the effects that relocation might have on women in this situation. Similarly, while support organisations are aware of adult survivors of child sexual abuse who move repeatedly in an effort to stay safe, it appears only peripherally in the research literature.
Farley and Kelly (2000) identified safe housing as a priority for women and girls trying to exit prostitution. For women working in prostitution, there are other compounding factors. Often precipitated into prostitution by poverty and/or abuse, women may then find themselves unable to leave if they incur fines as a result of being arrested for soliciting (Routes Out of Prostitution, 2003).
3.2 Consequences of violence against women for children and young people
Children and young people are affected by violence against women whether they are living with it or not. Young people who are affected by violence in their own lives are most likely to turn to other young people in the first instance for support (Young Women's Centre, 1997). The attitudes of young people are shaped by the society they live in, and many young people grow up believing that at least a degree of violence against women is acceptable (Burton and Kitzinger, 1998).
In a recent Minnesota study of women who had experienced violence by a partner, 114 women were interviewed about the extent to which their children were involved in the violence - 21% reported that their children had tried to get help when they were being assaulted, and 23% reported that their children had been physically involved in the events. Factors affecting children's attempts to intervene included the stability of the woman's financial and social situation, the frequency of the abuse, and the impact of the abuse on her life and health. Children were more likely to intervene when their mother's financial/social situation was less stable, when the abuse was frequent and the greater the impact on the woman's life and health. Children were less likely to intervene if the abuser was their biological father (Edleson et al, 2003). The authors suggest that these findings demonstrate a need for more careful assessments by all of the agencies involved in responding to domestic abuse, in order to improve the safety of women and children.
In this country too, there has been a development of work which considers the implications of domestic abuse for child protection (Hester, 2000; Humphreys, 2000; McGee, 2000) and examines the need for child protection services to develop an awareness of domestic abuse. In particular, it is argued, service providers must begin to understand how the abuse of women and children is used by abusive men to maintain control over them. In doing so, it is suggested, they will have to dismantle some of their own prejudices and misconceptions about where the responsibility for domestic abuse resides and place it back where it belongs - with the abusive man (Humphreys, 2000).
Forman (1991) viewed this from the other side of the glass, so to speak. She interviewed women whose children had been sexually abused and found that all of the women had experienced some form of abuse by their partner, and 17 of the women had experienced physical violence. This suggests that in addressing issues of child protection, it may also be necessary to consider whether the mother of an abused child may herself be in need of protection and support.
The effects on children of living with domestic abuse include difficulties with sleeping and eating, disruptive or very withdrawn behaviour, and delayed development (Hague, Kelly et al, 1995). Scottish Women's Aid has produced a series of reports based on the perspectives of children themselves, and it is clear from these that many children also experience high levels of fear and anxiety. The extent to which abusive men control the behaviour of children as well as women is also evident (Scottish Women's Aid, 1996).
Women's experiences, as indicated already, do not fit neatly into boxes. Nor do the experiences of children. As with research into violence against women, studies looking at childhood experiences of violence focus largely on a single type of violence (Saunders, 2003). Multiple experiences of violence and relationships between different types of violence are less explored, and Saunders advocates that, as with research into adult experiences of violence, a more integrated approach to researching how violence affects children and young people is required and should be developed (Saunders, 2003).
The impact of childhood and adult experiences of violence on women's mental health has already been acknowledged, including the links to substance misuse. Looking at the effect this then has on children provides another way to understand the effect that violence against women has on children. For example, the importance of recognising and addressing the impact on children of living with a parent with substance misuse issues has been acknowledged by the Scottish Executive (Scottish Executive, 2003).
Violence against women can be seen as an underpinning cause of difficulty for children across a broad spectrum of issues currently being addressed by the Scottish Executive.
3.3 Consequences of violence against women for society
Crisp and Stanko (2000) observe that relatively little research had been done into the financial implications of domestic violence, particularly within the U.K., and that within the body of research which had been done, there are wide variations in the methodologies adopted. They raise questions about the lack of accurate baseline data on which to base such research, and argue the need to move beyond awareness raising of the impact of domestic violence, and to develop effective monitoring systems which allow the cost and the benefits of different interventions to be measured. They also suggest that studies carried out in one part of the country can have relevance in another, that extrapolating and contextualising data can help to avoid needless duplication - "a broken arm is a broken arm in the city as well as in the country" (Crisp and Stanko in Taylor-Browne (ed.) 2001: 354)
Subsequently, the Women and Equality Unit commissioned research into the economic costs of domestic violence. The research aims to put a monetary value on the 'cost' of domestic violence. Methodologically, the research draws on the Home Office approach to costing crime, and applies this to data drawn from the 2001 British Crime Survey report on domestic violence. Although the full report is not yet available, 12 an interim report estimates that the cost of each 'female domestic homicide' is £1.1 million, based on lost economic output, the use of public services, and 'the human and emotional impact' (Walby, 2002). 13
If there is little on the economic impact of domestic violence, there is even less on sexual violence. An extensive database search yielded only one article specifically on the costs of rape. Post et al, writing in response to a Supreme Court ruling that rape was a "noneconomic violent crime", estimated the financial costs of sexual violence, including 'sex offense homicides' in the state of Michigan for 1996 to be $6.5 billion. This was based on an estimated 61,581 rapes and sexual assaults for that year, and estimates of $87,000 as the cost of a single incident of rape/sexual assault, and more than $3 million as the cost of a single 'sex-offense homicide'. The higher cost of homicide is attributed to the greater costs of health care, loss of productivity and lost quality of life (Post et al, 2002).
The impact of violence against women on society is about more than financial costs. Riger et al (2002) describe the "radiating impact" of domestic violence, based on an ecological approach to the issue. Using the image of a wheel, they identify the 'first order' effects, i.e. the effects on the woman, as the centre of the wheel. Second order effects, the spokes, represent the impact of violence on a woman's relationships with others, including her ability to function socially, educationally and economically. The outer rim of the wheel they identify as the third order effects, and this relates to the impact on other people in a woman's life, including her children and other family members (Riger et al, 2002). All layers in the wheel are connected. The authors use the example of how intimidation of family members (on the rim) is used by abusers to continue to control the woman. If the woman is reliant on family members for childcare to allow her to work, their reactions to intimidation by her abuser may have more than one effect on her, i.e. the fear such intimidation may engender in her, but also the threat to her livelihood (Riger et al, 2002: 196-198).
3.4 Consequences of violence against women - conclusions
- Violence against women has a significant impact on the lives of individual women, their health, their safety, their self-esteem, and their ability to participate in society
- Violence against women impacts on the lives and development of children, either as a result of witnessing violence against their mothers or other significant women in their lives, or because they themselves experience violence as a result of living in a violent world
- The scale of violence against women implies the need for a response from the whole of society which is only just beginning to be acknowledged, both through the provision of appropriate interventions and the development of strategies to challenge and change the situation.