Battered bodies & shattered minds: violence against women in Bangladesh.
Article Type: Clinical report
Subject: Abused women (Psychological aspects)
Abused women (Health aspects)
Violence (Prevention)
Violence (Laws, regulations and rules)
Violence (Varieties)
Authors: Wahed, Tania
Bhuiya, Abbas
Pub Date: 10/01/2007
Publication: Name: Indian Journal of Medical Research Publisher: Indian Council of Medical Research Audience: Academic Format: Magazine/Journal Subject: Biological sciences; Health Copyright: COPYRIGHT 2007 Indian Council of Medical Research ISSN: 0971-5916
Issue: Date: Oct, 2007 Source Volume: 126 Source Issue: 4
Accession Number: 173276706
Full Text: Violence against women is a common and insidious phenomenon in Bangladesh. The types of violence commonly committed are domestic violence, acid throwing, rape, trafficking and forced prostitution. Domestic violence is the most common form of violence and its prevalence is higher in rural areas. A higher prevalence of verbal abuse than physical abuse by partners has been observed. The reasons mentioned for abuse were trivial and included questioning of the husband, failure to perform household work and care of children, economic problems, stealing, refusal to bring dowry, etc. The factors associated with violence were the age of women, age of husband, past exposure to familial violence, and lack of spousal communication. The majority of abused women remained silent about their experience because of the high acceptance of violence within society, fear of repercussion, tarnishing family honour and own reputation, jeopardizing children's future, and lack of an alternative place to stay. However, severely abused women, women who had frequent verbal disputes, higher level of education, and support from natal homes were more likely to disclose violence. A very small proportion of women approached institutional sources for help and only when the abuse was severe, became life threatening or children were at risk. Interestingly, violence increased with membership of women in micro-credit organizations initially but tapered off as duration of involvement increased. The high acceptability of violence within society acts as a deterrent for legal redress. Effective strategies for the prevention of violence should involve public awareness campaigns and community-based networks to support victims.

Key words Bangladesh--policies--sexual abuse--violence--women

**********

In the last two decades violence against women, (gender-based violence) has emerged as the most pressing and intractable social problem across regional, social and cultural boundaries (1,2). Violence against women is recognized as a serious human rights violation and a pervasive public health problem that concerns all sectors of society. Irrespective of a nation's level of development, women are susceptible to exploitation, oppression and other types of demeaning violence from men in all societies where cultural norms, tradition and the legal system endorse women's subordination to men. In the South Asian Region, violence against women begins long before they are born and continues throughout their lives. The lives of unborn girls are terminated through sex selective abortions. Every sixth death of a female infant in India, Bangladesh and Pakistan is due to neglect and discrimination (3). In the Region, females face restrictions in mobility, usually have less to eat than their male counterparts, are denied proper education and health care, are often forced into early arranged marriages, have few opportunities of employment and are underrepresented in the governments. All too frequently women are targets of extreme forms of aggression such as incest, rape, public humiliation, trafficking, acid attacks and dowry deaths (3). Bangladesh is no exception. Violence against women is a common and insidious phenomenon in Bangladesh. Newspaper reports in the country clearly indicate that the prevalence of violence against women is very high. Evidence from research supports the above conviction. It is against this background that we have reviewed published literature on violence against women, primarily focusing on domestic violence, to assess and explore the prevalence, causes and consequences of the problem and to understand the strategies undertaken by the women victims themselves in order to plan effective prevention and intervention strategies. Furthermore, the knowledge of effective attitudes and practices among health service providers and law enforcers to support victims of violence is crucial for the formulation of appropriate policies and programmes.

Prevalence and forms of violence

Gender violence, namely violence against women, is now defined very broadly to include anything "involving use of force/coercion with an intent of perpetuation/promotion of hierarchical gender-relations in all social structures: family, community, workplace and society" (4). United Nations Commission of the Status of Women defined violence against women as "any act that results in, or is likely to result in physical, sexual or psychological harm or suffering to women" (5-7).

The types of violence commonly committed against women in Bangladesh are, domestic violence, acid throwing and burning, sexual harassment and indecent assault, rape, kidnapping and abduction, trafficking and importation for immoral purposes and forced prostitution. In every country where reliable large-scale studies have taken place, domestic violence has been found to be by far the most common, widespread and far-reaching form of gender based violence. Violent behaviours such as beating, pushing, slapping or throwing things by family members are considered domestic violence (8). Domestic and intimate partner violence involves physical, sexual and emotional abuse against women in the home, within the family or within an intimate relationship.

Several studies have indicated that domestic violence against women, especially violence perpetrated by a woman's husband, is a serious problem in Bangladesh (9-12). Although domestic violence includes child abuse, parent abuse and in-law abuse committed by male aggressors on female victims, available information from research indicated that the "most common type of violence in Bangladesh against women is domestic violence perpetrated by intimate partners or ex partners (13)". Many Bangladeshi women endure daily beatings, harassment for dowry, verbal abuse and acid attacks for refusing to comply with male demands. For many, home is not a haven but a place of pain and humiliation, where violence is an integral part of everyday life hidden behind closed doors and avoided in public discussion.

The estimated prevalence of physical violence against women by husbands in Bangladesh varies between 30 and 50 per cent". In earlier research the estimates on the magnitude of physical violence against women in Bangladesh were either focused on a particular socio-economic group or did not relate exclusively to intimate partner violence. The study by Schuler et al (14) addressed the prevalence of violence amongst economically disadvantaged women in rural Bangladesh. The study reported that 47 per cent of the women were beaten by their husbands in their lifetime and the prevalence of current abuse of women was 19 per cent in 1994. A different study by Steele and colleagues found the rate to be 32 per cent in another rural area of Bangladesh (15). Koenig et al (12) included women from all socio-economic strata but explored factors associated with abuse of women by the husband and in laws. However, the time period for abuse was not explicitly specified in this study. They reported that 42 per cent of the violence was perpetrated by the husband or the in-laws. Azim (10) observed the rate of violence to be 61 per cent in the urban areas of Bangladesh. However, the questions used in the earlier studies, with the exception of the study by Azim (10), did not have a direct and behaviourally explicit definition of physical violence, thus leaving the meaning of violence open for interpretation. As there were methodological differences among the earlier studies, the estimates of the magnitude of spousal physical violence were incomparable.

Using data from the most comprehensive population based survey of 2,702 women conducted in 2000-2001, collected in collaboration of ICDDR,B and Naripokkho, Naved et al (9) have shown a high prevalence of lifetime spousal physical violence against women, 40 per cent in the urban and 42 per cent in the rural areas of Bangladesh. About 19 per cent of urban and 16 per cent of rural women reported experiencing current physical abuse, defined as abuse during the twelve months prior to the interview. In both the urban and rural sites, 19 per cent of the women had experienced severe physical violence defined as hit with fist or something else or kicked, dragged or beaten up or choked or burnt or threatened or actually injured by a weapon or some other tool. For most of these women physical violence was not a one-time event. This was true regardless of the type of assault, ranging from slapping to burning. About 10 per cent of the ever-pregnant women in the urban and 12 per cent of such women in the rural area reported being physically abused during pregnancy. This suggests that physical violence does not escalate during pregnancy.

In another analysis using the ICDDR, B-Naripokkho survey data, Naved et al (16) observed that of the abused pregnant women, 37 per cent in Dhaka and 28 per cent in Matlab were punched or kicked in the abdomen.

In a study by Bhuiya and colleagues (11), it was reported that 50.8 per cent of the women were battered by their husbands, 2.1 per cent by other family members, and a small proportion (5.2%) by both their husbands and other family members. Hitting was mostly done with the use of hand (85.4%), followed by stick (63.5%), and kicking (51%) alone or in combination. The study reported that 20 per cent of women were physically abused by their husbands during pregnancy and 2.6 per cent by other family members as well. About, 13.7 per cent were throttled by their husbands and 2.1 per cent by other family members.

In a recent study, Ahmed (17) explored the effect of micro credit based development programmes of Bangladesh Rural Advancement Committee (BRAC) on domestic violence. The study reported that 17.5 per cent of women had experienced spousal violence during the study period of 4 months in 1999 and the proportion of abused women was greater among BRAC households (P=0.05). The study also demonstrated that violence decreased with age, more so for the poor non-member households. In addition, irrespective of BRAC membership status of the household, women with some schooling had reported less violence, while women with children had reported more. Furthermore, it was observed that, women above 35 yr of age, schooling of the household head, and the non-deficit economic status of the household were associated with a lower level of reported violence. Nevertheless, the study concluded that the higher level of domestic violence reported during the initial stages/phase of BRAC membership declined over time with the introduction of skill development or training among the women participants.

In a recent WHO multi-country study, based on the ICDDR, B-Naripokkho survey data, it was reported that more than 50 per cent of the women in Bangladesh were subjected to physical or sexual violence by intimate partners (18). The lifetime prevalence of partner violence (either physical or sexual) was observed to be 53.4 per cent in the urban and 61.7 per cent in the rural areas of Bangladesh. The magnitude of lifetime prevalence of physical violence, alone, against women was estimated to be 39.7 per cent in the urban and 41.7 per cent in the rural areas. The percentage of ever-partnered women, who had experienced severe physical violence, was 18.7 per cent for the urban and 19.4 per cent in the rural areas of Bangladesh. The most common act of violence reported by women was being slapped or having something thrown at them. In general, as the severity of the act of violence increased; the percentage of women having experienced that particular act decreased.

Further analysis of the ICDDR, B-Naripokkho sample survey of women in reproductive age group women revealed that the majority of women in Dhaka (60%) and Matlab (61%) had reported sexual or physical abuse or both by their husbands (19).

Emotional abuse: Emotional or psychological violence includes intimidation and humiliation where the victim is made to feel bad about oneself and is subjected to excessive controlling, curtailing and/or disruption of routine activities such as sleeping or eating habits, social relationship, access to money; verbal insults and so on (18,20).

In the study by Bhuiya and colleagues" a higher prevalence for verbal abuse than physical abuse against women was observed. The study reported that 66.8 per cent of women in the survey were verbally abused by their husbands and 23.7 per cent by other members of the family. In verbal abuse demeaning words were used only against the women in 70.9 per cent of the cases, and in 17.9 per cent of the cases the abusive words were used against only the parents of the women. About 6.8 per cent of the women in the study area were abused verbally on a daily basis, and for 37.8 per cent the abuse was occasional. Infrequent verbal abuse was suffered by 16.8 per cent of the women. In addition, it was observed that, during pregnancy, 37.9 per cent of the women endured verbal abuse from husbands and 11.6 per cent from other members of the family.

Analyzing data from the ICDDR, B-Naripokkho survey, the WHO multi-country study findings reported that two thirds or more of the women had experienced repeated emotional abuse for either a few or many times (19). Usually, the acts most frequently mentioned by women were insults, belittling and intimidation. In general, physical abuse in intimate relationships was almost always accompanied by severe psychological and verbal abuse. In addition, the findings from the study suggest that a woman who suffered violence, physical or sexual, from an intimate partner was significantly more likely to experience severe constraints on her physical and social mobility or, in other words, highly controlling behaviours by the intimate partner.

Qualitative research has consistently revealed that women often consider emotional abuse to be more devastating than acts of physical violence.

Sexual abuse: From the analysis of the ICDDR-B-Naripokkho survey data, Naved et al (16) reported that 37 per cent of the ever-married, reproductive age group women in Dhaka and 50 per cent in Matlab had experienced sexual violence within marriage.

From the ICDDR-B-Naripokkho survey data the WHO multi-country study reported that the percentage of women who had experienced sexual abuse by an intimate partner during their lifetime was 37.4 per cent in the urban area and 49.7 per cent in the rural areas of Bangladesh (18). Substantial overlap between sexual and physical violence was observed. Out of four victims of domestic violence, one was also observed to have experienced sexual abuse. In the rural areas of Bangladesh, women reported more sexual violence than physical violence and a high proportion of the women (33%) in the rural areas reported sexual violence only by an intimate partner.

Reasons for domestic violence

Bhuiya et al (11) reported that the most frequently mentioned reasons for verbal abuse included wife's questioning of the husband on day to day affairs (29.1%) followed by failure of the wife to perform household work satisfactorily (17.6%), economic hardship of the family (11.5%), failure of the wife to take proper care of the children (10%), not wearing the veil or conforming to other expected behaviour (5.5%), inability to bring money from the natal home (3%), not taking good care of in-laws and relatives (2%). The rest of the reasons included supporting the natal home, failure to have children, having too many children, dark complexion of children, family feud between the husband's family and natal family and suspected relation with others. Similarly, the most frequently mentioned reasons for physical abuse included questioning of the husband (29.9%), failure to perform household work satisfactorily (18.8%), economic problems (9.4%), failure to take care of children properly (5.1%), stealing (3.4%), refusal to bring dowry from natal home (2.6%).

Jahan (21) reported that family quarrel was the reason given by most of the women as the cause of violence. Family quarrel was referred to as a value term that included many specific reasons such as husband's dissatisfaction over wife's management of expenses, house keeping and child care, jealousy and suspicious nature, exaggerated sense of superiority over a wife as reflected in household decision making, sexual maladjustment and such. The reasons for discord were found to be exacerbated by stress resulting from prolonged illness in the family, sudden financial loss, loss of job and aggravation by in-laws. Non-payment of dowry was the next most important cause mentioned for violence.

Factors associated with domestic violence

Bhuiya et al (11) reported that the age of women, age of husband and membership of women in micro credit organizations had significant relationship with husbands beating of wives in univariate analysis. In the multivariate analysis, age of the husband and membership of micro-credit organizations showed statistically significant relationship with battering. Further, the study revealed that the odds of being beaten was lowest for women with husbands aged 50 yr or more and highest for women with husbands less than 30 yr of age (6 times more than the former). Moreover, women who were members of micro credit organizations were two times more likely to be abused by their husbands than those who were non-members.

Conversely, results of a BRAC study, by Hadi (22) reported that the prevalence of physical violence against participating women was 19 per cent compared to 40 per cent among non participating eligible non members. The risk was also lower for sexual violence among members (23). HoweVer, some studies contest this argument. Chowdhury and Bhuiya (24) studied data from 2,038 currently married women aged 15-55 yr and reported a higher level of physical violence among BRAC members than among non members. Nonetheless, it was observed that the prevalence of violence tended to decrease as the duration of membership with the micro credit organization increased. Violence peaked when membership was initiated but tapered off when other inputs such as training were offered. The logic behind the assumption was that greater economic opportunities for women, ensured through access to credit, awareness increasing activities and skill training, would enhance the self-esteem and status of women within households, improve spousal relationships and therefore reduce domestic violence (17).

Despite the lack of effect of women's contribution to family support on domestic violence in a multivariate analysis, an ethnographic study by Schuler et al (25) showed that group based credit programmes could reduce men's violence against women. These programmes could reduce women's vulnerability to violence by strengthening their economic roles and making their lives more public. However, women's role as wage earners was capable of provoking hostility and violence in the husband when and if the traditional gender norms were challenged by the women. Further, in the rural area it was observed that earning an income increased the risk of being abused. Khan (26) reported similar findings explaining that it may undermine the power relations between the couple and provoke violence in husbands.

Naved & Persson (27), from the previously mentioned ICDDR,B-Naripokkho population based survey data and in depth interviews of 28 abused women investigated the factors associated with domestic violence in urban and rural Bangladesh. Multilevel analysis revealed that in both areas, the strongest factor associated with husband's violence against the spouse was the history of abuse of the husband's mother by his father. The study reported that men witnessing physical violence in the family during childhood were 2.29 times more likely to physically abuse the wife. Past exposure to familial violence probably endorses poor emotional development in the perpetrator, or may result in adoption of physical strategies as a means of coping with conflict. Lack of spousal communication was the second most important factor associated with domestic violence. Nonpayment or partial payment of dowry also, increased the likelihood of the wife experiencing abuse. Educational attainment of the husband, and the age of the woman were the other two important variables associated with violence, however, the influence was observed to vary by region. In the urban area, husband's education above the sixth grade had a protective effect but in the rural areas educational attainment only had such an effect when the husband had a very high level of education probably because conventional gender norms are stronger in the rural areas. In the urban area participation in savings and credit groups created tensions and increased the risk of abuse. Further, women being younger than the husband had a negative effect on violence for urban women but not for the rural women. In the urban areas those in the younger age group of 15-19 yr were more likely than older women to be abused. Previous studies have shown similar findings, that a woman's relative youth was an important risk factor for violence within marriage. The explanation put forward is that women who are married at a young age usually start their marital lives in their husband's family home and as young wives their status is quite low. The findings from the study by NaVed et al also indicated that the presence of in-laws in the household was not associated with spousal abuse. Koenig et al (12) found that extended family structure was negatively related to violence in Bangladesh. Further, Naved et al (12) observed that once other factors were controlled for, household income was found to have no effect on physical violence.

Amin et al (28) assessed the impact of female education and mobility on the extent of female abuse in the rural areas of Bangladesh. The study reported that 78 per cent of the women had encountered verbal abuse, 42 per cent physical abuse and 18 per cent were threatened with divorce. Women were found to be more likely to be abused if they were older, not educated or less mobile and if their husbands were less educated and less wealthy. Women's ownership of land and husband's occupation were also statistically significant predictors of abuse.

Reasons for acceptance of violence

The WHO multi-country study (18) reported, from the data collected by ICDDR,B-Naripokkho survey, that 53.3 per cent in the urban and 79.3 per cent of women in the rural area believed that a man had a right to beat his wife under certain circumstances. The various circumstances included were--not completing housework adequately, refusing sex, disobeying the husband, or being unfaithful. The most commonly agreed upon reason for which beating was accepted as justified by women, was when the wife was unfaithful. The proportion of women who agreed that wife beating was justified under certain conditions was higher among women who had experienced partner violence than among those who had not. This may be an indication that women learn to accept or rationalize violence in circumstances in which they are victims themselves, or that women are at a greater risk of violence when it is accepted as the norm by a significant number of people. The circumstances under which wives have the right to refuse unwanted sexual demands from their husbands were also explored, e.g., situations including if she was sick, if she did not want to, if the husband was drunk, or if he mistreated her. Fewer women felt that a wife had the right to refuse sex because she did not want to than when her husband was drunk or abusive. About 45 per cent of urban and 35.8 per cent of rural women agreed that a wife had a right to refuse sex under all circumstances mentioned and 5.3 per cent urban and 11.3 per cent rural women agreed to none

The reasons for silence in abusive relationships and health seeking behaviour

The underreporting of violence is of serious concern and the major reasons for silence in abusive relationships are (i) the universal acceptance of gender inequality reinforced by social and religious beliefs in favour of male dominance; (ii) the deep seated reverence for family as an institution for continuing the existing social order and values related to regulating sexual behaviour, particularly female sexuality within the bounds of marriage; (iii) the prevalent bias amongst the middle class, who form the bulk of lawmakers, law enforcers, social norm setters and opinion makers, in regarding domestic violence as inconsequential; and (iv) the fear of social censure and loss of face (20).

The nature of family relationship forces women to struggle alone against violence where society encourages the husband to exercise his right to dominate and control the wife (29). In general, women are socialized to accept physical and mental abuses as part of a husband's marital prerogatives. Entrenched social attitudes define violence against women as normal practice and not as a punishable crime. Underlying the acceptance of violence against women is the deep-rooted social belief that women are fundamentally of less value than men. Violence against women is typically enmeshed in a complex web of institutionalized social relations that make women particularly vulnerable to it (14). Violence against women is rooted in the patriarchal ideology, which to a large extent is responsible for the internalization of female inferiority through a process of socialization, customs, religious laws and rituals. The legitimization and promotion of rigid gender roles, masculinity, toughness and male honour within the society perpetuates gender violence (30). The social forces that perpetuate gender based violence can be viewed from various approaches viz., cultural values that tend to rationalize violence, legal systems that are not sensitive to the rights of women, economic systems that subordinate women and political systems that marginalize women's need (7). The shame associated with domestic violence, rape and other forms of abuse may contribute to the fact that women are afraid of repercussions and stigma and consequently never tell anyone and often suffer the abuse in silence. Women with lowered self-esteem, severe depressive symptoms with minimal personal resources and least institutional support are more vulnerable to be abused by her spouse (31).

From the population based cross-sectional survey data of ICDDR-B-Naripokkho, Naved et al (9) reported that most of the abused women (66%) were silent about their experience. In addition, 75 per cent of the moderately abused women in the urban area and 86 per cent of them in the rural area never talked about their experience to anybody. The main reasons behind this silence was the fear of tarnishing family honour, stigma of ruining the woman's own reputation, securing the children's future, fear of repercussion from the husband, hopelessness, expectation that things would change, threat of murder, and belief that violence was the husband's prerogative. A high percentage of women (30% urban and 40% rural) reported that they had remained silent because of the shame they had felt, the fear they had of not being believed or being blamed for the violence. Moreover, 26 per cent of urban and 34 per cent of rural women felt that such disclosure would detract from family honour. Fear of repercussion was present particularly in cases where the violence experienced was severe.

Coping strategies of abused women

The findings from the analysis indicated that a third of the women did disclose or talk about their experience of violence (9). Severely abused women were more likely to disclose violence, almost 3 times as much as moderately abused women in the urban area and 8 times more than those in the rural area. Women who had frequent verbal disputes with their husbands were more likely to disclose violence in both urban and rural (5 times) areas. In the rural area, women with education above 10th grade were 3 times more likely to disclose their experience compared to women with no education. The study reported that only 2 per cent ever sought help from any institutional sources. The women who sought help from formal sources generally did so because they could not endure anymore (79% in urban and 84% in rural areas), husband threatened to do something to the children or actually hit children (32% in urban and 37% in the rural area), or because she was badly injured or afraid of being killed (21% urban and 31% rural women). The study revealed that about 60 per cent of urban and 51 per cent of rural abused women never received any help from others. Moreover, those who could count on their natal homes for support were more likely to report violence. Women experiencing both physical and sexual assault were more likely to disclose their sufferings. Unlike other studies the survey found that age, income earning by the women or history of violence in the woman's family was not associated with disclosure of violence in any sites.

Naved et al (9) reported that women in both urban and rural areas were more likely to talk to their parents (18% in urban and 19% in rural area) and siblings (16% in urban and 14% in the rural area). The abused women mentioned in-laws, parents and siblings as those from whom they received help most often. In the rural area, as women generally married outside their villages and lived away from their natal family, a high proportion of the rural women talked to their in-laws about the abuse (16%). In addition, 10 per cent of urban and 11.5 per cent of rural women disclosed violence to their neighbours. It was revealed that disclosure was not always a means of stopping violence but simply a coping strategy of the abused women to get some relief from mental stress. Urban women were less likely to use formal sources of support such as a local leader than the rural women, probably because they did not feel connected to a local leader in a high-density area.

Bhuiya et al (11) found that in 23.9 per cent of the abuse cases, the family members undertook a direct mediating role and appealed to the husband to not to abuse the wife. Younger children would cry (16.7%) when witnessing verbal and physical abuse and in 12.3 per cent of the cases would remain quiet. Grown up children would normally be on the mother's side and would confront the father. In 25 per cent of the cases, neighbours would advise the husband not to abuse the wife, in 30.8 per cent of the cases would remain indifferent and in 5 per cent of the cases would be unsympathetic to the woman. In 36 per cent of the cases the neighbours would intervene and take initiatives to stop the abuse. The neighbours in 43.2 per cent of the cases arranged arbitration to mitigate the problem.

Health consequences of violence

Men's violence against women compromises women's health and well-being in a wide variety of social settings (7,8,13,32). A growing body of evidence documents the consequences of violence on women's health and well-being, ranging from fatal outcomes such as homicide, suicide, and AIDS related deaths to non fatal outcomes such as physical injuries, chronic pain syndrome, gastrointestinal disorders, unintended pregnancies, pregnancy complications and sexually transmitted infections (31).

Psychiatric morbidity, such as depression, stress related symptoms, chemical dependency and substance abuse and suicide are consequences observed in the context of violence in women's lives over time (33). Abused women are more likely than others to suffer from depression, anxiety, psychosomatic symptoms, eating problems, sexual dysfunction and many reproductive health problems, including miscarriage and stillbirth, premature delivery, HIV and other sexually transmitted infections, unwanted pregnancies and unsafe abortions. Consequences of abuse, such as HIV/AIDS or unplanned pregnancies, may in themselves act as risk factors for further aggression, forming a cycle of abuse. Further, adolescents who had experienced sexual abuse were more likely to experience it again later in life.

At a global level, the health burden from violence against women aged 15-44 yr is comparable to that posed, in the age group, by HIV, tuberculosis, sepsis during child birth, cancer or cardiovascular diseases. Violence against women is a major cause of death and disability for women 16 to 44 yr of age (34). The World Bank estimates that rape and domestic violence together account for 5 per cent of the healthy years of life lost to a woman of reproductive age in developing countries (35). The economic cost Of violence against women is substantial involving direct medical and health care services, productivity loss leave alone the pain and anguish involved. Violence against women impoverishes individuals, families, communities and governments and reduces the economic development of each nation.

Bhuiya et al (11) found that nearly half (47.9%) of the women who were physically abused by husbands suffered injuries from the assault. In 78.3 per cent of the cases it was necessary to consult a health care provider (11). Approximately 57.1 per cent of the women who suffered from beatings by other family members sustained injuries and for nearly half of the cases, a health care provider was consulted.

About 25 per cent of physically abused women reported having lost children after they were born alive compared to 18 per cent women who were never abused (16). In Bangladesh, 13.8 per cent of maternal deaths in pregnancy were reported as resulting from injury/violence (36). In addition, children who witnessed violence were at a higher risk for a whole range of emotional and behavioural problems including anxiety, depression, poor school performance, low self-esteem, disobedience, nightmares and physical-health complaints. Paltiel (37) found that severe beatings usually perpetrated by the husband accounted for 49 per cent of household deaths in Bangladesh.

The WHO multi-country study (18), from the ICDDR,B-Naripokkho survey, reported that among the physically abused, 26.7 per cent in the urban and 24.8 per cent in the rural area reported that they had been injured as a consequence of an assault by an intimate partner. The majority of ever-injured women reported minor injuries (bruises, abrasions, cuts, punctures and bites), however, in some areas more serious injuries were relatively common. Data from the WHO study demonstrated that far from being an isolated event, most acts of physical violence by an intimate partner were part of a pattern of continuing abuse. About 16 per cent of ever abused women in the urban area and 11.6 per cent in the rural area reported that they had been injured more than 5 times as a consequence of an assault. In Bangladesh, at least 50 per cent of ever injured women reported that they had lost consciousness because of a violent incident. Of those who had ever been injured by an intimate partner, 80.4 per cent in the rural area and 68.5 per cent in the urban area reported having needed healthcare for an injury (whether healthcare was actually received or not).

In the WHO multi-country report (18), based on the ICDDR,B-Nafipokkho survey data it was revealed that 27 per cent of the urban and 25 per cent of the rural physically-abused women were injured as a consequence of violence. The injuries ranged from cuts and bruises and bites to broken limbs, broken teeth and bums. Married women who reported physical or sexual abuse by their husbands were 1.5 times (95% CI: 1.2-1.7) more likely to report problems in walking, 1.7 times (95% CI 1.4-2.0) more likely to report pain and 1.7 times (95% CI 1.5-2.1) more likely to report dizziness and 1.4 times (95% CI 1.2-1.6) more likely to report problems with memory.

Naved et al (16) observed from the ICDDR, B-Naripokkho survey that approximately 13.8 per cent of maternal deaths in pregnancy resulted from injury and violence. The 15-19 yr old women who were pregnant, or had recently given birth were nearly 3 times more likely to die from violence inflicted by others than the women of the same age who were not abused.

Mental health

Using a self-reporting questionnaire of 20 questions (SRQ-20) developed by the WHO as a screening tool for emotional distress, the mean score for women who had experienced abuse was found to be significantly higher than that for non abused women in both areas (7.9 as compared to 5.4 in the urban areas and 7.4 compared to 5.2 in rural areas) (18). Women who reported violence by an intimate partner were significantly more likely than women who had not experienced violence to report their general health to be poor.

From the survey data by ICDDR,B-Naripokkho, Naved et al (16) reported that 16 per cent of only physically abused ever married women, 12 per cent of only sexually abused and 28 per cent of those who had experienced both forms of violence reported suicidal thoughts in comparison to 7 per cent of those ever married women who never experienced any violence. In addition, it was reported that women who were physically or sexually abused by their husbands did much worse on a scale of mental health compared to those who had not experienced any forms of violence. However, 20 per cent of only physically abused women, 22 per cent of sexually abused and 35 per cent of those abused both ways reported attempting suicide compared to only 14 per cent of non-abused women (16).

Other forms violence against women

Acid attacks: Every week more than ten women in Bangladesh suffer from an acid attack that leaves them brutally disfigured and disabled (38). In 1999, there were 178 such instances reported and 101 in 1998 (39). In 2000, Odhikar, a human rights organization, recorded from reports of 8 daily national newspapers 186 incidences of acid attacks (40). Many of the victims were children between the ages of 6-15 yr. In a majority of the cases the main reasons cited were jealousy, refusal of sexual advances and revenge after an argument and dowry disputes.

Rape: Early sexual abuse is a highly sensitive issue that is difficult to explore in survey situations (18). The WHO survey reported that levels of sexual abuse before the age of fifteen years ranged from 1 per cent or less in rural Bangladesh to 7.4 per cent in Bangladesh city. Strangers were the most frequently mentioned perpetrators in the urban areas of Bangladesh. However, very low level of sexual violence by non-partners for women over 15 yr was reported (0.5% in rural and 7.6% in urban areas) in Bangladesh (18). Higher level of non partner violence was reported in the urban than in the rural areas. More than a fifth of the respondents in the urban area reported being physically or sexually abused by a non partner in Bangladesh. Generally the most frequently mentioned perpetrators were acquaintances or strangers. In the rural areas of Bangladesh, more than a fifth of the respondents who had experienced non partner physical violence reported that two or more perpetrators had assaulted them. The extremely low level of violence by non partners reported in Bangladesh may be a function of the great stigma associated with sexual violence in these rural settings. The early age of marriage prevailing in the society offers protection from the risk of sexual violence, through the guardianship of the husband. In 2002, a total of 1434 cases of rape were reported and 35 per cent were gang rapes and 10 per cent were rapes followed by murder (41).

Rape is the worst form of intimidation, the taboo surrounding sexuality and the "shame" and loss of honour generally prevent women from seeking justice in cases of rape. Moreover, the burden of proving the assault rests on the victim and the focus is more on the victim's character, possible provocation and temptation rather than the offence perpetrated by the assailant. The extent of this violence may be seen from the official figure of the rape victims (5,738 in 1991-1996) as quoted in the Parliament by the Home Minister (21). Taking into account the reasons for non reporting cited above, this might be a reflection of only the tip of the iceberg. The most shocking part is that 16.26 per cent, were minors. This is the manifestation of society's failure to protect female children from such devastating violation. Sexual abuse among young girls and adolescents is a violation of a young girl's basic rights and bodily integrity and may have profound health consequences such as behavioural and psychological problems, sexual dysfunction, relationship problems, low self-esteem, depression, thoughts of suicide, deliberate self harm, alcohol and substance abuse and sexual risk taking. In addition, women who are abused in childhood are at a greater risk of being physically and sexually abused as adults (18). Sexual abuse in childhood has also been linked to a range of negative reproductive health outcomes such as unwanted pregnancy and sexually transmitted infections, including HIV. A study in Bangladesh showed that 32 per cent of women working outside their homes experienced disruption of work due to incidents of domestic violence (36).

A distressing development in this area is the increased reports of incidents of violence against women committed by members of State agencies entrusted with the enactment of law and order in the country and the security of the people and the State. Instances of rape, sexual harassment, torture and other types of abuses, including physical and psychological violence and violation of basic human rights of women by members of the police, civil defense force and armed forces, including custodial rape, have given rise to apprehension and mistrust about the State's will and commitment towards protection of women against violence (21).

Early marriages

In Bangladesh where early marriages are arranged for a majority of girls, it was found that maternal mortality among teenage mothers was double the national figures. The trend of early marriages limits educational opportunities among adolescent girls and deprives them of opportunities in life.

Labour migration and trafficking

Trafficking in women and young girls includes all acts involved in capture and acquisition of girls and women for trade and transport with the intent to sell, exchange or use for such purpose as prostitution, sexual abuse, forced labour, bonded labour, slavery, etc. Three hundred thousand children from Bangladesh have been trafficked to brothels in India and 4500 women and children are trafficked to Pakistan annually (36). Sometimes the family are themselves responsible for trafficking of women and girl children. This problem though not new has increased tremendously in recent years. It is extremely difficult to assess the number of prostitutes in Dhaka but according to police sources their number in Dhaka and the adjacent port town of Narayangonj is nearly 50,000 (20). Some work in brothels and some are street prostitutes and some hotel call girls. Increasing poverty in the rural areas force young girls to come to the cities in search of work, which causes them to be vulnerable to the lure of false promises of pimps. The conservative attitude of the society which regards prostitutes as fallen makes rehabilitation of the prostitutes harder.

A study in Bangladesh on female labour migration and trafficking revealed that 82 per cent of the women went to the Gulf States with domestic visas and the rest went with company visas to work as cleaners in schools and hospitals (36). A majority of them admitted that sex work was an integrated part of their job and 10 per cent said that it was their only occupation (36). In addition, in population policy and programme implementation, poor rural women in many areas have been subjected to accepting methods of family planning, without proper counselling about side-effects, inadequate follow up and referrals. Instances of provision of obsolete or experimental methods also exist (20). On the other hand, many women have faced hostility from family members and community elders for accepting these methods. Instead of safeguarding poor rural women's health and welfare, the programme has, in effect, caused harm and violated women's rights to health and well-being (20).

Violence against women at the work place and loss of workdays due to violence

At the work place two major forms of violence take place (i) sexual violence, and (ii) violation of basic human rights of workers. Human rights violation include deliberate exploitation such as unequal and low wages, forced overtime work, harassment, deprivation of benefits such as maternity leave and locking the workers in during factory shifts. The informal sector especially the export oriented industries such as garments and shrimp farming offer alarming examples of this violation of basic rights (20).

In the study by Naved et al (16), 32 per cent of working women in Dhaka and 23 per cent in Matlab reported work disruption as a consequence of physical violence by husband (16), about 50 per cent of working women reported that they frequently suffered ill health because of violence and were frequently on leave. A study in Bangladesh showed that 32 per cent of women working outside their homes experienced disruption of work due to incidents of domestic violence (36).

Litigation

So far, the Bangladeshi legal system, based mainly on the British Penal Code and common law, has done little to diminish women's vulnerability to violence. The interventions undertaken by the government and non governmental organizations (NGOs) include the following: constitutional provisions, legislative actions, support services, shelter homes, advocacy, legal aid and counselling services. The Dowry Prohibition Act (1980), Cruelty to Women (Deterrent Punishment) Ordinance (1983), Family Court Ordinance (1985), and the Women and Children Repression (Special Provision) Act (1995) have all been very inadequately formulated and implemented, therefore the actual impact has been very marginal. However, it is worth mentioning that in Bangladesh, the Women and Children Repression (Special Provision) Act 1995, Section 8 provides a maximum penalty of life imprisonment and also, provides deterrent punishment for committing rape, murder and acid throwing.

However, in many instances of rape, molestation, abduction, and sexual harassment, technicalities and legal loopholes deprive women of justice, especially where the aggressors are in a dominant socio-economic position (20). Moreover, women's socio-economic powerlessness and ignorance of legal rights limit the scope of legal protection. In addition, litigation does not always guarantee that the desired judgment will be received. So far, imperfect understanding and inadequate knowledge of this multidimensional problem has produced limited success in this area. In many cases full enforcement and implementation of existing laws have not been achieved due to various factors, including the lack of awareness of women's rights among law enforcement agents. The scarcity of effective agencies offering supportive intervention and the excessive expenses and the time-consuming process involved in litigation also prevent many women, especially the poor and uneducated ones (who are the worst sufferers), from seeking redress through criminal proceedings. A review of interventions shows that the success of the attempts designed to reduce violence has been limited and the implications of the limited success very grave (20). However, interventions by NGOs and government have generated knowledge about the causes and consequences and the need to prevent violence against women. A review of successful methods of prevention and elimination of violence and effective legislative and policy intervention is absolutely necessary.

Existence of variations in definition and recognition of violence against women among the various agencies engaged in actions designed to resist/reduce violence against women prevails. The conventional categorization of abuse by the police is based on a very narrow definition. For instance, marital rape is not recognized in law and culture. Incest is a taboo topic. There are laws regarding rape and acid attacks but there are no clear law defining domestic violence. In Bangladesh, psychological violence and emotional abuse and deprivation and denial of basic rights, especially among married couples, is not given serious consideration either by family or police and therefore remains unreported. Wife-abuse (battering) is widely condoned and tolerated. These particular problems due to variations in definition and recognition make it extremely difficult to motivate people to contest violence against women and to formulate appropriate strategies (21). Tacit support by State agencies, especially in cases of custodial rape and State violence renders implementation of laws quite difficult (21). Of all the cases of rape in police custody brought to light so far, very few have been held accountable or chastised. Moreover, suppression of evidence by colleagues and foot-dragging over procedures by officials frequently helped the suspects in such cases. Collusion between local elites and religious leaders who abuse the traditional local arbitration procedure is one of the main factors that makes redress of abuse difficult.

Policies and programmes

Interventions to alleviate the problem of domestic violence have mostly been directed towards reducing the harm after the occurrence of the incident through medical treatment and counselling (11). Preventive efforts on the other hand have been limited to enactment of punitive legislative measures against the aggressor for abusing women. Although the occurrence of domestic violence directly involves husband, wife and other family members and is possibly perpetuated by the social context (22), a family and/or community approach to alleviate the problem has been largely absent so far.

Many of the contributing risk factors for sexual and physical assault have been identified in childhood and adolescence and these need to be incorporated into prevention goals. Peer attitudes, past experiences of child maltreatment and substance abuse in adolescence have all been linked to greater risk of domestic violence and sexual assault. Changing the norms and custom about relationships and providing the future generation with skills needed to foster healthy relationships is the only viable way to shift from a society where violence against women is widely condoned. Much of children's social learning takes place in schools and influences the development of behaviours and attitudes supportive of interpersonal violence in its many forms. Schools are an ideal place where primary prevention programmes can be introduced to a wide range of children. Discussion of personal safety and injury prevention in the classroom and integration of such discussion within the context of trusting relationships is necessary. Early and mid adolescence is another important phase of life when awareness of violence in relationships can be built and healthy ways of forming intimate relationships can be taught. As such some researchers agree that institutions should provide universal programmes to address and challenge violence against women.

Public awareness campaigns such as public service announcements and advertisements are common approaches to primary prevention providing information regarding the warning signs of violence and community resources for victims and perpetrators. Developing community-based networks for coordinating services and programmes are important in raising awareness for prevention of violence. Community-based educational activities should be used to raise awareness and knowledge about women's legal and social rights, and hence, empowering them to seek help for abuse. Effective implementation of legislation is constantly impaired by the lack of support from dominant interests in the community who legitimize violence as normal. Despite punitive legislation most women have to abide by the rules of a patriarchal social system, which reinforces gender inequalities.

Health providers are typically reluctant to ask women about experiences of abuse in fear of either offending the victims or getting involved into issues deeper than what they can handle. Providers often feel that they do not have the knowledge or skill to address the problem. They lack knowledge of national legislation as well as available services to which women may be referred. Poor co-ordination among and across agencies and institutions makes the process of negotiation complex, inefficient and confusing. In addition, there are only a few counsellors in Bangladesh with relevant training, thus capacity building should be prioritized.

Summary & conclusion

A very high level of domestic violence is prevalent confirming it as a major social and public health problem in Bangladesh. Recurrent physical and emotional abuse is widespread. A higher prevalence of emotional or verbal abuse than physical abuse has been reported in some studies. Existence of a substantial level of sexual abuse and overlap of sexual and physical abuse within marriage has been observed. In addition, a very low level of reported non-partner sexual abuse in Bangladesh has been revealed in the WHO multi-country study (18). However, the extremely low level of reported sexual violence by non partners maybe the manifestation of the profound stigma associated with sexual violence.

The health consequences of violence is grave and may result in fatal outcomes such as homicide, suicides to non fatal outcomes such as injuries, chronic pain syndrome, unintended pregnancies, pregnancy complications, and sufferings from depression, anxiety, suicidal thoughts and other psychosomatic symptoms. The economic cost of violence against women is substantial involving direct medical and health care services, productivity loss, death and disability apart from the pain and mental anguish suffered. Maternal death in pregnancy, a severe consequence of injury and violence, is prevalent in Bangladesh. Injuries from spousal physical abuse are quite common. Further, children who are witnesses of domestic violence are more likely to be the perpetrators of violence against women in later life. The strongest factor associated with husbands' violence against women was the history of abuse of the husband's mother by his father. Negative relationship between higher educational attainment of the husband and violence was observed. Non-payment and partial payment of dowry was also generally believed to instigate violence.

The findings that wife beating are more common amongst young couples was explained by changes in the life-stage, such as having grown up children and older age. Another explanation put forth, was that older women in abusive relationships probably develop strategies that decrease the frequency of violence or are less likely to report violence.

Majority of the abused women were found to remain silent about their experience of violence. The main reasons behind this silence was the high acceptance of violence within society, stigma and fear of greater harm, fear of challenging existing social norms and patriarchal values and tarnishing family honour, own reputation, lack of an alternative place to go to and fear of jeopardizing the children's future.

Interestingly, a higher rate of violence was observed against the beneficiaries of micro-credit programmes. This was explained by the increased tension that is associated with enhanced economic opportunities of women. However, it is believed that the level of violence decreases as the length of involvement with the micro-credit organizations increases,

The reasons for violence were many, yet mostly trivial. None of the reasons cited in the studies seemed serious enough to result in violence (11). As such it is normal to question why violence takes place? Is it due to the traditional norms that dictate the social belief that men are fundamentally superior to women? Is it the reflection of the deep-rooted patriarchal ideology that legitimizes and promotes rigid gender roles? Or is it the innate nature of men to perpetuate violence? The reasons for violence are a complex issue and should be understood and researched adequately.

The recent legal provisions of up-to-death penalty for violence against women in Bangladesh are one of the most significant measures to discourage women. However, the accessibility of legal services to women, especially to poor, rural women is limited. Legal services are city based and involve costs that may pose as an insurmountable hurdle for many. In addition, violence against women is neither condoned nor completely condemned. In general, the high acceptability of violence within the society acts as a deterrent for legal redress. As such the laws per se are unlikely to have a major impact on the reduction of violence against women. Obviously the fear associated with punitive judicial measures has its limitations. For the judicial measures to be effective the process of lodging complaints, legal battles and judgment should be transparent, fair, extremely well organized and uncomplicated. At present, the existing legal process is far from the idealistic condition. As such what can the alternative means be to prevent violence? One of the ways may be to increase awareness of basic human rights amongst women and the society. Increasing awareness about the forms of violence and its consequences on the immediate family and future generations may be the approach that can help to prevent violence. An effective strategy for the prevention of violence may be a community approach targeting both men and women. For interventions to be effective it may be necessary to target husbands, as they are the perpetrators of violence. Thus, knowledge about the effects of the interventions necessitates an in depth understanding of the social problem of abuse.

Many of the contributing risk factors for sexual and physical assault have been identified in childhood and adolescence and these need to be incorporated into prevention goals. Changing the norms and custom about relationships and providing the future generation with the skills needed to foster healthy relationships is a viable way to shift from a society where violence against women is widely condoned.

However, simply setting up services for victims of domestic violence will not improve the situation as only a small proportion of abused women access formal services for abuse. Public awareness campaigns such as public service announcements and advertisements are common approaches to primary prevention providing information regarding the warning signs of violence and community resources for victims and perpetrators. Developing community-based networks for coordinating services and programmes is important in raising awareness of violence and providing support to victims. Community-based educational activities may improve knowledge about women's legal and social rights and empower them to seek help for abuse. The prevailing social norms and ideologies that permit and encourage male violence must be directly and creatively addressed. The effort should involve careful research to identify messages and interventions that can change these attitudes. Education in the classroom and at community level as well as use of mass media are probable means that are essential for diffusion of these messages.

A large proportion of women accessing health services are, also, victims of violence. Interventions to support victimized women should involve health care provider settings. Violence against women is a multidimensional problem, which in its complexity is a major challenge for society. The very nature of the problem requires in depth knowledge of the complex web of social relations that has an impoverishing effect on individuals, families, and communities. Knowledge of effective attitudes and practices among health service providers and law enforcers to support victims of violence is crucial for the formulation of appropriate policies and programmes addressing the social issue concerned in the context of Bangladesh.

Acknowledgment

This review paper was done at ICDDR, B under the auspices of the Future Health Systems Project of ICDDR, B. The authors acknowledge the support extended by the Department for International Development (DFID), UK, through the "Future Health Systems: Innovations for equity--a research program consortium (Grant#H050474)"

Received April 20, 2007

References

(1.) Desjurlais R, Kleinman A. Violence and well-being. Soc Sci Med 1997; 45: 1143-5.

(2.) Fischback R, Herbert B. Domestic violence and mental health: Correlates and conundrums within and across cultures. Soc Sci Med 1997; 45: 1161-76.

(3.) Oxfam International. Towards ending violence against women in South Asia. Oxfam briefing paper: Oxfam International. Oxford, United Kingdom: Oxford International; 2004 August.

(4.) My Rights, who control? Kuala Lampur, Malaysia: APWLD (Asia Pacific forum on Women, Law and Development) 1990.

(5.) Economic and Social Council. Report of the Working Group on violence against women. Vienna: United Nations; 1992.

(6.) United Nations Declaration on the elimination of violence against women. In: Proceedings of the 85th Plenary Meeting, Geneva, December 20, 1993. Geneva: United Nations General Assembly, 2003. (Cited 2007 2 March) Available from: http:/ /daccessdds.un.org/doc/UNDOC/GEN/N94/095/05/PDF/ N9409505.pdf?OpenElement.

(7.) Heise L, Raikes A, Watts CH, Zwi A. Violence against women: a neglected public health issue in less developed countries. Soc Sci Med 1994; 39: 1165-79.

(8.) Heise L, Moore K, Toubia N. Sexual coercion and women's reproductive health: A focus on research. New York: Population Council; 1995.

(9.) Naved R, Azim S, Bhuiya A, Persson L Physical violence by husbands: Magnitude, disclosure and help seeking behavior of women in Bangladesh. Soc Sci Med 2006; 62: 2917-29.

(10.) Azim S. Naripokkho's pilot survey on violence against women in Bangladesh. Dhaka: Naripokkho; 2000.

(11.) Bhuiya A, Sharmin T, Hanifi S. Nature of domestic violence against women in a rural area of Bangladesh: Implication for preventive interventions. J Health Pop Nutr 2003; 2: 48-54.

(12.) Koenig M, Ahmed S, Hossain M, Mozumder A. Women's status and domestic violence in rural Bangladesh: Individual and community led effects. Demography 2003; 40: 269-88.

(13.) Heise L, Pitanguy H, Germain A. Violence against women: The hidden health burden. World Bank Discussion Papers No.255. Washington DC: The World Bank; 1994.

(14.) Schuler S, Hashemi S, Riley P, Akhter S. Credit programs patriarchy and men's violence against women in rural Bangladesh. Soc Sci Med 1996; 43: 1729-42.

(15.) Steele F, Amin S, Naved R. Savings/credit group formation and change in contraception. Demography 2001; 32: 267-82.

(16.) Naved R, Azim S, Persson L, Bhuiya A. Women's health and domestic violence against women in Bangladesh. Dhaka, Bangladesh: Urban Primary Health Care Project-Asia Development Bank; 2002.

(17.) Ahmed S. Intimate partner violence against women: Experiences from a woman-focused development programme in Matlab, Bangladesh. J Health Pop Nutr 2005; 23: 95-101.

(18.) WHO. Multi-country Study on women's health and domestic violence against women: Initial results on prevalence, health outcomes and women's responses. Geneva, Switzerland: World Health Organization; 2005.

(19.) ICDDR-B. Domestic violence against women in Bangladesh. Health Sci Bull 2006 June; 4(2).

(20.) Report of the subregional expert group meeting on eliminating violence against women: Violence against women in South Asia. Report No.: ST/ESCAP/2099. New York: United Nations; 2000.

(21.) Jahan R. Hidden danger: women and family violence in Bangladesh. Dhaka: Women for Women; 1994 p. 165.

(22.) Hadi A. Women's productive role and marital violence in Bangladesh. J Family Viol 2005; 20: 181-9.

(23.) Hadi A. Prevalence and correlates of the risk of marital violence in Bangladesh. J Interperson Viol 2000; 15: 787-805.

(24.) Chowdhury A, Bhuiya A. Do poverty alleviation programmes reduce inequities in health? The Bangladesh experience. In: Leon D, Walt G, editors. Poverty, inequality and health: an international perspective. Oxford: Oxford University Press; 2001 p. 312-32.

(25.) Schuler S, Hashemi S, Badal S. Men's violence against women in rural Bangladesh: Undermined or exacerbated by micro-credit programmes. Dev Prac 1998; 8: 148-57.

(26.) Khan M. Domestic violence against women: does development intervention matter? Dhaka: Research and Evaluation Division, Bangladesh Rural Advancement Committee; 1998.

(27.) Naved R, Persson L. Factors associated with spousal physical violence against women in Bangladesh. Stud Family Planning 2005; 36: 289-300.

(28.) Amin M, Khuda B, Islam A, Levin A. Do female education and mobility lead to less intra-family abuse? Dhaka: International Centre for Diarrhoeal Disease Research, Bangladesh; 1997.

(29.) Dobash RE, Dobash R. Women, violence and social change. London: Routledge; 1992.

(30.) Counts D, Brown J, Campbell J. Sanctions and sanctuary: cultural perspectives on the beating of wives. Boulder: CO: Westview Press; 1992.

(31.) Heise L, Ellsberg M, Gottemeller M. Ending violence against women in Latin American and Caribbean Region. A critical review of interventions. Latin America and the Caribbean Region: The World Bank, Poverty Sector Unit; 1999 October 2004.

(32.) Schuler M. Freedom from violence: Women's strategies around the world. New York: OEF International; 1992.

(33.) World Bank. World Development Report 1993: Investing in health. New York: Oxford University Press; 1993.

(34.) Domestic violence against women: Parliamentary Assembly of the Council of Europe 2002; 2002.

(35.) Jejeebhoy S. Associations between wife beating and fetal and infant death: Impression from a survey in rural India. Stud Family Planning 1998; 29: 300-8.

(36.) Naved R. A situational analysis of violence against women in South Asia. Paper commissioned by UNFPA CST Kathmandu for the Regional workshop on Parliamentary Advocacy for the prevention of violence against women in South Asia,. Dhaka, Bangladesh: United Nations Fund Population Activities (UNFPA); 2003.

(37.) Paltiel F. Women and mental health: A post Nairobi perspective. World Health Stat Q 1987; 40: 233-66.

(38.) Acid Survivors Foundation. Available from: http:// www.acidsurvivors.org/index_home.htm, accessed on March 2, 2007.

(39.) Rahman S. Reflections on women and violence in Bangladesh. Available from: http://www.hurights.or.jp/asia-pacific/no_24/ 05, accessed on March 5, 2007.

(40.) Odhikar. [cited 2007 5 March]; Available from: http:// www.odhikar.org, accessed on March 5, 2007.

(41.) South Asia Regional Campaign: We can end all violence against women. Available from: http://www.wecanendvaw.org/ bangladesh.htm, accessed on March 2, 2007.

Tania Wahed & Abbas Bhuiya

Social & Behavioural Sciences Unit, International Centre for Diarrhoeal Disease Research (ICDDR, B), Dhaka, Bangladesh

Reprint requests: Dr Abbas Bhuiya, Social & Behavioural Sciences Unit (ICDDR, B), Mohakhali, Dhaka 1000, Bangladesh email: abbas@icddrb.org
Gale Copyright: Copyright 2007 Gale, Cengage Learning. All rights reserved.